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
|
|---|---|---|---|---|---|
Attachment to the host mucosa is a key step in bacterial pathogenesis . On the apical surface of epithelial cells , members of the human carcinoembryonic antigen ( CEA ) family are abundant glycoproteins involved in cell-cell adhesion and modulation of cell signaling . Interestingly , several gram-negative bacterial pathogens target these receptors by specialized adhesins . The prototype of a CEACAM-binding pathogen , Neisseria gonorrhoeae , utilizes colony opacity associated ( Opa ) proteins to engage CEA , as well as the CEA-related cell adhesion molecules CEACAM1 and CEACAM6 on human epithelial cells . By heterologous expression of neisserial Opa proteins in non-pathogenic E . coli we find that the Opa protein-CEA interaction is sufficient to alter gene expression , to increase integrin activity and to promote matrix adhesion of infected cervical carcinoma cells and immortalized vaginal epithelial cells in vitro . These CEA-triggered events translate in suppression of exfoliation and improved colonization of the urogenital tract by Opa protein-expressing E . coli in CEA-transgenic compared to wildtype mice . Interestingly , uropathogenic E . coli expressing an unrelated CEACAM-binding protein of the Afa/Dr adhesin family recapitulate the in vitro and in vivo phenotype . In contrast , an isogenic strain lacking the CEACAM-binding adhesin shows reduced colonization and does not suppress epithelial exfoliation . These results demonstrate that engagement of human CEACAMs by distinct bacterial adhesins is sufficient to blunt exfoliation and to promote host infection . Our findings provide novel insight into mucosal colonization by a common UPEC pathotype and help to explain why human CEACAMs are a preferred epithelial target structure for diverse gram-negative bacteria to establish a foothold on the human mucosa .
During evolution bacteria have developed fascinating strategies to colonize multicellular organisms . A first critical step , which in many cases determines the outcome of the microbe-host encounter , is the ability of the microorganisms to establish themselves on mucosal surfaces [1 , 2] . Attachment to the mucosa is facilitated by specific bacterial adhesins , which firmly connect the microbe to the tissue [3 , 4] . Indeed , adhesin-mediated bacteria-host interactions prevent mechanical removal of the microbes via mucociliary cleansing or urinary flow , and can be seen as a prerequisite for efficient colonization . However , mucosal epithelia have several additional tissue-intrinsic defense mechanisms that protect the surface from adherent pathogens [5] . For example , in both stratified as well as single-layered epithelia the superficial cells are constantly replaced from a stem cell population . This tissue turnover also leads to shedding of cell-associated microbes from the epithelium reducing the bacterial burden . Epithelial tissue turnover can be very fast , as in the intestinal epithelium , where the superficial cells on the exposed villus folds are continuously replaced every day and where this process helps to maintain intestinal homeostasis . Indeed , slowing down tissue turnover in the intestinal tract can facilitate pathogen colonization [6 , 7] . Similar to the single-layered epithelium of the gut , stratified epithelia of the urogenital tract are also subject to continuous tissue renewal , albeit at a lower rate . However , exposure to high numbers of bacteria can trigger an accelerated turnover , whereby large amounts of superficial epithelial cells are released , a mechanism also known as exfoliation [8–12] . Exfoliation is an innate protective mechanism that , via rapid detachment and shedding of the infected superficial cells , limits colonization of the tissue by the microflora and ultimately prohibits further penetration of the bacteria [13] . By this process , even cell-associated bacteria can be removed from the tissue surface together with the infected cells . Recently , we could show that specialized bacteria , which colonize the human urogenital tract , are able to suppress the exfoliation response [14] . These bacteria utilize outer membrane adhesins , the so-called OpaCEA proteins , to bind to members of the CEACAM family , a group of immunoglobulin-related gylcoproteins expressed on the apical membrane of mucosal epithelial cells ( for review see [15] ) . CEACAM engagement by bacteria triggers activation of integrins , enhances matrix adhesion and reduces cell detachment of infected cells , ultimately facilitating bacterial colonization [14 , 16] . Besides Neisseria gonorrhoeae , which was studied in these previous investigations , also the closely related Neisseria meningitidis expresses CEACAM-binding Opa proteins and exploits mucosal CEACAMs such as CEACAM1 or CEA ( the product of the CEACAM5 gene ) for contacting host cells and for colonization of the nasopharynx [17–19] . In both instances , the pathogens selectively bind to human , but not other mammalian CEACAM family members [20] , suggesting that this exquisite recognition mechanism is a result of the co-evolution of these microbes with their sole natural host . Indeed , by interfering with epithelial exfoliation OpaCEA-expressing Neisseria should have a clear advantage during colonization of the human mucosa . It is currently unclear if the expression of OpaCEA proteins is sufficient to counteract exfoliation or if additional neisserial virulence factors are involved in this process . This aspect is of particular interest , as several unrelated bacterial pathogens , such as Haemophilus influenzae , Moraxella catarrhalis , and pathogenic strains of Escherichia coli , have been shown to possess distinct CEACAM-binding adhesins ( for an overview see [21] ) and might trigger similar processes . To address this question , we have investigated if engagement of human CEACAMs by a bacterial CEACAM-binding adhesin is sufficient to counteract exfoliation and to promote mucosal colonization . Indeed , we find that expression of a neisserial OpaCEA protein in non-pathogenic E . coli allows these bacteria to engage CEA on epithelial cells , to trigger increased integrin activity and cell-matrix adhesion , and to promote mucosal colonization . Importantly , uropathogenic E . coli ( UPEC ) , harboring a CEACAM-binding adhesin of the Afa/Dr family , also exploit this mechanism to block epithelial exfoliation and to boost their ability to colonize the urogenital tract in vivo . Together , our results establish human CEACAMs as a preferred epithelial target structure , which allows diverse gram-negative bacterial pathogens to suppress exfoliation and to efficiently colonize the human mucosa .
Previous work demonstrated that a CEACAM-binding adhesin is necessary to allow Neisseria gonorrhoeae to increase extracellular matrix binding of infected cells [14 , 16] . To test if CEACAM-binding alone is sufficient to trigger this process , we separated the CEACAM-binding OpaCEA adhesin from other gonococcal virulence determinants by expressing this neisserial outer membrane protein in E . coli . In the used expression plasmid pTrc , OpaCEA-protein expression is under control of the IPTG-inducible lac promoter [22] . Due to the leakiness of this promoter , OpaCEA protein-expression was already observed under non-inducing conditions in OpaCEA-expressing E . coli ( E . coli OpaCEA ) compared to an E . coli strain harbouring the empty pTrc plasmid ( S1A Fig ) . The OpaCEA protein expressed in E . coli showed a similar size to the native OpaCEA protein expressed in N . gonorrhoeae , but levels were only about 20% of that observed in gonococci ( S1A Fig ) . Both E . coli strains showed a comparable growth pattern in liquid culture , indicating that expression of the OpaCEA protein in E . coli at this level did not interfere with growth ( S1B Fig ) . In contrast , strong overexpression , e . g . upon IPTG induction , can retard E . coli growth ( S1C Fig ) . When expressed in E . coli , the neisserial Opa protein adhesin was functional with regard to CEACAM-binding , as E . coli OpaCEA was able to associate with the GFP-tagged amino-terminal domain of CEA ( CEA-N; Fig 1A ) . The control E . coli strain did not bind to soluble CEA-N , and no binding of either E . coli strain to the CEACAM8-amino-terminal domain ( CEA8-N ) was observed ( Fig 1A ) . This binding pattern of the heterologously expressed OpaCEA protein is in agreement with the CEACAM binding profile observed for N . gonorrhoeae expressing this OpaCEA protein ( Fig 1A ) . To investigate , whether E . coli OpaCEA interacts with CEA in a cellular context , we infected the human cervical epithelial cell line ME-180 for 2 h with the E . coli control strain or with E . coli OpaCEA . ME-180 cells endogenously express CEACAM1 , CEACAM6 , and CEA , which were located in the plasma membrane of uninfected cells ( S2A and S2B Fig ) . Upon infection with the E . coli control strain , the distribution of CEACAMs was unchanged and these bacteria did not associate with ME-180 cells ( Fig 1B and S2B Fig ) . Importantly , E . coli OpaCEA strongly adhered to ME-180 cells and triggered re-location and local concentration of CEACAMs at sites of bacteria-host cell contact ( Fig 1B and S2B Fig; arrowheads ) . Similarly , OpaCEA protein-expressing gonococci adhered in large numbers to ME-180 cells and induced CEACAM clustering ( Fig 1B and S2B Fig ) . Together , these results demonstrated that the neisserial adhesin is functionally expressed in E . coli , where it promotes interaction with CEA-expressing cells . CEA engagement by pathogenic N . gonorrhoeae has been shown to trigger increased matrix adhesion and to counteract the detachment of epithelial cells [14] . Therefore , we infected confluent monolayers of ME-180 cells , grown on collagen , for 14 h or left the ME-180 cells uninfected . Next , monolayers were washed to remove detached and loosely adherent cells and the remaining cells were stained with crystal violet . Elution and quantification of the dye in a spectrophotometer served as a measure of the remaining adherent cells . Interestingly , in samples infected with OpaCEA protein-expressing N . gonorrhoeae as well as in samples infected with OpaCEA protein-expressing E . coli no reduction in the amount of adherent cells compared to uninfected cells was observed ( Fig 1C ) . Indeed , ME-180 cells infected with CEACAM-binding bacteria even showed a slightly higher recovery than uninfected cells suggesting that cell-matrix adhesion is reinforced upon infection with OpaCEA protein expressing bacteria ( Fig 1C ) . In contrast to CEACAM-binding bacteria , cells infected with the E . coli control strain or with piliated , non-CEACAM-binding gonococci ( Ngo P+ ) showed pronounced detachment of cells ( Fig 1C ) . Microscopic observation of the infected cell cultures corroborated the massive loss of epithelial cells from monolayers infected with control E . coli or Ngo P+ , whereas cell numbers in samples infected for prolonged times with OpaCEA protein-expressing bacteria were even higher than in uninfected samples ( Fig 1D ) . Similar results were obtained , when primary human vaginal epithelial ( hVECs ) cells were infected for 14 h with either OpaCEA protein-expressing bacteria or non-CEACAM-binding strains . Again , expression of a CEACAM-binding adhesin was able to block the infection-induced detachment of the primary epithelial cells ( S3A and S3B Fig ) . Scanning electron microscopy revealed that ME-180 cells infected for 14 h with control E . coli reduced cell-cell-contacts and rounded up , whereas cells infected with OpaCEA protein-expressing E . coli remained well-spread on collagen similar to uninfected cultures and similar to cells infected with OpaCEA-protein expressing gonococci ( Fig 1E ) . These results demonstrate that CEACAM engagement via a CEACAM-binding adhesin , even when expressed in a heterologous background , is sufficient to counteract the detachment of infected epithelial cells in vitro . Both human primary vaginal epithelial cells as well as ME180 cervical carcinoma cells endogenously express members of the CEACAM family , in including CEA ( S2B and S3C Figs ) . To rigorously test the contribution of CEACAMs to this process , we employed 293 cells , a human cell line that lacks endogenous CEACAM expression . Upon transient transfection with a CEA-encoding plasmid , about 40% of the cell population showed surface expression of the receptor ( S4A Fig ) . Control transfected or CEA-expressing cells were then infected for 5 h with the indicated bacterial strains and used in cell adhesion assays on collagen . We again observed increased matrix binding of CEA-expressing cells upon infection with OpaCEA protein-expressing E . coli ( S4B Fig ) . The adhesion of CEA-expressing cells infected with E . coli OpaCEA was comparable to the increased cell adhesion seen upon infection with OpaCEA protein-expressing gonococci . In contrast , incubation with the E . coli control strain did not alter matrix adhesion of the infected cells ( S4B Fig ) . Importantly , 293 cells transfected with an empty control plasmid ( pcDNA ) did not display changes in adhesiveness , irrespective of the bacterial strain used for infection ( S4B Fig ) . These results demonstrate that it is the adhesin-CEACAM interaction , which serves as the trigger for increased extracellular matrix adhesion of the infected cells . Previously , CEACAM engagement by bacteria has been shown to induce CD105 expression in epithelial cells , which was a pre-requisite for enhanced matrix adhesion of infected cells [16] . Indeed , infection of CEA-expressing 293 cells with OpaCEA protein-expressing E . coli or N . gonorrhoeae resulted in the presence of CD105 on the surface of the cells , whereas uninfected cells or cells infected with control E . coli did not have detectable CD105 on their surface ( S4C Fig ) . In line with the idea that CEACAM-triggered expression of CD105 is critical for increased cell adhesion , expression of CD105 in 293 cells led to strongly elevated cell-matrix adhesion in the absence of bacterial infection or CEACAM stimulation ( S4B Fig ) . Though bacterial infection or CD105 expression resulted in alterations in cell-matrix adhesion , the amount of surface exposed integrin β1 was unaltered in all infected samples compared to the uninfected control ( S5A Fig ) . However , infection with CEACAM-binding gonococci or OpaCEA protein-expressing E . coli promoted a conformational change of integrin β1 as detected by the conformation-sensitive anti-integrin β1 antibody 9EG7 ( S5B Fig ) . Increased labeling by antibody 9EG7 indicated that CEA-engagement by bacteria led to pronounced activation of integrins . In contrast , integrin activity remained low in uninfected cells or cells infected with non-CEACAM-binding bacteria ( S5B Fig ) . The increased integrin activity clearly depended on the presence of CEA , as 293 cells transfected with the empty control vector did not alter the amount of surface integrin nor integrin activity upon infection with diverse bacteria ( S5C and S5D Fig ) . As a further control , cells were stimulated with Mn2+ , an exogenous activator of integrins . Upon Mn2+ addition , a similar , maximal integrin activity was observed in all samples demonstrating that the total activatable integrin levels on the cell surface were similar ( S5B and S5D Fig ) . Together , these findings imply that E . coli OpaCEA , via engagement of CEACAMs , can trigger CD105 expression , which in turn enhances integrin activity in vitro . Because OpaCEA-expressing E . coli was able to enhance integrin activity and suppress cell detachment in vitro , we next analyzed bacterial colonization of the urogenital tract . For this purpose , we used either wildtype mice or transgenic mice , expressing human CEA on all mucosal surfaces ( CEAtg mice ) [14 , 23] . Accordingly , 8–10 week old female mice were vaginally infected with 1 x 106 bacteria and colonizing bacteria were recovered by urogenital swabs 24 hours later . Only few bacterial colonies of the E . coli control strain could be isolated from wildtype mice and a slightly elevated ( ~3-fold ) recovery of this non-CEACAM binding strain from CEA-tg mice was observed ( Fig 2A ) . In contrast , more than 30-fold higher numbers of OpaCEA protein-expressing E . coli were recovered from CEAtg mice than from wildtype mice ( Fig 2A ) . Analysis of re-isolated OpaCEA protein-expressing E . coli showed that expression of the Opa adhesin was unaltered by in vivo growth conditions ( Fig 2B ) . Immunohistochemical staining of tissue sections from the urogenital tract revealed that OpaCEA-expressing E . coli were closely associated with the CEA-positive tissue surface ( Fig 2C ) . In the case of E . coli control only few bacteria could be detected on the mucosal surface ( Fig 2C ) . Importantly , CD105 was expressed by mucosal epithelial cells of CEAtg mice in contact with OpaCEA-protein expressing E . coli , whereas the E . coli control strain did not trigger CD105 expression ( Fig 2D ) . Furthermore , E . coli did not trigger CD105 expression in wildtype mice , irrespective of the Opa protein status of the bacteria ( S6 Fig ) . These data demonstrate that OpaCEA-expressing E . coli are able to associate with CEA-positive epithelial cells and trigger CD105 expression in the urogenital tract in vivo . Together , these data suggest that a CEACAM-binding adhesin is not only necessary , but also sufficient to promote colonization of the mucosal surface via stimulating CD105 expression and enhanced integrin activity in superficial epithelial cells . Besides the neisserial OpaCEA proteins , additional CEACAM-binding bacterial proteins have been described . For example , uropathogenic E . coli harbouring the Afa/Dr locus of afimbrial adhesins have also been shown to engage CEACAMs including CEACAM1 and CEA [24] . Therefore , we used the uropathogenic E . coli ( UPEC ) strain A30 , which expresses the AfaE-III adhesin ( E . coli AfaE-III ) encoded within the afa gene cluster on a large virulence plasmid [25] . To monitor AfaE-III-dependent events , we cured strain A30 from the virulence plasmid generating the AfaE-III-negative strain E . coli ΔAfaE-III , which lacks the afa gene cluster ( S7A Fig ) . The wildtype UPEC strain and the ΔAfaE-III strain grew with similar growth kinetics ( S7B Fig ) . Clearly , E . coli AfaE-III was able to associate with human CEACAM1 and CEA in the form of soluble GFP-tagged receptor domains and this property was lost in E . coli ΔAfaE-III ( Fig 3A ) . Similar to other CEACAM-binding bacteria such as gonococci , E . coli AfaE-III selectively bound to human CEACAM1 , but not to CEACAM1 orthologues from other mammalian species , including mouse , dog and cattle ( Fig 3B ) . Furthermore , E . coli AfaE-III strongly associated with CEA-expressing ME-180 cells and clustered CEA on the surface of the infected cells , whereas E . coli ΔAfaE-III hardly attached to the cell surface ( Fig 3C ) . These results suggest that E . coli pathovars associated with urogenital infections target human CEACAM family members present on epithelial cells of the urogenital tract via their AfaE-III adhesin . To investigate , if a CEACAM-binding pathogenic E . coli strain is able to modulate the matrix-adhesion of infected epithelial cells , we again employed 293 cells transfected either with a GFP-encoding control vector or a CEA-encoding expression vector . As observed for CEACAM-binding gonococci , E . coli AfaE-III promoted cell-matrix adhesion in 293 cells expressing CEA , but not in control transfected cells ( Fig 4A ) . Neither the E . coli control strain nor E . coli ΔAfaE-III led to enhanced extracellular matrix adhesion of infected cells ( Fig 4A ) . In agreement with the enhanced extracellular matrix adhesion , CEA-expressing 293 cells did not detach upon infection with E . coli AfaE-III , whereas detachment of cells infected with non-CEACAM-binding E . coli ( E . coli control strain or E . coli ΔAfaE-III ) , could be readily detected under the microscope ( Fig 4B and 4C ) . Treatment of the infected cells with Mn2+ , a general inducer of integrin activity , increased the matrix adhesion of cells infected with the E . coli control strain or E . coli ΔAfaE-III , but did not further enhance collagen binding of cells infected with CEACAM-binding E . coli AfaE-III ( Fig 4D ) . Moreover , infection of CEA-expressing 293 cells with E . coli AfaE-III , but not E . coli ΔAfaE-III , triggered CD105 expression by 293 cells ( Fig 4E ) . These results indicate that E . coli AfaE-III is able to exploit CEACAMs to modulate host cell adhesion and to counteract bacteria-induced cell detachment via CD105 expression and integrin activation . In line with this idea , infection with E . coli AfaE-III , but not with but not E . coli ΔAfaE-III , resulted in enhanced integrin β1 activity in CEA-expressing cells ( Fig 4F and 4G ) . These results demonstrate that CEACAM-binding pathogenic strains of E . coli can modulate host cell adhesion and integrin activity on human epithelial cells . Because CEACAM-binding by E . coli AfaE-III results in enhanced integrin activity in vitro , we wondered whether these pathogens have an advantage during colonization of the urogenital tract . Therefore , wildtype or CEAtg mice were vaginally infected for 24 h with 106 E . coli AfaE-III or with E . coli ΔAfaE-III . 24h after infection , colonization was analyzed by dilution plating . Whereas only low numbers of bacteria were recovered from the urogenital tract of wildtype mice , recovery of E . coli AfaE-III from CEAtg mice increased more than 80-fold ( Fig 5A ) . In contrast , numbers of non-CEACAM-binding E . coli ΔAfaE-III were only slightly ( ~3-fold ) elevated in CEAtg mice compared to wildtype mice ( Fig 5A ) . Immunohistochemistry revealed that numerous E . coli AfaE-III were found associated with the epithelial surface of CEAtg mice , which stained positive for human CEA ( Fig 5B ) . Furthermore , superficial epithelial cells of CEAtg mice infected with E . coli AfaE-III showed local expression of CD105 ( Fig 5C ) . While occasionally non-CEACAM-binding E . coli were also detected on the mucosal surface of CEAtg mice , no increase in CD105 expression was evident ( Fig 5C ) . Together , these results indicate that the enhanced colonization of CEAtg mice by E . coli AfaE-III might be due to modulation of epithelial cell-extracellular matrix adhesion via CD105 expression and integrin activation , which would provide the mechanistic explanation for the observed phenotype . In the case of gonococcal infection , bacteria-triggered CD105 expression on the epithelial surface translates into suppression of host cell exfoliation in vivo [14] . To investigate the level of host cell exfoliation in response to bacterial infection , we used scanning electron microscopy ( SEM ) of the urogenital tract . In uninfected wildtype and CEAtg animals , the surface of the upper vaginal epithelium showed few detaching superficial cells , indicating low tissue turnover under these conditions ( Fig 6A ) . However , infection of wildtype mice with E . coli AfaE-III or E . coli ΔAfaE-III triggered a dramatic increase in exfoliation of superficial epithelial cells ( Fig 6B and 6C ) . In strong contrast , infection with the CEACAM-binding UPEC strain did not result in an increased exfoliation of epithelial cells in CEAtg mice ( Fig 6B and 6C ) . Despite the presence of numerous E . coli AfaE-III as well as E . coli ΔAfaE-III on the vaginal epithelium , only the non-CEACAM-binding strain E . coli ΔAfaE-III led to a strong increase in exfoliation ( Fig 6B and 6C ) . The low number of exfoliating cells in CEAtg mice upon infection with E . coli AfaE-III was comparable to the uninfected situation indicating that CEACAM-binding by the AfaE-III adhesin is able to completely suppress the exfoliation response ( Fig 6B and 6C ) . Together , these results suggest that CEACAM-binding pathogenic E . coli , such as E . coli AfaE-III , can interfere with the detachment of infected epithelial cells in a manner reminiscent of OpaCEA-expressing Neisseria gonorrhoeae . Suppression of exfoliation clearly is a means to increase the likelihood of a successful and lasting colonization . Therefore , CEACAM-triggered interference with epithelial exfoliation seems to be more common amongst human pathogens than previously appreciated and appears to be an evolutionarily favourable strategy to colonize human mucosal surfaces .
Regulation of epithelial exfoliation is a particularly effective and rapid innate defense mechanism modulating mucosal colonization by microorganisms . However , there is only limited knowledge how pathogens themselves regulate this process and which molecular factors affect cell exfoliation during the course of an infection . In this study we provide novel insight into the role of bacterial and host determinants , which modulate the exfoliation of epithelial cells . Based on prior observations with Neisseria gonorrhoeae , we were able to confirm that engagement of members of the CEACAM family on the mucosal surface of the upper vaginal epithelium results in suppression of exfoliation . Importantly , the observed host cell responses were independent of the bacterial background , in which the CEACAM-binding adhesin was expressed , as both OpaCEA protein-expressing gonococci as well as E . coli were able to block exfoliation . These results demonstrate that CEACAM engagement is not only necessary , but also sufficient to promote increased host cell-extracellular matrix adhesion and to counteract exfoliation . Interestingly , a pathogenic UTI isolate of E . coli , which expresses the Dra/AfaE CEACAM-binding adhesin , was able to induce a similar host cell phenotype in vitro , characterized by CEACAM-triggered upregulation of CD105 , increased integrin activity , and enhanced host cell adhesion to the extracellular matrix . CEACAM engagement allowed these pathogens to blunt epithelial exfoliation leading to enhanced mucosal colonization in vivo . Based on these findings we propose that CEACAM-binding adhesins have independently evolved in multiple gram-negative bacterial pathogens , including pathogenic Neisseriae , E . coli pathovars , Haemophilus influenzae and Moraxella catarrhalis , as a means to facilitate the initial , species-specific contact with the mucosa of an appropriate host organism and to counteract the detachment of superficial cells . One of the best studied examples of bacteria-induced exfoliation is taking place in the bladder , where incoming bacteria trigger massive shedding of the superficial umbrella cells , a specialized cell type covering the luminal surface of the bladder urothelium [10 , 26] . Indeed , while in some organs the epithelium regenerates constantly , the mammalian urinary bladder can shift from a physiological mode of slow tissue turnover to a highly proliferative status as a result of epithelial injury [27 , 28] . UPEC strains that infect the bladder epithelium are characterized by the possession of specific adhesin gene clusters , such as the operons encoding for the type 1 pilus , the P pilus , or the Afa/Dr family of adhesins , as well as the secretion of toxins such as α-hemolysin ( HlyA ) [29 , 30] . In particular , the FimH adhesin-mediated contact of UPEC with bladder cells has been shown to induce pronounced exfoliation and tissue renewal [10] . Detachment of the large superficial urothelial cells , which seems to be accompanied by apoptosis , then affords the pathogen access to deeper strata of the bladder epithelium , where some bacteria invade , multiply , and persist in undifferentiated epithelial cells [31 , 32] . Accordingly , a small fraction of FimH-expressing UPEC seems to profit from exfoliation , as increased urothelial stem and early progenitor cell proliferation provide an expanded protective niche for UPEC in the bladder [27] . In line with the idea that in some instances the bacteria might benefit from the host tissue response , Dhakal and Mulvey recently identified the pore-forming toxin α-hemolysin ( HlyA ) as the bacterial effector that induces exfoliation upon infection with UPEC [33] . At sublethal concentrations , HlyA activates host cell proteases resulting in breakdown of integrin-associated proteins such as paxillin , thereby weakening cell-matrix attachment . Interestingly , the finding by Dhakal and Mulvey that a secreted factor promotes cell detachment , also demonstrates that host cell contact is not a pre-requisite for the induction of epithelial exfoliation . This is also reflected in our current and prior studies , where non-adherent bacteria were able to trigger this process in vitro and in vivo [14 , 16] . Clearly , the non-pathogenic E . coli or non-opaque gonococci used in these studies do not secrete toxins , pointing to the existence of additional soluble triggers for host cell detachment . It has been speculated that conserved bacterial factors such as LPS could initiate this host response [8] . However , infection of TLR-4-deficient mice with E . coli also resulted in a strong increase in exfoliation , suggesting that LPS , or more precisely LPS sensing by TLR-4 , is not required to trigger exfoliation ( S8 Fig ) . Therefore , future efforts should be directed towards identifying the relevant ( and most likely conserved ) bacterial feature ( s ) , which initiate epithelial exfoliation . Amongst UTI isolates of E . coli , type 1 pilus or P pilus expression in combination with the secretion of HlyA prevails , which is in line with the idea that these pathotypes profit from epithelial exfoliation [13] . However , a further pathotype of UTI has been described ( pathotype V ) , which lacks HlyA secretion and expresses Afa/Dr-related adhesins [30] . Indeed , non-hemolytic , Afa/Dr-possessing E . coli are found in less than 4% of fecal isolates , but make up almost 10% of UTI isolates suggesting a prominent enrichment of strains with this genetic makeup during urogenital colonization [30] . Even higher rates of Afa/Dr-expressing strains have been reported from cystitis cases in children and pyelonephritis cases of pregnant women [34 , 35] . Our results with a non-hemolytic , AfaE-III-expressing strain now demonstrate that this pathotype can engage human CEACAMs on the surface of the urogenital mucosa to suppress exfoliation . Though our model relies on infection of the upper vaginal epithelium in mice , rather than the bladder epithelium , pathogenic E . coli can also infect the human genital tract mucosa of both male and female and can be acquired similar to gonococci by sexual transmission [36] . Therefore , our results indicate that amongst isolates some strains may , in contrast to HlyA-expressing strains , be able to suppress exfoliation via the possession of CEACAM-binding Afa/Dr adhesins . Clearly , CEACAM-binding is only found in a subfamily of Afa/Dr adhesins expressed by E . coli , the so-called Afa/Dr-I family , including Dr , F1845 and AfaE-III adhesins [24 , 37] . The Afa/Dr adhesins belong to the large group of chaperone/usher ( CU ) type adhesins , which can be located on fimbriae or which can occur in the form of an afimbrial surface sheath ( such as in the case of Afa adhesins ) [38 , 39] . Similar to other CU systems , the Afa adhesins are part of a small gene cluster ( afaA–afaF ) , which , besides the adhesin , encodes the accessory proteins required for surface expression of the adhesin proper [38] . The afa gene clusters comprise genes of transcriptional regulators ( afaA , afaF ) , a periplasmic chaperone ( afaB ) , and an integral outer-membrane protein ( afaC ) , which serves as the assembly platform , the so-called usher , for the surface display of the adhesin [40] . In the case of the afa-3 gene cluster , there are two genes ( afaD and afaE ) , which encode proteins with adhesive and invasive properties towards mammalian cells [41–44] . Isolated , recombinant AfaE-III derived from the afa-3 operon of the UPEC strain A30 has been shown to bind to CD55 as well as members of the CEACAM family including CEACAM1 , CEA , and CEACAM6 [37] . However , AfaE orthologues from other pathogenic E . coli strains , such as AfaE-I from strain KS52 , seemingly lack CEACAM binding [37] . For AfaE-III , the CEACAM binding interface of the adhesin has been structurally defined [45] . Though the six amino acids of AfaE-III engaged in the binding interface with CEA seem to be largely conserved in the CEACAM-binding Dr protein from strain IH11128 , the CEACAM-binding adhesin F1845 encoded by the daaE gene from strain C1845 does not share a single identical amino acid at the corresponding positions [37 , 46 , 47] . The lack of a defined CEACAM-binding motif currently prohibits the use of the vast sequence information about E . coli adhesins to predict possible binding interactions with human CEACAMs . Furthermore , a comprehensive functional analysis of CEACAM-binding properties among UTI-causing E . coli isolates and a correlation with the clinical manifestations is currently lacking . It is therefore interesting to note , that a recent survey by Qin et al . found a strong association between the presence of the afa gene cluster and recurrent infections of the lower urogenital tract , which were characterized by a lack of systemic symptoms [48] . In contrast , the afa operon was not detected in strains isolated from acute pyleonephritis or cystitis patients [48] . These findings are in line with the idea that some pathogenic strains of E . coli , e . g . afa-3 harboring strains , by the help of CEACAM-binding adhesins , are able to sustain long-term accommodation in the urogenital tract without causing systemic pathology . Interestingly , also many asymptomatic bacteriuria ( ABU ) E . coli isolates are hlyA-negative and carry afa gene clusters [49] . Based on our previous study , we have performed an additional extended PCR screening of 126 ABU isolates from various sources and find that 9 . 3% of these isolates encode afa/dra genes . These ABU strains can colonize the urinary tract with high bacterial numbers for many weeks or months without provoking an overt innate immune response [50] . Accordingly , suppression of exfoliation via Afa/Dr binding to CEACAM may also promote the stable asymptomatic colonization of the bladder by ABU strains . Given the HlyA-triggered induction of exfoliation versus Afa/Dr-mediated suppression of exfoliation , it can be envisioned that the presence of the respective virulence factors might dictate the distinct behaviour of these strains and the clinical outcome . Importantly , the majority of CEACAM-binding bacteria characterized to date does not colonize the urogenital tract , but rather inhabits the nasopharynx , such as Neisseria meningitidis , Moraxella catarrhalis and Haemophilus influenzae [51–53] . In each case , the bacteria employ structurally unrelated adhesins to contact CEACAMs on the luminal surface of the host tissue , where CEACAM-binding adhesins , including AfaE-III , bind to the protein part and not the glycan part of the receptor [24 , 45 , 54–56] . It is easily conceivable that all these human-restricted pathogens exploit CEACAMs for securing successful colonization of their sole natural host . Indeed , recent experiments with model organisms have demonstrated that adhesin-CEACAM interactions contribute to improved recovery of these bacteria after experimental infection of the nasopharyngeal mucosa [19 , 57] . Our in vivo experiments with CEAtg-mice indicate that the recovery of non-CEACAM binding E . coli from CEA-transgenic animals is also slightly , but consistently elevated compared to wildtype animals . It could be speculated that by expressing human CEA in mice , an additional glycoprotein is added to the murine mucosa that could allow additional low affinity , glycan-based interactions . Though such weak interactions might not be apparent in in vitro binding assays with soluble CEA-domains , they could contribute to a slightly improved colonization in vivo . Clearly , there is a strong further increase in colonization of the urogenital tract by CEACAM-binding E . coli and this correlates with the ability of these strains to suppress exfoliation in vivo , an effect not observed for non-CEACAM-binding bacteria . This suggests that potential low affinity , glycan-mediated binding interactions are not sufficient to mediate suppression of epithelial exfoliation . In contrast to the situation in the urogenital tract , it is currently unknown , if the improved colonization of the nasal cavity of CEACAM1 transgenic mice observed for N . meningitidis or the CEACAM-dependent colonization of chinchillas by Haemophilus influenzae is linked to a suppression of exfoliation . However , at least in stratified regions of the nasopharyngeal epithelium similar processes might occur as observed for AfaE-III-expressing E . coli or OpaCEA protein expressing N . gonorrhoeae in the urogenital tract . Together , a detailed understanding of the molecular processes initiating the exfoliation of epithelial cells and the host-tailored countermeasures by successful pathogens holds the promise to provide novel avenues to interfere with or to protect from bacterial colonization right from the start of the bacterial host cell encounter .
The E . coli and E . coli OpaCEA strains were derived from DH5α and have been described previously [22] . The heterologous expression of Opa proteins in E . coli allows these strains to interact with human CEACAMs in a manner analogous to Opa protein expressing gonococci and meningococci [18] . The opaque phenotype was regularly controlled by Western blotting of bacterial lysates with the monoclonal anti-Opa protein antibody ( clone 4B12/C11 ) . The uropathogenic E . coli harbouring the afa-3 gene cluster ( strain A30; isolated from a cystitis patient [25 , 58] is a non-hemolytic , serotype O75 strain expressing the CEACAM-binding AfaE-III adhesin ( E . coli AfaE-III ) . All E . coli strains were grown on LB agar plates with or without adequate antibiotics and cultured at 37°C . To select for afa-deficient variants of E . coli AfaE-III , bacteria were sequentially cultured overnight at 37°C , then for 24 h at 40°C , at 42°C , and at 45°C . Serial dilutions of the final culture were plated on LB agar and individual colonies were tested by colony-PCR for the presence of the afa locus ( primer pair afa-f: 5’-ggcagagggccggcaacaggc-3’; afa-r: 5’-cccgtaacgcgccagcatctc-3‘ ) and the K5 capsule determinant ( K5-f: 5‘-cagtatcagcaatcgttctgta-3‘ / kpsII-r: 5’-catccagacgataagcatgagca-3’ ) as described [59] . In addition , we have attempted to generate a complemented strain , but have not succeeded in re-expressing the afa-gene cluster of E . coli strain A30 , despite subcloning into low-copy or inducible vectors . Sequencing of subcloned gene clusters in multiple independent clones revealed that all the tested subclones carried point mutations resulting in premature stop codons or frame shifts within the coding regions of the afa-gene cluster , indicating that the maintenance of heterologous fimbrial gene clusters in multiple copies in E . coli laboratory strains may select for subcloned PCR products with accumulated mutations preventing expression . For determining growth curves of E . coli strains , LB broth cultures were initiated and grown at 37°C for 4 h . These cultures were used as 1:25 inoculation into 100 ml LB medium at 37°C . Absorbance at 600nm was recorded every 60 min over a course of 10–25 h using a Libra S4 spectrophotometer ( Biochrom , Cambridge , UK ) . For infection , bacteria were suspended in DMEM , the optical density of the suspension at 600 nm ( OD600 ) was used to estimate the number of bacteria according to a standard curve , and the bacteria were added to the cells at the indicated multiplicity of infection ( MOI ) . Neisseria gonorrhoeae strains used in this study were described previously and were derived from strain MS11 [16] . The gonococci were either non-piliated and expressed a CEACAM-binding Opa protein ( Ngo OpaCEA; strain N309 ) , were non-piliated and expressed a heparansulphate proteoglycan-binding Opa protein ( Ngo OpaHSPG; strain N303 ) [22] , or the bacteria did not express an Opa protein ( non-opaque phenotype ) , but expressed pili to bind to human cells ( Ngo P+ , strain N280 , a non-opaque derivative of MS11-F3 [60] ) . The opacity and the piliation status of the used bacteria were regularly monitored by colony morphology as well as Western Blotting with monoclonal anti-Opa antibodies . Gonococci were grown on GC agar plates ( Difco BRL , Paisley , UK ) supplemented with vitamins , chloramphenicol ( 10 μg/ml ) and erythromycin ( 7 μg/ml ) at 37°C , 5% CO2 and subcultured daily . The human cervix carcinoma cell line ME-180 ( ATCC , Rockville , MD ) and the embryonic kidney cell line 293T ( 293 cells; ACC-635 , DSMZ , Braunschweig , Germany ) were cultured in DMEM containing 10% calf serum ( 293 cells ) or DMEM containing 10% FCS ( ME-180 cells ) at 37°C in 5% CO2 and subcultured every second to third day . 293 cells were transfected by calciumphosphate co-precipitation using a total of 5 μg of plasmid DNA for each 10 cm culture dish and employed in experiments 2 days after transfection . In some cases , 293 cells were serum-starved in DMEM containing 0 . 5% CS . The human vaginal epithelial cell ( hVEC ) line MS74 was obtained from A . J . Schaeffer ( Feinberg School of Medicine , Northwestern University , Chicago , IL ) , cultured on gelatine-coated dishes in DMEM containing 10% fetal calf serum ( FCS ) , and subcultured every third day . The expression plasmids used in this study included the commercially available vectors pcDNA3 . 1 Hygro ( pcDNA; Invitrogen , Karlsruhe , Germany ) , pEGFP-N1 , pLPS-3’-EGFP and pDsRed2-C1 ( Clontech , Palo Alto , CA ) . Plasmid pcDNA3 . 1 CEA ( pcDNA-CEA ) was described previously [61] . The expression plasmid pcDNA3 . 1 CEACAM1-4L-HA ( pcDNA-CEACAM1 ) was constructed by PCR amplification of the human CEACAM1-4L cDNA ( generous gift of W . Zimmermann , LMU München , Germany ) with primers CEACAM1-HA-sense ( 5’-GGGAAGCTTGCCATGGGGCACCTCTCAGCCCCACTTCAC-3’ ) and CEACAM1-HA-anti ( 5’-GGGGACGTCATAGGGATACTGCTTTTTTACTTCTGAATAAATTATTTCTG-3’ ) and was cloned into the HindIII-AatII—digested plasmid pBluescript CEACAM3-HA [62] before further subcloning via HindIII-NotI into pcDNA3 . 1 Hygro ( Invitrogen ) . The vector pLPS-3'-RFP2 was constructed by PCR amplifying the RFP2-coding sequence from vector pDsRed2-C1 with primers RFP2-AgeI-sense 5’-ATAACCGGTCGCCTCCTCCGAGAACGTCATCACC-3’ and RFP2-NotI-anti 5’-ATAGCGGCCGCTTACAGGAACAGGTGGTGGCGGCC-3’ and subcloning into the AgeI/NotI sites of pLPS-3’-EGFP resulting in the exchange of the GFP by the DsRed2 coding sequence . The human CD105 cDNA was transferred by Cre-mediated recombination from pDNR-dual CD105 [16] into pLPS-3'-RFP2 resulting in RFP2 fused to the carboxy-terminus of full-length CD105 . GFP-fusion proteins of the human CEACAM1 , CEA , and CEACAM8 amino-terminal IgV-like domains as well as the corresponding constructs encoding the amino-terminal IgV-like domains of CEACAM1 orthologues from dog ( cCEACAM1 ) , the two CEACAM1 alleles from cattle ( bCEACAM1a; bCEACAMb ) , and from mouse ( mCEACAM1 ) have been constructed and employed previously [20 , 63] . 293 cells were transfected with 5 μg plasmid DNA encoding secreted GFP-fusion proteins of the N-terminal domains of human or mammalian CEACAM family members . After 24 h , the medium of the transfected 293 cell cultures was replaced by serum-reduced OptiMEM ( Life Technologies , Darmstadt , Germany ) . Two days later the culture supernatants ( supe ) containing the secreted fusion proteins were collected , centrifuged for 15 min at 5000 rpm and either stored at -20°C or used immediately for bacterial pull-down experiments . The GFP-derived fluorescence was analysed using a Varioskan Flash reader ( Thermo Scientific ) to adjust equal amounts of the secreted fusion proteins . For pull-down experiments , indicated bacteria were suspended in PBS and binding to the indicated receptor protein contained in cell culture supernatants was determined essentially as described [63] . Immunofluorescence staining , cell lysis and Western blotting were performed as described previously [62] using mAbs against CEACAMs ( clone D14HD11 ) against CEACAM1 ( clone GM-8G5 ) , against CEACAM6 ( clone 9A6 ) ( all from Aldevron , Freiburg , Germany ) , against CEA ( clone COL-1; Zymed , San Francisco , CA ) , against green fluorescent protein ( GFP; clone JL-8; BD Biosciences ) , against E . coli LPS ( AbD Serotec , Oxford , UK ) , or against murine CD105 ( clone MJ7/18; Southern-Biotech , Birmingham , USA ) . Mouse monoclonal antibodies against human CD105 ( clone P4A4; provided by Developmental Studies Hybridoma Bank ( DSHB ) , University of Iowa ) or against Opa proteins ( clone 4B12/C11; generous gift of Marc Achtman , University of Warwick , UK ) as well as rat monoclonal anti-integrin β1 ( clone 9EG7; provided by D . Vestweber ( MPI for Molecular Medicine , Münster , Germany ) ) and rat monoclonal anti-integrin β1 ( clone AIIB2; DSHB ) were purified from hybridoma culture supernatants . Further , rabbit polyclonal antisera raised against paraformaldehyde-fixed N . gonorrhoeae MS11 ( IG-511 ) was custom produced by Immunoglobe ( Himmelstadt , Germany ) . All secondary antibodies were from Jackson Immuno-Research ( West Grove , PA ) . For scanning electron microscopy , ME-180 cells were seeded at 2 . 5 x 105 cells/well in 24-well plates on acid-washed glass coverslips coated with 25 μg/ml collagen and grown to confluency . Medium was replaced with DMEM , 0 . 5% FCS for 8 h . Then cells were infected for 14 h at a MOI of 20 or left uninfected . Samples were fixed in situ for at least 1 h at 4°C . The samples were washed and dehydrated in a graded series of aceton on ice . After critical point drying samples were sputter-coated with 5 nm gold-palladium in a BAL-TEC SCD 030 and examined at 15 kV accelerating voltage in a Philipps 505 scanning electron microscope using the secondary electron detector . Images were digitally recorded with a DISS 5 system ( remX GmbH , Bruchsal , Germany ) and processed in Adobe Photoshop 6 . C57BL/6J mice transgenic for human CEA ( CEAtg mice ) have been described before [14 , 23] . Wildtype C57BL/6J mice ( originally obtained from Elevage Janvier , Le Genest Saint Isle , France ) and CEAtg mice as well as TLR4+/+ ( HeN ) and TLR-/- ( HeJ ) female mice were maintained under specified pathogen-free conditions under a 12-h light cycle in the animal facility of University of Konstanz in accordance with the institutional guidelines . The CEAtg mice were kept heterozygous for the transgene by crossing male CEAtg mice with female WT mice . Offspring of these crosses ( age 3–4 weeks old ) was genotyped by PCR . Experiments involving animals were performed in accordance with the German Law for the Protection of Animal Welfare ( Tierschutzgesetz ) . The animal care and use protocol , including the protocol of experimental vaginal infection of female mice , was approved by the appropriate state ethics committee and state authorities regulating animal experiments ( Regierungspräsidium Freiburg , Germany ) under the permit file numbers G-10/108 and G-15/43 . Experimental vaginal infection of female mice with E . coli strains was performed as previously described for N . gonorrhoeae [14] . Briefly , CEAtg and wildtype mice were subcutaneously injected with 17-β-estradiol 4 days prior to infection . The drinking water was supplemented with ampicillin ( 1 mg/ml ) to reduce the overgrowth of commensal bacteria during hormone treatment . Ampicillin containing water was changed to normal drinking water 1 day before infection . Mice were inoculated intravaginally with 106 CFU of the different E . coli strains suspended in 20 μl of PBS . 24h later , the mucosa-associated bacteria were re-isolated by cotton swaps . Serial dilutions of re-isolated bacteria were plated on LB agar ( E . coli AfaE-III ) or LB agar containing ampicillin ( E . coli OpaCEA and E . coli ) . In order to analyse the Opa protein profile of re-isolated bacteria , single colonies were expanded on agar plates , lysed and analysed by Western blotting using Opa specific antibodies . The genital tract of infected animals was excised and the longitudinally opened vaginal and uteral tissue was mounted and analysed by scanning electron microscopy essentially as described previously [14] . For immunohistochemistry , tissue samples were immediately fixed with 4% paraformaldehyde for at least 24 h and transferred to 10% sucrose , 0 . 05% cacodylic acid for 1 h at 4°C . Next , samples were transferred to 20% sucrose for 1 h and then into 30% sucrose at 4°C over night . Organs were mounted in embedding medium ( Cryo-M-Bed; Bright Instrument , Huntingdon , UK ) and frozen at -20°C . 10 μm thick sections were cut at -20°C using a cryostat ( Vacutom HM500 , Microm , Germany ) . Sections were incubated with a mouse monoclonal antibody against CEA ( clone COL-1; dilution 1:200 ) or a rat monoclonal antibody against murine CD105 ( clone MJ7/18; dilution 1:1200 ) together with a polyclonal rabbit antibody against N . gonorrhoeae MS11 ( dilution 1:100 ) or a polyclonal rabbit antibody against E . coli ( dilution 1:200 ) . Detection of the primary antibodies was accomplished by incubation with a combination of Cy5-conjugated goat-anti-rabbit antibody ( 1:250 ) and rhodamine-conjugated goat-anti-rat antibody ( 1:250; in the case of CD105 detection ) or Cy3-conjugated goat-anti-mouse antibody ( 1:250; in the case of CEA detection ) . Cell nuclei were visualized by the addition of Hoechst 33342 ( 1:30 , 000; Life Technologies , Darmstadt , Germany ) in the final staining step . Samples were analysed with a TCS SP5 confocal laser scanning microscope ( Leica , Mannheim , Germany ) . Images were digitally processed with Photoshop CS ( Adobe Systems , Mountain View , CA ) and merged to yield pseudo-coloured images . 293 cells were transfected with pcDNA3 . 1 Hygro ( pcDNA ) , pcDNA-CEA , or pLPS-3’-RFP2-CD105 . Cells were infected or not with E . coli , E . coli OpaCEA , E . coli AfaE-III , or gonococci for 14 h . After infection , cells were stained with monoclonal antibodies against CEA ( clone COL-1 ) or against human CD105 ( clone P4A4 ) for 1 h at 4°C . Following washing , samples were stained with a Cy2-conjugated goat-anti-mouse antibody for 30 min , 4°C . ME180 cells and hVEC cells were analysed for the presence of endogenous CEACAMs using mouse monoclonal antibodies specific for CEACAM1 ( clone GM-8G5 ) , CEACAM6 ( clone 9A6 ) , or CEA ( clone COL-1 ) . Stained samples were analysed for Cy2-derived fluorescence by flow cytometry on an LSRII ( BD Biosciences ) using FACS Diva software . Cell adhesion and cell detachment assays were performed essentially as described [14 , 16] . Briefly , the wells of 96-well plates were coated with PBS containing collagen type 1 from calf skin ( ICN Biomedicals , Irvine , CA ) for 24 hours at 4°C . 293 cells were transfected with pEGFP-N1 ( GFP ) , CEA or pLPS3’-CD105 ( CD105 ) . After serum starvation , cells were infected or not with the indicated bacterial strains at a MOI of 30 for 8 hours . Following infection , the cells were detached and kept in suspension medium ( DMEM , 0 . 2% BSA ) with or without 1 mM Mn2+ ( 1 h at 37C° ) , and then replated at 4 x 104 cells/well onto collagen-coated wells . Cells were allowed to adhere for 90 min in the presence or absence of 1 mM Mn2+ at 37°C , before non-adherent cells were removed by gentle washing with PBS . For the Mn2+-treated samples , the washing buffer also contained 1 mM Mn2+ . Adherent cells were fixed and stained for 60 min with 0 . 1% crystal violet in 0 . 1 M borate , pH 9 . After washing and drying , the crystal violet was eluted in 10 mM acetic acid and the staining intensity was measured at 550 nm with a Varioskan Flash ( Thermo Fisher Scientific Oy Microplate Instrumentation ( Vantaa , Finland ) . For measuring integrin activity , cells were serum-starved over-night , before they were infected or not with the indicated bacteria at an MOI of 30 for 8 h . Following infection , cells were detached by limited trypsin/EDTA digestion that was stopped by addition of soybean trypsin inhibitor ( 0 . 5 mg/ml in DMEM ) . Detached cells were kept in suspension medium ( DMEM , 0 . 2% BSA ) for 1 h at 37°C , and then replated at 5 x 104 cells/well into wells coated with collagen type 1 . After 75 min , wells were treated or not with 1 mM MnCl2 , incubated for 5 min and transferred to ice . Cells were fixed with 4% paraformaldehyde in PBS for at least 30 min , washed with PBS and permeabilized with Triton X-100 ( 0 . 1% in PBS ) for 15 min . Cells were washed with PBS and blocked with 2% BSA in PBS ( blocking buffer ) for 20 min , before incubation for 1 h with rat monoclonal antibodies against active integrin ( clone 9EG7; dilution 1:600 ) or against total integrin ( clone AIIB2; dilution 1:750 ) in blocking buffer as described [14] . After washing and incubation with ProteinA/G-HRP ( 1:250 ) , 100 μl/well of substrate solution ( substrate solution was prepared by mixing 10 ml of 2 . 4 mg/ml tetramethylbenzidine in 10% acetone , 90% ethanol with 0 . 5 ml of 30 mM potassium citrate , pH 4 . 1 ) were added . The enzymatic colour reaction was stopped using 2 M H2SO4 ( 100 μl/well ) and the absorbance was determined at 450 nm in a Varioskan Flash ( Thermo Scientific ) microplate reader . For cell adhesion , cell detachment , and integrin activity assays , values were analysed for normal distribution and mean values were compared by two-tailed unpaired t-test . For in vivo infection assays , including enumeration of exfoliating cells , differences between samples were assessed using the Mann-Whitney U-test . Differences between samples with p<0 . 001 are indicated by *** .
|
Mucous surfaces are a hallmark of the nasal cavity and the throat as well as the intestinal and urogenital tracts . These surfaces serve as primary entry portals for a large number of pathogenic bacteria . To get a foothold on the mucosa , bacteria not only need to tightly attach to this tissue , but also need to overcome an intrinsic defence mechanism called exfoliation . During the exfoliation process , the outermost cell layer , together with attached bacteria , is released from the tissue surface reducing the microbial burden . A comprehensive understanding of the molecular strategies , which bacteria utilize to undermine this host defence , is currently lacking . Our results suggest that different bacterial pathogens have found a surprisingly similar answer to this problem by targeting a common set of proteins on the tissue surface . Accordingly , these bacteria express unrelated proteins that engage the same host receptors called CEA-related cell adhesion molecules ( CEACAMs ) . Binding of microbes to CEACAMs triggers , via intracellular signaling pathways , an increased stickiness of the infected cells . Thereby , the pathogens suppress the release of superficial host cells from the tissue and effectively block exfoliation . Detailed mechanistic insight into this process and the ability to manipulate exfoliation might help to prevent or treat bacterial infections .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"bacteriology",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"hek",
"293",
"cells",
"pathogens",
"biological",
"cultures",
"microbiology",
"neisseria",
"gonorrhoeae",
"cloning",
"epithelial",
"cells",
"bacterial",
"diseases",
"integrins",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"adhesins",
"infectious",
"diseases",
"neisseria",
"microbial",
"physiology",
"escherichia",
"coli",
"infections",
"cell",
"adhesion",
"medical",
"microbiology",
"extracellular",
"matrix",
"animal",
"cells",
"microbial",
"pathogens",
"biological",
"tissue",
"cell",
"lines",
"molecular",
"biology",
"bacterial",
"physiology",
"cell",
"biology",
"anatomy",
"virulence",
"factors",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2016
|
Uropathogenic E. coli Exploit CEA to Promote Colonization of the Urogenital Tract Mucosa
|
Trachoma control programs utilize mass azithromycin distributions to treat ocular Chlamydia trachomatis as part of an effort to eliminate this disease world-wide . But it remains unclear what the community-level risk factors are for infection . This cluster-randomized , controlled trial entered 48 randomly selected communities in a 2×2 factorial design evaluating the effect of different treatment frequencies and treatment coverage levels . A pretreatment census and examination established the prevalence of risk factors for clinical trachoma and ocular chlamydia infection including years of education of household head , distance to primary water source , presence of household latrine , and facial cleanliness ( ocular discharge , nasal discharge , and presence of facial flies ) . Univariate and multivariate associations were tested using linear regression and Bayes model averaging . There were a total of 24 , 536 participants ( 4 , 484 children aged 0–5 years ) in 6 , 235 households in the study . Before treatment in May to July 2010 , the community-level prevalence of active trachoma ( TF or TI utilizing the World Health Organization [WHO] grading system ) was 26 . 0% ( 95% CI: 21 . 9% to 30 . 0% ) and the mean community-level prevalence of chlamydia infection by Amplicor PCR was 20 . 7% ( 95% CI: 16 . 5% to 24 . 9% ) in children aged 0–5 years . Univariate analysis showed that nasal discharge ( 0 . 29 , 95% CI: 0 . 04 to 0 . 54; P = 0 . 03 ) , presence of flies on the face ( 0 . 40 , 95% CI: 0 . 17 to 0 . 64; P = 0 . 001 ) , and years of formal education completed by the head of household ( 0 . 07 , 95% CI: 0 . 07 to 0 . 13; P = 0 . 03 ) were independent risk factors for chlamydia infection . In multivariate analysis , facial flies ( 0 . 26 , 95% CI: 0 . 02 to 0 . 49; P = 0 . 03 ) and years of formal education completed by the head of household ( 0 . 06 , 95% CI: 0 . 008 to 0 . 11; P = 0 . 02 ) were associated risk factors for ocular chlamydial infection . We have found that the presence of facial flies and years of education of the head of the household are risk factors for chlamydia infection when the analysis is done at the community level . ClinicalTrials . gov NCT00792922
Trachoma is an ocular infection caused by Chlamydia trachomatis . The World Health Organization ( WHO ) has implemented treatment guidelines that include mass antibiotic administration , facial cleanliness campaigns and environmental improvements in an effort to control and ultimately eliminate the disease [1] . The risk factors for infection are poorly characterized on a community level . This study establishes a knowledge base that assesses these community-level risk factors in an effort to guide the WHO and other programmatic efforts . Some of the risk factors thought to be predictive of trachoma include facial cleanliness ( nasal or ocular discharge and presence of flies on the face ) , access to a latrine , access to clean water , and education . Many studies have identified individual-level risk factors , such as hygiene and latrine use [2] , [3] , but trachoma is a communicable disease with community-level risk factors and treatment . Individuals can be affected by their neighbors , regardless of whether they themselves are compliant with their prescribed antibiotic treatment or adherent with other recommended environmental changes . Here , we analyze community-level risk factors for ocular chlamydia infection in Niger in an effort to improve trachoma control and elimination strategies . In this arm of the Partnership for the Rapid Elimination of Trachoma ( PRET ) trial [4] , [5] , we assess 48 communities before treatment and determine the importance of several community-level risk factors for ocular chlamydial infection .
Niger , in the Sahel region of West Africa , is one of the poorest and least developed countries in the world , ranking in the bottom 1% of countries in the 2010 United Nations Human Development Index [6] . Trachoma is endemic within Niger's population of roughly 14 million [7] , with estimated regional prevalences of between 5% and 49% in children aged 1–9 years [8] . Our study methods have been previously described [4] and are briefly summarized below . In Niger , contiguous administrative units are known as grappes , and are referred to as communities in this manuscript . A community is the smallest population unit for which health services are organized and within which trachoma programs are implemented . The study took place in the Matameye district in the Zinder region of Niger . Communities were selected from among 6 health centers ( Centre de Santé Intégrée or CSIs ) and were eligible for inclusion if they had an estimated total population of between 250 to 600 persons , generally encompassing between 50 and 100 children in the eligible age range for treatment . Other community inclusion criteria were distance >4 kilometers from the center of any semi-urban area ( communities which are close to an urban center are believed to have a lower prevalence of trachoma ) , and prevalence of active trachoma ( TF and/or TI ) ≥10% in children aged 0–5 years . There were a total of 235 eligible communities in the 6 CSIs of which 72 ( 31% ) satisfied the inclusion criteria for community size and 48 of these were selected for inclusion in the study . In a 2×2 factorial design , 48 communities were randomly allocated into 4 treatment arms with 12 communities in each arm ( Figure 1 ) . Randomization of communities and sentinel individuals to the treatment arms was done using RANDOM and SORT functions in Excel ( Version 2003 ) by TP and BN . Note that only pretreatment results are presented here . To determine the impact of mass antibiotic administration on clinical trachoma and ocular chlamydia infection , a random sample of 50 to 100 children aged 0 to 5 years was established as the sentinel group for the study in all enrolled communities prior to treatment . No adjustments were made for missing individuals from the census and all analyses were performed at the community level on an intention-to-treat basis . Baseline data were collected on the following measures: trachoma clinical grade , facial cleanliness , ocular swabs , and ocular photographs . Grader validation was done in a 2-step process: in the first step , research leaders attended a workshop conducted in February 2008 in Ethiopia where trachoma is hyper-endemic , to standardize methods for the trial . Certification of researchers for trachoma grading required a chance corrected agreement ( kappa statistic ≥0 . 6 ) with an experienced grader ( RB ) over the scoring signs of clinically active trachoma ( TF and/or TI in the WHO system ) in validation exercises in both the classroom ( photographic collection ) and the field . In the second step , clinical graders in Niger were eligible to perform ocular grading for the trial if they had attained a chance corrected agreement ( kappa statistic ≥0 . 6 ) with a certified grader over the scoring signs of clinically active trachoma . The pretreatment visit for this trial was conducted in Niger from May to July 2010 , where 3 ophthalmic nurses ( TSO's , or Technicien Superior en Ophtalmologie ) received a kappa score on photo grading validation of 0 . 96 , 1 . 00 and 0 . 88 ( against senior grader RB ) . A population census of all 48 study communities was conducted by trained personnel , masked to the treatment arm , prior to the baseline visit . Study personnel were also masked to the prevalence of clinical trachoma and ocular chlamydia infection . The census team moved from house to house , enumerating all residents in all households , collecting baseline characteristics information . The 100% census allowed us to create a sampling frame from which we could randomly select sentinel children with equal probabilities . After obtaining consent , conjunctival examination for trachoma and conjunctival swabbing for chlamydial PCR were performed on all sentinel children [9] , [10] . Clinical grading of the right everted superior tarsal conjunctiva was performed using a 2 . 5× magnifying loupe and adequate sunlight or a torch light according to the WHO simplified grading system [11] . Prior to conjunctival swabbing , a trained photographer took a minimum of 2 photographs of the right eyelid of all child participants using a Nikon D-series camera and a Micro Nikon 105 mm; f/2 . 8 lens ( Nikon , Tokyo , Japan ) . After conjunctival examination , a Dacron swab was passed firmly 3 times over the right upper tarsal conjunctiva , rotating 120 degrees between each pass [9] , [10] . To assess for field contamination , negative field control swabs were collected from 5 randomly selected children in each community . The negative field controls were collected immediately after the ocular swabs were collected; a new swab was passed within 5 cm of , but not contacting , the participant's exposed conjunctiva [9] . Examiners changed gloves before examining each new participant . All of the samples were placed immediately at 4°C in the field and frozen at −20°C within 10 hours . Swabs were shipped at 4°C to University of California , San Francisco , CA , USA , where they were stored at −80°C until processing [12] . The Amplicor PCR assay ( Roche Diagnostics , Branchburg , NJ , USA ) was used to detect C . trachomatis DNA and samples were pooled for processing to save time and cost , as previously described [13] . For laboratory standardization and validation between PRET study sites , a set of 20 samples ( both positive and negative ) was sent to a laboratory at Johns Hopkins University ( Baltimore , MD , USA ) for processing . Results were considered valid only when agreement between labs was at least 90% . During the census , community-level data were collected on household characteristics , including number of years of education completed by head of household; distance to the primary water source >30 min walk; presence of a latrine; and awareness of a programmatic face-washing campaign in the community within the previous year . Facial cleanliness was measured for each child who presented for an exam , including the presence of ocular discharge ( on the eyelashes or eyelids ) , nasal discharge ( on nares , cheeks , or lips ) , and presence of flies on face ( presence of ≥1 fly on the face while observing for 3 seconds ) . Data were entered into a database ( MS Access v2007 ) developed at the Dana Center , Johns Hopkins University as previously described [5] . Data were double-entered by different data entry clerks and discrepancies were resolved by reference to the original forms . Queries of all data discrepancies were identified using MS Access prior to any analysis . The intention-to-treat statistical analysis was performed using the statistical package R ( http://www . r-project . org ) , 2 . 12 for MacIntosh . All pretreatment characteristics were summarized at the community level . Univariate associations with ocular chlamydia infection in children aged 0–5 years were tested using linear regression , with permutation-based significance testing because of non-normality . Multivariate regression was conducted using demographic and clinical ( non-trachoma ) predictors by selecting the best univariate predictors ( no more than 4 included predictors , because the number of communities was only 48 ) . Because stepwise regression may result in overfitting [14] , we used Bayes model averaging based on the Bayes information criterion to conduct multiple regression ( R package BMA , v . 3 . 14 ) [15] . This technique yields estimates for each regression coefficient and a posterior probability that each differs from zero ( posterior effect probabilities ) . Because of heteroskedasticity of the outcome variable , we reported confidence intervals derived from bootstrap resampling of the residuals following ordinary linear regression [16] ( with 10 , 000 replications ) for each regression coefficient . Ethical approval for this study was obtained from the Committee for Human Research of the University of California , San Francisco and le Comite Consultatif National d'Ethique du Ministere de la Sante Publique , Niger ( Ethical Committee , Niger Ministry of Health ) . Oral consent was obtained from the village leaders , and written ( thumbprint ) consent from the study participant ( or the child's parent or guardian ) at the time of examination . The study was carried out in accordance with the Declaration of Helsinki . An independent data and safety monitoring committee appointed by the PRET study executive committee oversaw the design and implementation of the study and performed annual reviews of quality assurance and adverse events .
The 48 randomly selected communities included 24 , 536 total individuals in 6 , 235 households ( Table 1 ) . Exams were conducted and ocular swabs collected from a total of 4 , 484 children aged 0–5 years from May to July 2010 . The mean prevalence of clinical trachoma ( TF ) was 26 . 0% ( 95% CI: 21 . 9% to 30 . 0% ) and the mean prevalence of ocular C . trachomatis infection was 20 . 7% ( 95% CI: 16 . 5% to 24 . 9% ) in the 48 communities ( Table 2 ) . The community-level prevalence of infection and the community-level prevalence of TF were highly correlated ( Pearson correlation 0 . 63 , 95% CI: 0 . 42 to 0 . 78; P<0 . 001 ) . A total of 238 negative field control swabs were collected; 12 ( 5 . 0% ) were PCR positive for chlamydia . As part of a quality control investigation , it was determined that 11 of these 12 ( 91 . 7% ) field controls were collected by a single examiner and steps were immediately taken to improve the field training for specimen collection in all field workers to reduce contamination . All positive field controls had corresponding positive ocular swabs . The pre-specified level of agreement between laboratories at UCSF and Johns Hopkins University was maintained at >90% . In the univariate model ( Table 3 ) using simple linear regression , there was increased risk of ocular chlamydia infection in communities where children had nasal discharge ( 0 . 29 , 95% CI: 0 . 04 to 0 . 54; P = 0 . 03 ) or flies on the face ( 0 . 40 , 95% CI: 0 . 17 to 0 . 64; P = 0 . 001 ) . Other significant predictors of chlamydia infection were years of education of the head of household ( 0 . 07 , 95% CI: 0 . 007 to 0 . 13; P = 0 . 03 ) , proportion of children aged 0 to 1 year ( −0 . 96 , 95% CI: −1 . 73 to −0 . 19; P = 0 . 02 ) and prevalence of clinical trachoma TF ( 0 . 65 , 95% CI: 0 . 42 to 0 . 88; P<0 . 001 ) ( Table 3 ) . In the multivariate analysis using Bayes model averaging ( Table 4 ) , among the measured risk factors , chlamydia infection in communities was associated only with flies on face ( 0 . 26 , 95% CI: 0 . 02 to 0 . 49 , P = 0 . 03 ) and level of education of head of household ( 0 . 06 , 95% CI: 0 . 008 to 0 . 11 , P = 0 . 02 ) . Other characteristics , including gender , ocular discharge , nasal discharge , number of individuals per household , water access >30 mins , and latrine access , were not significant risk factors in this model . Ordinary linear regression gave confidence intervals which were slightly wider than those derived from bootstrap resampling .
We have found that communities with higher percentages of younger children , nasal discharge , facial flies , and number of years of education of the head of the household are associated with higher community prevalence of chlamydia infection in univariate analysis . Community-level facial flies and years of education of head of household are significantly associated with the prevalence of chlamydia infection in multivariate analysis . Flies are thought to be important in trachoma transmission but studies have been done only on the individual level [2] , [5] , [17]–[19] or have used clinical outcomes rather than more objective laboratory measurements [2] , [17] , [18] . Chlamydia DNA has been found on 15% of flies in areas hyperendemic for trachoma [20] and so flies are believed to be vectors for the spread of chlamydia in endemic areas [21] . Our study is the first to show the association between community-level fly density flies and community-level ocular chlamydia infection . Other studies performed on the individual-level have concluded that general education is associated with less trachoma [5] , [22]–[25] , but our study in Niger found the opposite and chlamydia infection was correlated directly with the self- reported years of education of the head of the household . Trachoma is felt to be disease which aggregates among the poor and uneducated [26] , and studies have shown health education can improve trachoma control [27] . However , a study in Mali showed the odds of trachoma was higher in households where children attended a traditional school compared to households where children attended a modern school . Mali shares a border with Niger of greater than 860 km and is likely more similar to Niger than more distant areas . The structure of the schools and the way in which children congregate in these schools is just as important as the years of education that have been received . We have used community-level predictors and community-level outcomes because trachoma is a communicable disease; poor hygiene and specific behavior of others in the same household and neighborhood may increase the risk of infection in the entire community [26] , [28]–[30] . For these reasons the WHO strategy for reducing trachoma is implemented at the community level with community-wide interventions . Other trachoma studies have been cluster-randomized but then analyzed at the level of the individual [10] , [27] , [31] . Like our study in Niger , the Tanzania arm of the PRET study found that facial flies are associated with ocular chlamydia infection [5] . However , this individual-level study also showed that facial cleanliness is a risk for ocular chlamydia infection , an association which we were unable to identify . Reinfection is known to occur at the community level and an individual-level approach does not take this into account . In our study , we look for risk factors for higher community prevalence of ocular chlamydia infection . The clinical exam for trachoma has been shown to be unreliable and poorly correlated with infection in some situations [32]–[34] . We chose to use the more objective , masked , outcome of ocular chlamydia infection by PCR in our study . Note that we did include the clinical exam in the multivariate predictor model of infection by design , because TF and TI are consequences of infection rather than causes of it . However , in a different context , programs that have access to clinical surveys may be interested in the associated level of infection that we have measured here ( Table 5 ) . It is important to keep in mind the clinical exam for trachoma ( TF ) is close to 90% sensitive but only 30% specific in latent class analysis , leading to overtreatment in some situations [35] . Furthermore , the poor correlation between the clinical signs of trachoma and the laboratory evidence of infection with C . trachomatis , becomes more problematic as the community-level of infection decreases following mass treatment [36] . Nevertheless , the clinical exam is inexpensive and easy to perform . It will remain an important tool for the WHO in their treatment guidelines and continued efforts to understand the relationship between clinical trachoma and infection with the causative bacteria is critical . There are several limitations to this study . First , we evaluated only a sample of sentinel children in the communities rather than all individuals and this may have produced bias in our estimates . Note that laboratory workers were masked to the identities of the sentinel children chosen and the communities in which they lived . Second , although all of the selected communities for inclusion in the study were evaluated successfully as planned , some individuals who were randomly selected for inclusion were missing; if individuals were not missing at random , this could also have created bias . All analyses were done on an intention-to-treat basis with no adjustments for missing individuals . Third , the evidence for risk factors of infectious ocular chlamydia that we have found in rural Niger may not be generalizable to other trachoma-endemic areas in other countries and continents because of differences in important , unmeasured cultural or environmental characteristics . The finding that more years of education of the head of household is associated with higher levels of chlamydia infection is counterintuitive . The study questionnaire asked specifically for the number of years of formal education that were completed . Religious education , a form of learning that is very common in this area , was not included in this estimate . If the education of adults is by methods other than formal education in households , our measurement could have been an underestimate of the total amount of education interfering with our ability to capture an association with infection . In summary , we have found that facial flies and years of education of the head of the household are associated with community-level prevalence of ocular chlamydia infection . Further analyses will be performed as treatment begins in the study and continues over the next 3 years .
|
Trachoma is one of the most important neglected tropical diseases because it is the leading cause of blindness from an infection in the world . There are about 1 . 3 million persons blind from the disease and many more at risk of blindness in the future . It is caused by the common bacterium Chlamydia trachomatis and can be treated with mass drug administrations ( MDA ) of azithromycin . We have begun a clinical trial in Niger , a country with limited resources in Africa , to determine the best treatment strategy . Our study from May to July 2010 , which began before MDA's were given , showed that 26% of children aged 0–5 years were infected with the disease . In these children , we found that discharge from the nose , presence of flies on the face , and the number of years of education completed by the head of the household were risk factors for infection in 48 different communities . We hope to use this information about risk factors of infection to help guide future studies for trachoma and also to help with the WHO goal of eliminating the disease worldwide by the year 2020 .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"ophthalmology",
"neglected",
"tropical",
"diseases",
"trachoma"
] |
2012
|
Community Risk Factors for Ocular Chlamydia Infection in Niger: Pre-Treatment Results from a Cluster-Randomized Trachoma Trial
|
How neuronal diversity emerges from complex patterns of gene expression remains poorly understood . Here we present an approach to understand electrophysiological diversity through gene expression by integrating pooled- and single-cell transcriptomics with intracellular electrophysiology . Using neuroinformatics methods , we compiled a brain-wide dataset of 34 neuron types with paired gene expression and intrinsic electrophysiological features from publically accessible sources , the largest such collection to date . We identified 420 genes whose expression levels significantly correlated with variability in one or more of 11 physiological parameters . We next trained statistical models to infer cellular features from multivariate gene expression patterns . Such models were predictive of gene-electrophysiological relationships in an independent collection of 12 visual cortex cell types from the Allen Institute , suggesting that these correlations might reflect general principles relating expression patterns to phenotypic diversity across very different cell types . Many associations reported here have the potential to provide new insights into how neurons generate functional diversity , and correlations of ion channel genes like Gabrd and Scn1a ( Nav1 . 1 ) with resting potential and spiking frequency are consistent with known causal mechanisms . Our work highlights the promise and inherent challenges in using cell type-specific transcriptomics to understand the mechanistic origins of neuronal diversity .
A major goal of neuroscience has been to understand the mechanistic origins of neuronal electrophysiological phenotypes . Such electrical features help define the computational functions of each neuron [1 , 2] , and further , specific electrophysiological deficits contribute to brain disorders such as epilepsy , ataxia , and autism [3–5] . The molecular basis of neuron electrophysiology is complex . There are over 200 mammalian ion channel and transporter genes whose products influence a neuron’s electrophysiological phenotype [6–9] . Numerous additional genes regulate channel functional expression through initiating gene transcription and alternative splicing , post-translational modifications , and trafficking channels to and from the membrane surface [10–12] . Even morphological features contribute to cellular electrophysiology [13] . Recent genetic studies in human epileptic and neuropsychiatric patients provide convergent evidence , as mutations in many genes reflecting multiple functional pathways are associated with these disorders [4 , 14–16] . In light of this complexity , the gold standard employed by neurophysiologists is to use gene knockouts or pharmacology to assay how electrophysiological function changes following protein disruption [7 , 8] . However , these single-gene focused methods are relatively low-throughput and many potentially relevant genes have yet to be studied for their electrophysiological function . Cell type-specific transcriptomics , enabling genome-wide assay of quantitative mRNA expression levels , provides a lucrative avenue for discovering novel genes that might contribute to specific aspects of cellular physiology [17 , 18] . Correlation-based approaches have been proposed that pair single-cell expression profiling with patch-clamp electrophysiology [19–21] . These approaches leverage the biological variability observed across a collection of cells to identify gene expression patterns correlated with cellular phenotypic differences . Generalizing from these studies has proven challenging however , since they typically have been focused on a limited number of cell types . Similarly , and perhaps more critically , there are typically hundreds to thousands of genes correlated with electrophysiological variability[22] . Thus it has been difficult from this data to pin down how individual genes might shape specific cellular phenotypes . Though making use of larger and more diverse collections of cell types could provide a potential solution , collecting such reference data is immensely resource- and labor-intensive . Here , we present an approach for correlating cell type-specific transcriptomics with neuronal electrophysiological features . We leverage neuroinformatics methods to build a novel reference dataset on brain-wide neuronal gene expression and intrinsic electrophysiological feature diversity . The compiled dataset reflects the neuronal characterization efforts of hundreds of investigators as well as our efforts to compile and normalize these data for unified mega-analysis [23–25] . From this data , we identified hundreds of genes whose expression levels significantly correlate with specific electrophysiological features ( e . g . , resting potential or maximum spiking frequency ) . Illustrating the generalizability of these results , we could use these correlations to predict the ephys parameters of an independent neocortex-specific dataset from the Allen Institute . In addition , many of these genes have been further found to directly regulate neuronal electrophysiology , suggesting that some of the correlations reported here likely reflect novel causal relationships . Our findings present a major step for understanding how a multitude of genes contribute to cell type-specific phenotypic diversity .
To construct our primary dataset for gene-ephys correlation analysis , we adapted and combined two databases developed and curated by our group . The first , NeuroExpresso , a database containing microarray-based transcriptomes collected from samples of purified mouse brain cell types under normal conditions [23] . The second , NeuroElectro , a database of rodent neuronal electrophysiological profiles manually curated from the published literature reflecting intracellular ephys characterization of normal , non-treated cells [24 , 25] . From NeuroElectro’s initial publication , we have massively expanded the resource from 331 to 968 articles and have made essential improvements that allow more fine-grained annotation of neuron subtypes and curation of more electrophysiological features . Given the methodological heterogeneity of the primary data comprising these databases , we applied a number of quality control filtering and cross-laboratory standardization approaches ( see Methods and S1 Fig ) . These include careful re-analysis of neuron type-specific transcriptomes for cellular contamination ( e . g . , astrocytes , glia ) and statistical approaches to normalize ephys measurements for lab-specific experimental conditions ( e . g . , animal age and slice recording temperatures ) . We obtained neuron type-specific paired gene expression and ephys data by carefully aligning these databases on cell type identity , making use of our detailed annotations of each sample’s specific cell type ( Fig 1A , left ) . This harmonization allows us to merge cell types defined using orthologous criteria , e . g . , gene expression data derived from transgenic lines with ephys data collected from cells defined by traditional morpho-electric criteria [26] . The final “discovery” reference dataset is composed of 34 neuron types sampled throughout the brain and reflects cell types with diverse circuit roles , neurotransmitters , and developmental stages ( summarized in Table 1 and S2 Table ) . For validation we utilized an independent dataset characterizing neurons from adult mouse primary visual cortex collected by the Allen Institute for Brain Science . Here , genetically labeled cells were characterized either for their transcriptomic profiles , using single-cell RNA sequencing ( scRNAseq ) [27] , or their electrophysiological properties , using patch-clamp electrophysiology in vitro with standardized protocols ( http://celltypes . brain-map . org/ ) . Importantly , for both expression and ephys characterization , the same mouse lines for genetically labeling specific populations of cells were used , making it straightforward to combine samples post-hoc , yielding a final “validation” dataset composed of 12 unique cell types ( Table 2 ) . Averaging data across labeled single cells within a mouse line also helps mitigate the influence of cell-to-cell variability and technical “dropouts” in the scRNAseq data [18] . Given the smaller number of cell types present in the AIBS dataset we chose to use these data primarily for validation and generalization of findings made using the discovery dataset . Note that for both the discovery and validation datasets , electrophysiological and gene expression values are from separate cells . Our primary analysis focus was to understand how cell type-specific expression of individual genes might statistically explain the variance in electrophysiological parameters observed across cell types ( Fig 1A , right ) . For example , how does Scn1a ( Nav1 . 1 ) expression correlate with neuronal maximum firing rates ? Which genes are most correlated with cellular resting membrane potentials ? We primarily chose to employ a single-gene focused approach ( utilizing Spearman rank correlations ) because of sample size considerations , reasoning that we did not have enough unique cell types in both the discovery or validation datasets to rigorously pursue a combinatorial gene approach . However , as this single-gene focus limits our ability to identify highly combinatorial and/or redundant or degenerate gene-ephys relationships [28 , 29] , we further pursued a machine learning approach where we used sparse , regularized linear models to relate multivariate gene expression to ephys features . For each of the 34 neuron types in the NeuroExpresso/NeuroElectro discovery dataset , we obtained a gene expression profile for 11 , 509 genes and 5–11 intrinsic electrophysiological properties ( mean = 9 +/- 2 ephys properties per cell type; described in S1 Table ) . We first asked whether there are individual genes whose quantitative mRNA expression levels correlate with systematic ephys diversity in both the discovery and AIBS validation datasets . Using the discovery dataset , after first filtering for genes with sufficiently high and variable expression across cell types ( see Methods ) , we found a total of 653 genes ( of 2694 tested ) correlated with at least 1 of the 11 ephys properties at padj < 0 . 05 ( padj indicates Benjamini-Hochberg false discovery rate adjusted p-value ) . 1095 genes were identified at padj < 0 . 1 and 217 genes were identified at padj < 0 . 01 . As an illustrative example of one gene-ephys correlation , we found that expression levels of the gene Nkain1 correlated with input resistance ( Rin ) values across cell types in the discovery dataset ( Fig 1B and 1C; Spearman correlation , rs = 0 . 86; padj = 1 . 7*10−7 ) . We also saw this trend recapitulated when only considering within-cell type changes observed across cortical basket cell and Purkinje cell development , with Nkain1 expression and Rin decreasing dramatically as these cells mature ( S2 Fig ) . In the AIBS validation dataset , after summarizing the single-cell data to the level of cell types , we further found a consistent Nkain1- Rin correlation amongst adult visual cortex cell types ( Fig 1D; rs = 0 . 71 ) . Little is known about Nkain1 protein function , except that it interacts with the Na+/K+ pump β-subunit and likely modulates the pump’s function and membrane localization [30] . Intriguingly , the Na+/K+ pump has a known role in establishing cellular volumes and input resistance [31] . We provide a summary of the total number of genes identified as significantly correlated with each of the 11 ephys properties in Fig 2A and the full list of gene-ephys correlations in S3 Table . We initially noticed that different ephys properties were significantly correlated with varying numbers of genes . For example , at the somewhat conservative threshold of padj < 0 . 05 , we found no genes correlated with action potential threshold voltage ( APthr ) , despite there being many genes previously implicated with this feature [5 , 32] . In contrast , there were over 200 genes significantly correlated with either Vrest or AHPamp . However , we consider it unlikely that all of these genes reflect a direct causal relationship , as gene-gene correlations driven by gene co-regulation create ambiguity . We note that in the discovery dataset , not all ephys properties were available for each cell type , with 19–34 cell types quantified per ephys property . Furthermore , since correlation p-values are in part related to sample size , we found a positive relationship between the total number of genes associated with each ephys property and the number of cell types where the ephys property was quantified ( R2 = 0 . 30; S3 Fig ) . Next , given that ephys properties tend to be correlated with one another [21 , 25] , we asked if pairs of correlated ephys properties also tend to share associated genes . For example , cellular measurements of membrane capacitance ( Cm ) and Rin are highly anti-correlated ( rs = -0 . 69 in the discovery dataset ) ; furthermore , of the 80 genes significantly associated with Cm , 36 were also associated with Rin . Though some pairs of ephys properties share common biophysical mechanisms and could be thus regulated via common genes ( e . g . , Cm and Rin are both dependent in part on cell size ) , correlations between ephys properties likely limit the specificity of the relationships reported here . We next used the AIBS dataset to validate the significant correlations observed in the discovery dataset . We predicted that gene-ephys correlations discovered in our brain-wide dataset should generalize to the transcriptomic and electrophysiological diversity among adult visual cortex cell types . Because of the limited number of cell types available in the validation dataset relative to the discovery dataset , we were generally underpowered to identify statistically significant relationships using the AIBS dataset alone for most electrophysiological properties ( S3 Table and S4 Table ) . We therefore chose to compare results between the discovery and validation datasets as: 1 ) overall consistency , defined by the global rank correlation between results from the two datasets ( Fig 2B ) ; and 2 ) consistency for the subset of gene-ephys relationships meeting our threshold for significance in the discovery dataset ( padj < 0 . 05 ) . Overall , we found positive , but modest , agreement between the two datasets , with most ephys properties showing a positive correlation ( Table 3 ) . However , APthr , Rheo , and Tau are notable exceptions and might reflect challenges in normalizing these ephys features from the cross-study NeuroElectro database [25] . Focusing specifically on significant gene-ephys correlations identified in the discovery dataset , we found that the majority of these , 61 . 2% , reflecting 420 individual genes , were consistent in the validation dataset , with consistency defined as a matching correlation direction and with an absolute value of rs > 0 . 3 ( Table 3 ) . The degree of consistency between the NeuroExpresso/NeuroElectro and AIBS datasets is encouraging given their dissimilarity in design and content . For example , the AIBS cell types dataset is sampled from a single brain region ( visual cortex ) at one developmental stage ( adult ) . Moreover , there are considerable technical differences between the datasets , such as transcriptome quantification via single-cell RNAseq vs pooled-cell microarrays or between standardized versus heterogeneous ephys data collection . In the remainder of the manuscript , we focus on incorporating multivariate methods and further characterizing the significant gene-ephys correlations from the discovery dataset that have evidence for further validating in the AIBS dataset . Given the relatively high correlation between the expression of single genes and specific ephys properties , we next wondered if we could construct statistical models to predict ephys parameters from gene expression patterns . Using the discovery dataset , we trained sparse , regularized statistical models to predict cell type-specific ephys values from multivariate gene expression ( using a consensus set of 2603 genes with high variance in the discovery dataset that were also available in the AIBS validation dataset ) . Across the set of 11 ephys properties , we used leave-one-out cross-validation ( LOOCV ) to evaluate how well gene expression patterns can predict the ephys parameters of cell types not used for model training . For most ephys properties , such as action potential amplitude ( Fig 3A , R2LOOCV = 0 . 63 ) and maximum firing rate ( Fig 3C , R2LOOCV = 0 . 58 ) , we found considerable predictive power between cell type-specific gene expression and ephys ( summarized results across ephys properties shown in ( Fig 3E ) . We further noted that , qualitatively , ephys properties with relatively poor predictive performance also tended to be those with fewer genes identified as significantly correlated with that feature , such as APthr and APhw ( Table 3 ) . Next , we asked if the statistical models that were originally trained on the discovery dataset could further be used to predict the ephys properties of the cell types in the AIBS validation dataset , even though technical differences would likely limit the accuracy of such cross-dataset prediction . We first applied simple normalizations to help align the RNAseq-based expression values and ephys measurements to those from the discovery dataset ( see Methods ) . After using the models to predict AIBS ephys values from the single cell-based gene expression patterns , we found good accuracy for some ephys properties , such as APamp ( Fig 3B , R2AIBS = 0 . 37 ) and FRmax ( Fig 3D , R2AIBS = 0 . 98 ) . We tended to find similar generalization performance between the discovery and validation datasets for a number of ephys properties , with membrane time constant ( Tau ) and cellular capacitance ( Cm ) being notable outliers ( Fig 3E ) . While individual poorly predicted ephys properties and cell types should be investigated further , these results speak to the generalizability of the gene expression-ephys relationships described here . Such findings suggest that these relationships could be used to potentially inform on cellular phenotypes when only expression data are available . A key question is whether any of the univariate gene-ephys correlations we observed are due to direct causal relationships supported by specific evidence . To this end , we made use of the existing literature on gene-ephys relations . We focused on ion channel genes ( Fig 4A ) , reasoning that these would be most likely to have been directly tested for electrophysiological function . We manually searched the literature for such experiments , since at present this data is not reflected within a comprehensive database ( the current NeuroElectro database reflects experiments done under standard or control conditions , not genetic or pharmacological manipulations ) . We present a brief summary of our gene-centered literature search alongside highlights from our correlation-based analysis below , with the complete results provided in S5 Table . Of 31 significant and validated ion channel-ephys correlations , we found 17 had been directly tested through genetic manipulations or channel-specific pharmacology ( reflecting 12 unique ion channel genes ) . To compare our correlations to individual results from direct experiments , we first mapped our correlations to predicted causal effects; for example , knocking out a gene whose expression is positively correlated with maximum firing rate should tend to lower firing rates , all else being equal . We found that of 17 total tested ion channel-ephys correlations , 11 were consistent with literature evidence , 2 showed mixed evidence , 1 showed no effect on the ephys property , and 3 were inconsistent . Here , we defined inconsistent evidence as those where a predicted increase ( or decrease ) in an ephys property was reflected by a change in the opposite direction in the literature; mixed evidence were those where some manipulations were consistent but others were inconsistent ( e . g . , pharmacology versus gene knockout ) . Below , we provide specific illustrative examples from this literature search . Scn1a , encoding the sodium channel Nav1 . 1 , was positively correlated with maximum firing rate ( Fig 4B; NeuExp/NeuElec rs = 0 . 86 , AIBS rs = 0 . 36 ) , with the highest Scn1a expression observed in adult cortical PV interneurons and Purkinje cells . In a mouse model of Dravet syndrome with a hemizygous gene deletion ( i . e . , Scn1a +/- ) , it was observed that fast-spiking PV interneurons cells could no longer fire at their characteristically high frequencies ( Fig 4C ) , with a smaller but significant effect also observed in Sst-expressing Martinotti cells [5] . However , the same change was not seen in layer 5 pyramidal cells , which express ~3–4 fold less Scn1a relative to PV cells ( in NeuroExpresso and AIBS ) , potentially suggesting that total expression levels might mediate the effect of hemizygous Scn1a deletion . Intriguingly , in a haploinsufficiency model of Dravet syndrome , directly upregulating Scn1a expression using long non-coding RNAs rescued the firing phenotype in PV cells and lowered seizure number and duration [36] . We found 4 ( of 5 total ) ion channel genes correlated with Vrest that were consistent with literature evidence . Hcn3 , encoding a slow HCN channel variant [6] , was positively correlated with Vrest ( Fig 4D; NeuExp/NeuElec rs = 0 . 82 , AIBS rs = 0 . 57 ) . Blocking HCN-current using ZD7288 across multiple cell types consistently made Vrest more hyperpolarized ( Fig 4E ) [34 , 37] . Gabrd , Kcnk1 , and Itpr1 , were each negatively correlated with Vrest and each gene reflects a different mechanistic route towards Vrest hyperpolarization ( Fig 4F and S4 Fig ) . For example , Gabrd encodes the δ-subunit of the GABAA receptor and mediates extrasynaptic tonic inhibition , effectively turning the GABAA receptor into a chloride channel[38] . Thus , increased Gabrd expression , or pharmacologically increasing its activity ( Fig 4F and 4G ) [35] would tend to hyperpolarize cells through the chloride reversal potential ( median ECl = -72 mV , based on reported internal and external solutions ) . Similarly , Kcnk1 , encoding the K2P1 . 1 2-pore potassium channel , hyperpolarizes Vrest through the potassium reversal potential ( EK ~ -100 mV ) [39] . Itpr1 activity releases calcium from intracellular stores and hyperpolarizes Vrest through calcium-activated potassium channels [40 , 41] . Taken together , each of these genes reflect distinct and potentially degnerate routes towards modulating cellular Vrest . We found evidence for two ion channel subunits , Kcna1 and Kcnab2 , regulating multiple distinct electrophysiological properties ( S4 Fig ) . For example , Kcna1 , encoding the delayed rectifier potassium channel Kv1 . 1 , was negatively correlated with action potential half width ( NeuExp/NeuElec rs = -0 . 70 , AIBS rs = -0 . 52 ) and positively correlated with rheobase ( NeuExp/NeuElec rs = 0 . 69 , AIBS rs = 0 . 66 ) . These correlations were corroborated by Kcna1 genetic knockouts or pharmacological block in auditory brainstem neurons and are consistent with known mechanistic insight about Kv1 . 1 function [42–44] . While the previous examples are encouraging , not all of our findings were concordant with previous literature . For example , we saw that Kcnb1 , encoding the Kv2 . 1 channel , was negatively correlated with spike afterhyperpolarization amplitude ( AHPamp ) ( S5A and S5B Fig; NeuExp/NeuElec rs = -0 . 70 , padj = 0 . 0033; AIBS rs = -0 . 62 ) . Based on this correlation , we would expect that decreasing Kv2 . 1 functional expression should increase AHPamp values . However , convergent genetic and pharmacological evidence suggests the opposite: decreasing Kv2 . 1 activity or expression decreases AHPamp values [45 , 46] . Delving deeper , the Kcnb1- AHPamp correlation appears driven in part by gross differences between excitatory and non-excitatory cell types , with excitatory cells strongly expressing Kcnb1 and also having small AHPamp relative to non-excitatory cell types ( S5C Fig ) . Thus though there is likely some mechanistic explanation for why excitatory cells tend to express more Kcnb1 , this does not appear to be directly related to AHPamp per-se . This example suggests that caution is needed before interpreting each correlation reported here as a direct causal relationship . To summarize , we found multiple examples of direct regulation of specific ephys properties by individual genes identified through our correlation-based methodology . In the remainder of the results , we highlight additional genes that may be of relevance in future studies . Encouraged that many of the univariate ion channel gene-ephys associations discovered through our analysis were consistent with previous experimental manipulations , we next expanded our attention to other classes of genes . From the larger list of correlations identified in our analysis ( S3 Table ) , we have highlighted below a small number of individual gene-ephys correlations . Multiple genes known to regulate ion channel functional expression and localization were identified in our analysis ( Fig 5A and 5B ) . For example , two genes regulating the localization of sodium channels , L1cam and Fgf14 , were correlated with Vrest in our analysis and the direction of correlation was further supported by previous experiments [47 , 48] . Along this theme , our analysis identified novel associations between Nedd4l and Slmap with Vrest , Ank1 with maximum firing frequency , and Nkain1 with Rin ( as shown in Fig 1 ) . Nedd4l , identified as an epilepsy gene through whole-exome sequencing [14] , ubiquitinates voltage-gated sodium and potassium channels [49]; Slmap , associated with Brugada syndrome , controls the trafficking and surface expression of voltage-gated sodium channels in cardiac and muscle cells but remains unstudied in neurons [50] . Ank1 , a member of the ankyrin family , has recently been shown to coordinate the localization of specific Nav subunits to nodes of Ranvier [51] . Though we found the highest expression of Ank1 in fast-spiking cells , including Purkinje and PV interneurons , its function remains completely uncharacterized in these cells . We noted several transcription factors in our list of associated genes , including some that have known roles in the nervous system that are compatible with possible , but unknown , roles in the regulation of cellular ephys ( Fig 5C ) . For example , we found Zbtb18 ( a . k . a . , RP58 , Zfp238 ) to be negatively correlated with Vrest . Though Zbtb18 has yet to be studied for its potential electrophysiological effects , this gene has been shown to be required for the normal development of neocortical glutamatergic cells [52 , 53] and its human homolog has recently been identified as a causative gene for autism and neurodevelopmental disorders [54] . As another example , Zscan21 ( a . k . a . , Zipro1 or Zfp38 ) positively correlated with input resistance here and has been shown to be involved in the normal proliferation of progenitor cells into cerebellar granule cells [55] . Among genes correlated with membrane capacitance and input resistance , we noticed that many of these were cytoskeletal proteins or otherwise associated with regulating neuronal differentiation and dendritic morphology , including Cap2 , Chn1 , Stmn4 , Bex1 , and Tpm4 ( S6 Fig ) . In summary , this analysis presents suggestive evidence for many novel gene-ephys relationships . Though we do not expect all of these novel associations to reflect direct causal relationships , by focusing on gene classes that are compatible with possible regulation of ephys , we can further hone the list of associated genes to those that might be of further interest for follow-up investigation .
The results presented here are restricted to a limited range of situations . First , we can only identify genes where mRNA , as measured in dissociated cells [59] , is an adequate readout of a gene’s functional activity at the protein level . Future datasets employing RNA-seq , proteomics , or techniques to capture non-somatic mRNA will likely be able to identify more genes where alternative splicing and post-translational modifications are essential for understanding gene function [10–12] . Second , the univariate approach that forms the majority of our study assumes a gene’s contribution to electrophysiology is similar and monotonic across cell types . This single-gene focused analysis likely misses genes that contribute to complex ephys features in ways that are biologically degenerate and are highly non-linear or combinatorial [28 , 29] . For example , Kv3-family ion channels , including Kcnc1/Kv3 . 1 , have been implicated in helping fast-spiking cells maintain narrow spike widths [32 , 60] , but we did not identify Kcnc1 as correlated with AP width in our analysis . Further utilizing multivariate approaches ( like shown in Fig 3 ) and incorporating other information sources , such as how proteins interact to form functional complexes , might reveal additional signals and help mitigate spurious correlations . However , pursuing such approaches will likely necessitate larger datasets than are currently available . Third , the focus of our analysis is to explain how ephys differences across cell types emerge through gene expression . It remains an open question whether the same genes driving large across cell type differences would also be the same genes that are defining subtler within cell type differences , like amongst olfactory bulb mitral cells or CA1 pyramidal cells [1 , 2 , 58] . As the patch-seq methodology , enabling transcriptomic and ephys characterization from the same single-cell [19 , 20] , is further developed and applied , we eagerly anticipate testing these hypotheses . However , small changes in expression of individual genes , as expected within a single cell type , are difficult to reliably detect using current technologies , in part , due to relatively limited sample sizes and technical challenges like “dropouts” [18] . Indeed , while these patch-seq studies have demonstrated their utility in classifying individual cells into types [19 , 20] , how variance in expression of specific genes gives rise to within cell type ephys differences remains largely unaddressed . Fourth , ephys property correlations and gene co-expression limits the potential specificity of any causal prediction made here . For example , some pairs of ephys properties , like AHPamp and Rin , are correlated but probably do not share common biophysical underpinnings ( S3B Fig ) . Because of this common correlation , genes significantly associated with one ephys feature are more likely to be also associated with other ephys features , potentially spuriously . Similarly , many pairs of genes show correlated expression across samples ( i . e . , gene co-expression ) . Gene co-expression often reflects biologically meaningful signals , such as co-regulation by common transcription factors or shared membership in biological pathways and cellular compartments [61] . However , co-expression makes interpreting individual gene-ephys associations difficult and likely contributes to why we found many more genes for some ephys properties than we would naively expect , such as Vrest and AHPamp . Future analysis approaches that explicitly consider co-expression might prove useful [62] . Lastly , the heterogeneous nature of the compiled NeuroExpresso/NeuroElectro dataset [23 , 25 , 59] might limit our power to see possible biologically relevant signals and could explain our failure to find genes for some ephys features . For example , because data in NeuroElectro are compiled from different studies collected in the absence of standards for how some ephys properties are defined [24 , 63] , this likely limits our downstream attempts at normalization . Similarly , the cell types reflected in the aggregated dataset are likely composed of multiple transcriptomic or morphologically-defined subtypes [27 , 64] . However , the overall consistency with the AIBS Cell Types dataset , where data were collected using standardized conditions and protocols , suggests that the results shown here are not entirely the result of technical artefacts due to data compilation . Our findings suggest a number of directions for future study . Can specific gene-ephys relationships be used as biomarkers to detect electrophysiological changes in a disease or treatment context ? For example , if Scn1a/Nav1 . 1 is upregulated in a cell type , does that serve as a reliable indicator of hyper-excitability ? Given the relative ease and growing popularity of single-cell transcriptomics on dissociated cells and nuclei [18 , 27] , could the multivariate gene expression-based statistical models we developed be useful in imputing ephys phenotypes from transcriptomic signatures alone ? Lastly , are the gene-ephys correlations reported here predictive of cell-to-cell variability reported within the same cell type ? In summary , our results suggest that large-scale transcriptomics can prove useful in helping elucidate the biophysical basis for the rich electrophysiological diversity seen amongst neuron types throughout the brain .
To obtain neuron type-specific transcriptomic data , we made use of the NeuroExpresso database ( neuroexpresso . org ) , described previously [23] . Briefly , the database contains transcriptomic studies collected from mouse brain cell types sampled under normal conditions . We specifically utilized the microarray-specific subset of NeuroExpresso . These samples were collected using purified , pooled-cell microarrays with transcriptomes quantified using the Affymetrix Mouse Expression 430A Array ( GPL339 ) or Mouse Genome 430 2 . 0 Array ( GPL1261 ) . We further only used probesets that were shared between both platforms . Transcriptomic samples were quality controlled and manually curated for cell type identity and basic sample metadata , including animal age , array platform , and purification method . Transcriptomic samples are from adult mice unless explicitly mentioned . The samples were subjected to RMA normalization and an additional round of quantile normalization in order to obtain a uniform distribution of signals across samples . When a single gene was represented by multiple probesets , the probeset with highest variability across samples was chosen to represent the gene . We note that we have re-annotated the cell type labels used here from those used in the NeuroExpresso database and web resource . For the purpose of obtaining a large corpus of cell types , we made use of a small number of cell type-specific transcriptomic samples excluded from analysis in the original NeuroExpresso publication ( e . g . , developmentally immature samples ) . Specifically , for two major cell types with transcriptomic data collected at varying ages , cortical parvalbumin-positive ( PV ) interneurons labelled by the G42 mouse line and cerebellar Purkinje cells [22 , 65] , we kept samples collected at different ages separate and used of samples collected from animals aged less than P14 . We further included data representing cortical Htr3a- and Oxtr-expressing cells from Gene Expression Omnibus ( GEO ) accession GSE56996 [66] and layer 2–3 and layer 6 pyramidal cells from GSE69340 [67] . The complete listing of transcriptomic samples , annotated cell types , and references is provided in S2 Table . Following data compilation , we filtered genes to retain only those with 1 ) high mean expression; and 2 ) highly variable expression across cell types in the combined dataset . Specifically , for each gene , g , we calculated its expression mean , μg , and standard deviation , σg , across the collection of 34 cell types in the combined discovery dataset . Next , we calculated a global mean , μglobal defined as mean ( μg1:gN ) , and standard deviation , σglobal defined as mean ( σg1:gN ) across the total set of genes . Here , μglobal = 7 . 5 and σglobal = 0 . 75; for context , background expression levels were approximately ~6 . 0 ( log2 expression units ) . We filtered genes where μg > μglobal and σg > σglobal , leaving 2694 from 11667 total genes quantified . Lastly , we summarized each cell type by the mean expression per gene across samples . To obtain neuron type-specific electrophysiological measurements , we used an updated version of the NeuroElectro database ( neuroelectro . org ) , originally described in [24 , 25] . Briefly , we populate the NeuroElectro database using manual curation to extract information on electrophysiological measurements such as resting membrane potential and input resistance ( described in S1 Table ) from the results sections of published papers using intracellular electrophysiology . These ephys features were chosen because they were frequently reported across articles and were calculated using relatively consistent criteria from article to article . Curators also annotate a set of relevant methodological information , including species , animal age , electrode type , preparation type , recording temperature , and use of liquid junction potential correction . Because it was uncommon for a single study to characterize both a cell type’s transcriptomic and electrophysiological parameters , we developed a neuroinformatics-based strategy for pairing gene expression and ephys datasets from different studies based on common cell type identity . We first manually re-annotated the cell type identity of each transcriptomic sample from NeuroExpresso using a descriptive semantic label ( shown in S2 Table ) , defined by a minimally sufficient number of defining features ( including brain region and marker gene expression or projection pattern [69] ) . For example , the transcriptomic samples corresponding to cerebellar granule cells in NeuroExpresso were purified using the L10a-Neurod1 mouse line , where GFP is specifically expressed in the ribosomes of these cells [70] . Here , we merely annotated these samples using the label , “cerebellar granule cells” ( CB gran ) . We next identified all curated electrophysiological data within NeuroElectro corresponding to this same major cell type , making use of the manual annotations for each electrophysiological sample’s cell type identity ( n = 9 articles for CB granule cells ) . We note that subtle differences between how CB granule cells are labelled in the L10a-Neurod1 mouse line and how CB granule cells are targeted by lamina and morphology for ephys recordings would tend not to be preserved after this data harmonization step . Lastly , we note that these cell types reflect broad cellular classes and likely encompass multiple morpho-electric or transcriptomic subtypes [27 , 64] . To pair transcriptomic to ephys datasets explicitly defined by different ages ( e . g . , P7 and P25 ) , we matched animal ages +/- 2 . 5 days . For example , the samples corresponding to “Ctx G42 P15” reflect neocortical parvalbumin-positive interneurons labeled by GFP in the G42 mouse line aged P15 +/- 2 . 5 days . Because we tended to have fewer data points after subsetting the cortical G42 cells into different age groups , for one ephys property , APthr , we excluded APthr values from these cells since they varied widely ( ~10mV ) across studies from the same time point . The harmonized and processed cell type-specific data for the discovery and validation datasets has been made publically available at http://hdl . handle . net/11272/10485 . To obtain specific gene sets , we made use of Gene Ontology annotations ( as of August 2016 ) . We used the GO term 0005216 corresponding to “ion channel activity” to identify ion channels; the term 0015075 corresponding to “ion transmembrane transporter activity” in addition to Nkain1 to identify ion transporters; the term 0007010 corresponding to “cytoskeleton organization” to identify cytoskeletal genes; the term 0007399 corresponding to “nervous system development” to identify developmental genes; and the term 0034765 to identify “regulation of ion transport” in addition to the genes L1cam , Slmap , and Ank1 . To obtain a comprehensive manually curated listing of transcription factors , we used the Transcription Factor Checkpoint resource [72] .
|
Brain cell types have different electrical features , determined by the genes that each cell expresses . By combining data from hundreds of articles studying individual cell types in isolation , we developed a dataset that combines neuron gene expression patterns with their electrical characteristics . We asked if patterns of gene expression could predict a neuron’s electrical features; for example , if a neuron that expresses more of a sodium channel also tends to fire action potentials more frequently . We found hundreds of such statistical correlations that also replicated across brain cell types and regions . These relationships provide a starting point for understanding how alterations in the gene expression result in alterations in electrical functioning of neurons and brain circuits .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"membrane",
"potential",
"brain",
"electrophysiology",
"electrophysiological",
"properties",
"electrophysiology",
"neuroscience",
"ion",
"channels",
"genome",
"analysis",
"bioassays",
"and",
"physiological",
"analysis",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"proteins",
"gene",
"expression",
"electrophysiological",
"techniques",
"biophysics",
"physics",
"biochemistry",
"cellular",
"neuroscience",
"cell",
"biology",
"neurophysiology",
"physiology",
"neurons",
"genetics",
"transcriptome",
"analysis",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"genomics",
"computational",
"biology"
] |
2017
|
Transcriptomic correlates of neuron electrophysiological diversity
|
The antagonistic actions of Polycomb and Trithorax are responsible for proper cell fate determination in mammalian tissues . In the epidermis , a self-renewing epithelium , previous work has shown that release from Polycomb repression only partially explains differentiation gene activation . We now show that Trithorax is also a key regulator of epidermal differentiation , not only through activation of genes repressed by Polycomb in progenitor cells , but also through activation of genes independent of regulation by Polycomb . The differentiation associated transcription factor GRHL3/GET1 recruits the ubiquitously expressed Trithorax complex to a subset of differentiation genes .
Epigenetic control of cell fate by the opposing action of the repressive Polycomb group proteins ( PcG ) and the activating Trithorax group proteins ( trxG ) is a mechanism found throughout evolution [1]–[3] . These families of chromatin modifiers were first described as regulators of HOX gene expression in Drosophila [4] , [5] . The mammalian counterparts of PcG and trxG have since been identified and their role in the regulation of multiple cellular processes , outside of HOX gene regulation , has begun to emerge [1] . Although trxG has been depicted as a de-repressor of PcG repressed genes in Drosophila , it remains unclear if in mammalian differentiation trxG mediated gene regulation is only through antagonizing PcG mediated gene repression , or if trxG can regulate gene activation independent of PcG [1] . The epidermis , the outermost skin layer , is a multi-layered epithelium containing proliferative progenitors at the base that migrate towards the surface while simultaneously undergoing differentiation to form an effective barrier which prevents dehydration and protects the organism against toxins and invasion of microorganisms [6] , [7] . Epidermal differentiation involves the coordinated expression of numerous genes including those involved in protein cross-linking , lipid metabolism , and cell adhesion . As such this system represents an excellent model for dissecting the transcriptional and regulatory changes required for differentiation . Some key transcription factors regulating this process have been identified [8] , including GRHL3/GET1 [9] , [10] , and more recently the contribution of epigenetic regulation has begun to emerge . DNA methylation [11] and histone deacetylation [12] , as well as remodeling of chromatin by BRG1 [13] and Mi-2beta [14] , have been described in regulating various stages of epidermal differentiation and homeostasis . Furthermore , the chromatin organizer Satb1 , shown to be directly regulated by the transcription factor p63 , regulates the expression of certain epidermal differentiation associated genes [15] . Epidermis-specific deletion of Ezh2 , a PcG histone methyltransferase of the polycomb repressive complex 2 ( PRC2 ) that catalyzes the methylation of Lys27 on histone 3 ( H3K27 ) , led to the premature expression of some but not all late differentiation genes [16] . Consistently , it was recently reported that epidermal deletion of the PRC2 related protein JARID2 leads to a loss of H3K27me3 at many of the Ezh2 affected epidermal differentiation genes [17] . Moreover , the PRC1 related protein Cbx4 was shown to maintain human epidermal stem cells in the proliferative state while also preventing them from senescence [18] . Complementing these findings , JMJD3 , a histone demethylase that removes the repressive H3K27me3 , promotes epidermal differentiation [19] . Thus , there is abundant evidence indicating that PcG-mediated H3K27 methylation maintains keratinocytes in the progenitor state and release from this repression contributes to epidermal differentiation . Yet , addition and removal of the H3K27me3 mark does not fully explain differentiation associated gene activation as many differentiation genes are not affected by interference with either Ezh2 or JMJD3 [16] , [19] . We now show that the trxG components , histone H3K4 methyltransferase MLL2 and WDR5 , play an important role in epidermal differentiation and , in part , act to regulate gene expression independent of PcG . Furthermore we demonstrate that a subset of epidermal differentiation genes are activated by GRHL3 mediated recruitment of trxG .
The highly conserved Grainyhead transcription factors control epidermal differentiation and barrier formation in organisms ranging from worm to human by directly or indirectly regulating the expression of key genes involved in these processes [9] , [10] , [20]–[22] . One Grainyhead homologue , the mouse Grhl3/Get1 , is a critical regulator of the epidermal differentiation program [9] , [22] . To study GRHL3 gene-regulatory mechanisms we utilized the in vitro calcium induced differentiation model of normal neonatal human epidermal keratinocytes ( NHEK ) . In this system GRHL3 is expressed at low levels in undifferentiated cells and at higher levels when cells are induced to differentiate ( Figure S1A ) , reminiscent of its high expression in the most differentiated layer of mouse skin . Likewise the GRHL3 target Transglutaminase 1 ( TGM1 ) , a Ca2+-dependent enzyme that functions in the formation of the cornified cell envelope by crosslinking proteins such as Involucrin and Filaggrin , is increased 3-fold upon calcium-induced differentiation ( Figure 1A ) . The human TGM1 promoter contains a conserved GRHL3 binding site ∼800 bp upstream of the transcription start site ( TSS ) ( Figure 1B ) within a 2 . 5 Kb region that mediates correct temporal and spatial expression in transgenic mice [23] . To determine if TGM1 is a direct target of GRHL3 in NHEKs , we performed chromatin immunoprecipitation ( ChIP ) assays with a GRHL3 antibody in undifferentiated and differentiated NHEKs with three primer pairs tiling the TGM1 promoter ( Figure 1B ) . Consistent with a differentiation-dependent increase in TGM1 expression , we observed a differentiation-dependent increase in GRHL3 occupancy at the predicted binding site in the TGM1 promoter ( Figure 1C ) ; this binding was specific as no binding was detected upstream or downstream ( Figure 1C ) . Higher occupancy of GRHL3 on the TGM1 promoter in differentiated NHEKs correlated with increased H3K4me3 , a histone modification associated with active promoters ( Figure 1D , 1E ) . Additionally we found low levels of H3K27me3 in differentiated NHEKs compared to slightly higher levels in undifferentiated NHEKs and RT4 cells , a human bladder epithelial cell line which expresses GRHL3 but virtually no TGM1 ( Figure 1A and 1D–1F , Figure S1A ) . RT4 cells also displayed very low levels of the H3K4me1 and H3K4me3 modifications consistent with the low level of TGM1 expression ( Figure 1F ) . These experiments illustrate a distinct chromatin landscape in TGM1-expressing and non-expressing cell types and increased H3K4 methylation at the TGM1 promoter correlating with increased GRHL3 binding and TGM1 expression during differentiation . To determine whether increased H3K4 methylation at the TGM1 promoter depends on GRHL3 binding , we utilized siRNAs to knock down GRHL3 in NHEK cells followed by calcium-induced differentiation for 48 hours . As expected there is a greater than 2-fold reduction in TGM1 mRNA levels upon knockdown of GRHL3 ( Figure S1B ) . Correspondingly , we observed a strong decrease in H3K4 mono- , di- , and tri-methylation at the TGM1 promoter in GRHL3 depleted cells ( Figure 2A ) . These results indicate that increased H3K4 methylation at the TGM1 promoter during differentiation of NHEK cells is facilitated by GRHL3 . As H3K4 methylation is mediated by the SET domain-containing trxG components , we examined the expression of these and core trxG complex members in NHEKs; all are expressed at a relatively stable level during human keratinocyte differentiation ( Figure S2A ) . We also assessed expression of GRHL3 , MLL1 , MLL2 , and WDR5 transcripts by qPCR in human skin ( Figure S2B ) and mouse skin separated into dermal and epidermal compartments ( Figure S2C ) . Transcripts encoding all four factors are easily detected in human skin with WDR5 and MLL1 having the highest expression ( Figure S2B ) . In mouse skin that was separated into epidermis and dermis , we found that Grhl3 and Mll2 were enriched in the epidermis while both Mll1 and Wdr5 are expressed at a similar level in the epidermis and dermis ( Figure S2C ) . We also assessed the expression of these same proteins in human skin by immunofluorescence and observed expression of MLL1 , MLL2 , and WDR5 throughout the epidermis with MLL2 and WDR5 showing higher expression in the more differentiated cells and MLL1 expression being absent from the nucleus of basal cells ( Figure S2D–S2H ) . Upon siRNA knockdown of individual trxG members followed by differentiation , there were varying effects on the expression of TGM1 . Knockdown of MLL1 and MLL2 ( also known as KMT2B and ALR ) caused a greater than 2 fold reduction in TGM1 expression , knockdown of MLL4 caused less of a reduction , and knockdown of ASHL1 caused a slight increase in TGM1 expression ( Figure 2B , Figure S2I ) . TGM1 mRNA expression was also decreased upon knockdown of WDR5 , a non-enzymatic core component of the methyltransferase complex , further supporting the role of trxG dependent histone methylation in TGM1 expression ( Figure 2B ) . ChIP assays revealed that MLL2 but not MLL1 occupied the TGM1 promoter in differentiated NHEK cells ( Figure 2C ) . MLL2 recruitment to TGM1 depends on GRHL3 as there is a significant reduction in both GRHL3 and MLL2 at the TGM1 promoter in GRHL3 depleted cells ( Figure 2D ) . Conversely , when MLL2 was knocked down , GRHL3 recruitment to the TGM1 promoter was not significantly affected ( Figure 2E ) . Together these findings indicate that TGM1 is a direct target of MLL2-mediated H3K4 methylation in NHEKs , and that this recruitment is GRHL3-dependent . To better understand the broader roles of GRHL3 and MLL2 in keratinocyte differentiation we utilized microarrays to define the global differentiation gene expression program in NHEKs at various time points during differentiation; 2 , 583 genes are significantly differentially expressed ( p<0 . 005 , fold-change>1 . 25 ) ( Figure S3A , Table S1 ) . The most overrepresented Gene Ontology ( GO ) terms included , “epidermis development” , “regulation of cellular proliferation” , “regulation of cell motion” , “cornified envelope” , and “positive regulation of gene expression” ( Figure S3B ) . K-Means clustering of these genes revealed four clear patterns ( Figure 3A–3D ) . First , a “progenitor” cluster; genes most highly expressed in undifferentiated cells with falling expression during the time course ( Figure 3A ) , including E2F3 and CDC6 , both of which have been linked to differentiation induced cell cycle arrest in keratinocytes [24] , [25] . Second , an “early” cluster; genes expressed at low levels in undifferentiated cells that were up-regulated most highly at one hour of differentiation ( Figure 3B , Figure S4 ) , including many key transcription factors like Jun and Fos , AP1 components that play an important role in epidermal differentiation [26] . Furthermore , this cluster contains KLF4 and HES1 , both of which have been shown to play a role in the induction of terminal differentiation [27] , [28] . Third , an “intermediate” cluster; genes up-regulated most highly at three to six hours post induction with many genes related to kinase activity , including SRF [29] , and positive apoptosis regulators including SOCS3 [30] ( Figure 3C , Figure S4 ) . Fourth , a “late” cluster; genes most highly upregulated at 24 and 48 hours , at the end of the time course with an overrepresentation of genes related to epithelial differentiation , keratinization , desmosomes , and the cornified layer including KRT1 , KRT10 , and FLG , as well as members of the Epidermal Differentiation Complex ( Figure 3D , Figure S3 ) . This cluster also contains CDKN2A ( Ink4a ) a powerful inhibitor of cell cycle progression shown to play a role in inhibiting G1 to S transition in the epidermis , which is critical for epidermal terminal differentiation [31] . In summary , through gene expression profiling over a dense time course of NHEK differentiation , we are able to recapitulate key aspects of normal epidermis differentiation and classify novel genes in this process . To further investigate the hypothesis that GRHL3 recruits MLL2 to target promoters to activate gene transcription during epidermal differentiation we studied the effect of loss of GRHL3 and MLL2 on global gene expression in differentiated NHEK cells . We found 323 and 4 , 281 differentially expressed genes ( p<0 . 001 ) when we depleted GRHL3 and MLL2 , respectively , indicating a more general role for MLL2 than GRHL3 in keratinocyte transcription ( Figure S5A–S5B and Tables S2 , S3 ) . DAVID analysis of down regulated genes revealed over-represented GO terms such as “cornified envelope” and “epidermal differentiation” in the GRHL3 siRNA dataset , indicating that GRHL3 plays a similar differentiation-promoting role in human keratinocytes as in mice ( Figure 3E ) . Down regulated genes in the MLL2 siRNA experiment were also enriched for “cornified envelope” and “epidermis development” supporting the idea that MLL2 plays a crucial role in epidermal keratinocyte differentiation ( Figure 3F ) . Knockdown of MLL2 in undifferentiated NHEKs had no effect on proliferation , apoptosis or senescence while knockdown of the core component WDR5 lead to a slight but significant decrease in proliferation and no change in apoptosis or senescence ( Figure 3G–3I ) . Comparing the lists of significantly down-regulated genes in each data set , we found a statistically significant overlap ( Figure 3J ) , and using MotifMap [32] , we detected a statistically significant over-representation of predicted GRHL3 binding sites in the down-regulated genes ( Figure S5C ) , providing additional evidence that GRHL3 and MLL2 co-occupancy regulates some of these genes . The GRHL3 regulated geneset overlapped significantly with both the early and intermediate clusters ( p<0 . 001 ) from our time course study of NHEK differentiation . The intersection with the late cluster was even more significant ( p<1×10−17 ) , fitting with the idea that , as in mouse development , GRHL3 is an important regulator of terminal differentiation in human keratinocytes ( Figure S5D ) . When the genes affected by MLL2 siRNA were overlapped with the same four clusters , the most significant enrichment was found with both the intermediate and late clusters although there was a significant overlap with all four clusters , suggesting a broader role for MLL2 in the differentiation process ( Figure S5E ) . Further exploration of the differentially expressed genes in common between the GRHL3 and MLL2 siRNA experiments showed the strongest overlap with the late cluster of genes , indicating that these two factors converge on terminal differentiation genes ( Figure 3K ) . These findings support our hypothesis that GRHL3 acts , in part , to regulate gene expression through collaboration with MLL2 and that MLL2 itself is a crucial regulator of epidermal differentiation . To test whether GRHL3 and MLL2 co-occupy the promoters of epidermal differentiation genes other than TGM1 , we identified a set of genes with the following criteria: 1 ) down regulated by siRNAs against both factors; 2 ) GO terms related to epidermal differentiation; and 3 ) containing a high scoring GRHL3 binding site within a −2 to +1 Kb region around the transcription start site . These putative GRHL3/MLL2 target genes were then tested for the presence of GRHL3 , MLL2 and H3K4me3 by ChIP assays on chromatin from either scrambled or GRHL3 siRNA treated differentiated NHEKs . Consistent with our hypothesis many of the tested target genes were in fact bound by GRHL3 and MLL2 at their proximal promoters and displayed reduced H3K4me3 levels in GRHL3 siRNA treated cells , with the exception of SPRR2B which had no change in H3K4me3 ( Figure 4A–4C ) . We also performed ChIP assays for MLL1 and SET1 occupancy and found that SPRR2B is also a target of both MLL1 and SET1 while EPHX3 is also a target of MLL1 ( Figure 4D–4E ) . We also examined GRHL3 occupancy at these genes in cells depleted of MLL2 and found that with the exception of BLNK , GRHL3 occupancy was not significantly decreased , further supporting the hypothesis that GRHL3 contributes to MLL2 recruitment to target gene promoters and not vice versa ( Figure 4F ) . In order to test whether GRHL3 could directly bind trxG complex members , we performed Co-IP experiments on extracts from differentiated NHEK cells , and from 293T cells transfected with HA-GRHL3 . While we could not detect GRHL3 in MLL2 immunoprecipitates in NHEK cells , we readily detected GRHL3 in extracts precipitated with WDR5 , a core component of the trxG methyltransferase complex ( Figure 5A ) . HA-GRHL3 was detected in both MLL2 and WDR5 immunoprecipitates in 293T cells , while WDR5 , but not MLL1 or SET1 , was readily detected in 293T extracts precipitated with HA antibody ( Figure 5A , Figure S6A–S6B ) . Thus , GRHL3 appears to interact strongly with WDR5 and to a lesser degree with MLL2 , consistent with the idea that GRHL3 can recruit trxG to target promoters . The above results led us to perform ChIP-seq experiments to define genome-wide GRHL3 and WDR5 co-occupancy in differentiated NHEK cells , identifying 25 , 340 GRHL3 peaks and 48 , 269 WDR5 peaks ( FDR<5% ) ( Figure S6C ) . For both proteins there is a statistically significant enrichment in occupancy at promoters compared to the average genomic distribution ( Figure S6D ) . Strikingly , we found that 88 percent of genes that contained a GRHL3 peak also had an overlapping WDR5 peak , further supporting the hypothesis that GRHL3 recruits WDR5 to gene regulatory regions ( Figure 5B ) . GO analysis of these co-occupied targets revealed enrichment for processes like “cell differentiation” , “positive regulation of gene expression” , “regulation of programmed cell death” , “cell-cell adhesion” , and “regulation of lipid biosynthetic processes” , all important components of epidermal keratinocyte differentiation ( Figure 5C ) . Forty three percent of genes differentially expressed during keratinocyte differentiation had co-occupancy of WDR5 and GRHL3 ( Figure S6E ) , and there was a relatively even distribution of WDR5 , GRHL3 , and WDR5/GRHL3 co-occupied genes in our four differentiation clusters ( Figure S6F ) . There is also a statistically significant overlap of genes bound by GRHL3 that were downregulated upon GRHL3 depletion by siRNA , as well as between genes bound by WDR5 and downregulated upon MLL2 depletion , supporting the idea that many of these genes are direct targets of GRHL3 and MLL2 , respectively ( Figure 5D–5F ) . Together these findings indicate a role for trxG in human epidermal keratinocyte differentiation and a novel role for GRHL3 in recruiting this complex to gene promoters . To understand the relationship between trxG and PcG in differentiation and to determine if there are genes that are regulated by trxG independent of PcG , we performed ChIP assays at one and three hours after calcium-induction to examine the dynamics of H3K4me3 and H3K27me3 at the differentiation associated genes we had identified as common targets of GRHL3 and MLL2 . Interestingly , different sets of genes showed unique dynamics of these marks , one group , including the genes CRCT1 , SBSN , and EPHX3 showed initially high levels of H3K27me3 followed by a drastic decrease , coupled with only a modest increase in H3K4me3 ( Figure 6A , Figure S7A ) . A predominant mechanism of PRC recruitment , described in human ES cells , is through interactions with unmethylated CpG islands [33]–[35] . It is therefore intriguing that SBSN , whose promoter contains no CpG islands , had high levels of H3K27me3 in undifferentiated keratinocytes; there is a broad region of H3K27me3 with a peak ∼800 bp upstream from the TSS ( Figure S7B ) . Taking advantage of the publically available ENCODE histone modification ChIP-seq data we studied globally the overlap of H3K27me3 with CpG islands and their 500 bp flanking regions . While in ES cells 43 percent of H3K27me3 regions did overlap with CpG islands , only 17 percent of H3K27me3 regions in undifferentiated NHEKs overlapped with CpG islands . We also carried out this analysis on H3K27me3 regions in Normal Human Lung Fibroblasts and found they also had a 17 percent overlap with CpG islands . This finding supports the hypothesis that in more differentiated cells types such as epidermal keratinocytes and lung fibroblasts , PRC recruitment and H3K27me3 are largely mediated by mechanisms unrelated to unmethylated CpG islands . A second group , BLNK and IVL , displayed much less change in H3K27me3 , with a more prominent increase in H3K4me3 ( Figure 6B , Figure S7C ) . These two groups are likely regulated by a combinatorial PcG/trxG mechanism . In contrast a third group of differentiation genes , including TGM1 and SCEL show nearly no H3K27me3 in progenitor or differentiated cells , but display a dramatic increase in H3K4me3 upon differentiation ( Figure 6C , Figure S7D ) . These findings support a model where activation of differentiation genes in epidermal progenitors is either mediated through a joint , reciprocal PcG/trxG coregulation , or by trxG alone ( Figure 6D ) .
Our study presents three major findings significant for understanding the control of epithelial cell differentiation . First we defined a role for Trithorax proteins WDR5 and MLL2 in activating the differentiation gene program in epidermal progenitor cells . Through siRNA knock down experiments , whole genome gene expression microarrays and ChIP experiments , we demonstrate that the histone methyltransferase , MLL2 , activates many genes involved in different stages of epidermal progenitor cell differentiation . We also discovered , through ChIP-sequencing experiments , that a core component of the histone methyltransferase complex , WDR5 , occupies promoters of a large subset of genes involved in epidermal progenitor differentiation . While previous work has demonstrated that WDR5 is involved in maintaining the ES cell state [36] , our work shows that WDR5 is also important for promotion of terminal differentiation in the epidermal lineage . The most probable explanation for the diversity in WDR5 function is selective recruitment to the appropriate promoters by cell- and differentiation-specific DNA binding proteins . Thus , in ES cells trxG complexes are recruited to gene targets through direct interactions with Oct4 , while in epidermal keratinocytes recruitment is through interactions with epidermal transcription factors such as GRHL3 . The SET domain containing proteins in the trxG complex are the subunits that are enzymatically responsible for methyltransferase activity , therefore described as “writers” . These proteins include the MLL and SET proteins , the mammalian homologues to drosophila trx . The apparent non-redundant functions of the mammalian MLL proteins , as demonstrated by the embryonic lethality in mice upon deletion or truncation of individual family members [37]–[40] , suggests that different MLL proteins may act as unique components in trxG complexes aiding in their gene specific recruitment . Identifying the specific SET domain containing MLL protein , MLL2 , as having a role in promoting epidermal differentiation broadens our understanding of the role of MLL proteins in differentiation of a self-renewing tissue . We also observed MLL1 binding to some epidermal differentiation genes , suggesting that other MLLs and SET may play a role in activating the epidermal differentiation program . Recently mutations in MLL2 have been found with high incidence rate in patients with Kabuki syndrome , a syndrome with multiple congenital abnormalities and intellectual disabilities [41]–[47] . No skin abnormalities have been described in Kabuki syndrome which may be because the syndrome is caused by haploinsufficiency [41] , [47] and the predicted fifty percent reduction in MLL2 expression may not be sufficient to affect epidermal differentiation in vivo . Secondly we define a previously unknown role for the transcription factor GRHL3 in the recruitment of a trxG complex to promoters of genes , leading to increased H3K4 methylation and gene expression . The critical role for Grhl3 in epidermal differentiation was previously shown in the mouse [9] , [10] , and we now demonstrate a similar role in activation of the human epidermal differentiation gene expression program . Our genome wide gene expression analysis on cells depleted of either GRHL3 or MLL2 revealed 109 genes co-regulated by these two factors , most of which have an increase in expression late in epidermal progenitor cell differentiation . Furthermore , we determined that GRHL3 directly interacts with WDR5 in differentiated keratinocytes and through ChIP-sequencing experiments we established a global co-localization of these two factors , with a significant enrichment of occupancy at genes involved in processes crucial for proper epidermal progenitor differentiation . Thus , we hypothesize that GRHL3 recruits trxG to epidermal differentiation promoters through WDR5 . While the expression of MLL2 and other trxG components do not change during epidermal differentiation , the expression of GRHL3 does increase , possibly explaining how trxG is directed to its differentiation associated gene targets in a differentiation dependent manner . As GRHL3 plays roles in the differentiation of other epithelia , including bladder epithelium [48] , we speculate that trxG may be similarly involved in activation of other epithelial differentiation programs . Lastly we propose models of trxG mediated regulation of differentiation that are dependent and independent of functional interactions with PcG . While previous studies demonstrated roles for the PcG complex in maintaining the progenitor state [16] , and demethylation of H3K27 for de-repression of epidermal differentiation genes [19] , the current work suggests that this mechanism alone cannot fully explain activation of the differentiation program . This finding is consistent with the aforementioned studies as they found that only a subset of differentiation associated genes are affected upon disruption of either Ezh2 or JMJD3 [16] , [19] . Activation of differentiation genes that are suppressed by PcG in progenitor cells appears to be associated with recruitment of trxG and increased H3K4me3 at the gene promoters during the differentiation process . This mode of regulation for differentiation genes is reminiscent of the antagonistic actions of trxG and PcG in fate determination in Drosophila [49] . However , the promoters of a subset of differentiation genes , including TGM1 , whose activation is associated with H3K4me3 are not marked by H3K27me3 in the progenitor state; these genes primarily rely on a trxG mediated mechanism for activation and their low expression in progenitor cells does not appear to be associated with PcG mediated H3K27me3 . These findings are consistent with in vitro studies where trxG could activate transcription on a chromatin template without the presence of PcG proteins [50] and where during induced gene activation , MLL2 can associate transiently with the Myc locus [51] . However , it should be pointed out that the set of epidermal differentiation genes that we find to be independent of regulation by PcG in the NHEK differentiation model may have been regulated by PcG earlier in their lineage similar to that observed for GATA3 gene activation during T cell development [52] . In summary , our work supports a previously unappreciated role for trxG in promoting expression of the human epidermal progenitor differentiation program; reveals the role of a transcription factor GRHL3 in recruiting this complex to its gene targets; and uncovers a function for trxG mediated gene activation in differentiation that is independent of overcoming PcG mediated repression .
Neonatal Human Epidermal Keratinocytes ( NHEK ) were purchased from LifeLine Technologies and grown according to the manufacturer's instructions in Dermalife medium ( LifeLine Tech ) supplemented with Dermalife growth factors ( LifeLine Tech ) . For ChIP-qPCR experiments , cells were grown for two days and for ChIP-seq experiments cells were grown for 24 hours in medium supplemented with a final concentration of 1 . 8 mM CaCl2 to induce differentiation . For timecourse experiment , cells were seeded into a 6 well plate and induced to differentiate at 50% confluency by addition of 1 . 8 mM CaCl2 and collected at 0 , 1 , 3 , 6 , 12 , 24 , and 48 hours after induction . RT4 cell were maintained in McCoy's 5A medium ( Gibco ) supplemented with 10% FBS . 293T cells were grown in DMEM ( Gibco ) supplemented with 10% FBS . For siRNA experiments , NHEK cells were subjected to reverse transfection using Lipofectamine RNAi max ( Life Tech Inc . ) per manufacturer's protocol . 15 hours post transfection , medium was changed to Dermalife supplemented with 1 . 8 mM CaCl2 and cells were allowed to grow for 48 hours . The following Silencer Select siRNAs ( Ambion ) were used at a final concentration of 25 nM: GRHL3 ( cat#s33754 ) MLL1 ( cat#s8819 ) MLL2 ( cat#s15604 ) MLL4 ( cat #s18831 ) AshL1 ( cat #s31702 ) WDR5 ( cat #s225470 ) Negative #1 ( cat#4390843 ) . For HA-GRHL3 overexpression , plasmid was transfected into 293T cells with Lipofectamine 2000 ( Life Tech Inc . ) per manufacture's protocol . Proliferation assays were performed by quantifying total ATP content via the ApoSENSOR ATP Luminescence luciferase reporter assay ( cat# K254-1000 , Bio Vision Inc . ) . TUNEL assays were performed using In Situ Cell Death Detection Kit , Fluroscein ( cat# 11 684 795 910 , Roche ) . Senescence assays were performed using Senescence β-Galactosidase Staining Kit ( cat# 9860 , Cell Signaling Technology , Inc . ) . Immunofluorescence was performed as previously described [53] using 4% PFA fixed tissue . The following antibodies were used: MLL1 ( Thermo Scientific , cat#PA5-11264 , 1∶100 ) , MLL2 ( Abcam , cat#AB32474 , 1∶100 ) , WDR5 ( R&D Systems , cat#AF5810 , 1∶100 ) , K5 ( Covance , 1∶1000 ) , K10 ( Covance , 1∶1000 ) , p63 ( Santa Cruz , sc-8431 , 1∶200 ) AlexaFluor anti-Rabbit 488 ( Invitrogen , cat#A11008 , 1∶500 ) , AlexaFluor anti-Goat ( Invitrogen , cat#A11078 , 1∶500 ) and AlexaFluor anti-mouse 594 ( Invitrogen , cat#A11005 , 1∶500 ) . Cells were collected and lysed in Trizol , followed by Chloroform extraction . RNA was extracted from the aqueous phase using Ambion PureLink RNA mini kit per manufacturer's protocol . RNA concentration and quality were quantified on a NanoDrop . All experiments were performed with biological duplicates . Experiments were performed as previously described [54] except Affymetrix Human Gene 1 . 0 ST arrays ( 26 , 869 probe sets ) were used and washed according to manufacturer's recommendations ( Affymetrix , Santa Clara , CA ) . For the time course ANOVA was performed , using MeV software [55] , [56] to analyze genes for differential expression . K-means clustering was performed on the differentially expressed genes as determined by ANOVA with a p<0 . 005 and greater than +/−1 . 5 fold change . siRNA microarrays were analyzed with Cyber-T [57] . The Neg siRNA was used as the control and either Grhl3 siRNA-treated or Mll2 siRNA-treated samples as experimental . Gene Ontology analysis was performed on all datasets using DAVID [58] , [59] . Statistical significance of overlaps was calculated using the Fisher's exact test . The microarray data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus [60] and are accessible through GEO series accession numbers GSE37570 , GSE37049 , and GSE38628 . ChIP assays were performed as previously described [48] with the following changes: 24 ug of sonicated chromatin was used for each IP and enrichment was calculated as a percent of input sample compared to an IgG control IP and normalized to a control genomic region ( n≥3 ) . The following antibodies were used : Grhl3 ( Andersen Lab ) MLL2 ( AbCam cat# ab32474 ) MLL1 ( Bethyl cat#A300-374A ) SETD1A ( Abcam cat# ab70378 ) IgG ( Sigma cat#15006-10MG ) H3K4me1 ( AbCam cat# ab8895 ) H3K4me2 ( AbCam cat# ab32356 ) H3K4me3 ( Milipore cat#07-473 ) H3K27me3 ( AbCam cat#ab6002 ) WDR5 ( AbCam cat#ab56919 ) Primer sequences available upon request . For mRNA expression analysis cDNA was prepared using iScript cDNA kit ( Biorad Laboritories ) and RT-PCR was performed using SsoFast for Probes and SsoFast EvaGreen ( Biorad Laboratories ) master mixes in CFX384 Real-Time PCR Detection System ( Biorad Laboratories ) . GAPDH or RPLPO were used as endogenous controls . RT-PCR was performed using the following primers or probes ( n≥3 ) : Taqman Probes: WDR5: Hs00424605_m1 TGM1: Hs00165929_m1 Krt10: Hs00166289_m1 Grhl3: Hs00297962_m1 Grhl1: Hs00227745_m1 Ppl: Hs00160312_m1 Primers: Brip1 Fwd: TTACCCGTCACAGCTTGCTAT Rv: TCCCACTAAGAGATTGTTGCCA Ets1 Fwd: AGACGGAAAAAGTCGATCTGGA Rv: TGCTTGGAGTTAATAGTGGGACA Fos Fwd: CGGGCTTCAACGCAGACTA Rv: GGTCCGTGCAGAAGTCCTG Jun Fwd: TGGAAACGACCTTCTATGACGA Rv: GTTGCTGGACTGGATTATCAGG Klf4 Fwd: GCGCTGCTCCCATCTTTCT Rv: TGCTTGACGCAGTGTCTTCTC Creb5 Fwd: CCCTGCCCAACCCTACAATG Rv: GGACCTTGCATCCCCATGAT NHEK cells were differentiated for two days prior to cell collection . 293T cells were collected 2 days post transfection with HA-GRHL3 . Cells were lysed in 1%NP-40 lysis buffer on ice for 1 hour with vortexing every 5 minutes . Protein extract was pre-blocked with protein A- agarose beads ( Invitrogen ) for 45 minutes at 4C with constant rotation . Protein extract was incubated with 5 ug of indicated primary antibodies overnight at 4C: IgG ( Sigma ) , WDR5 ( Abcam ) , MLL1 ( Bethyl ) , MLL2 ( Abcam ) , SETD1A ( Abcam ) , HA ( Covance ) , and Grhl3 ( Andersen Lab ) . Samples were immunoprecipitated using pre-blocked protein A Dynabeads ( Invitrogen ) for 1 hour at 4C , followed by washing in PBS and elution in loading buffer at 100C for 10 minutes . Protein samples were run on a 4–20% gradient gel ( Invitrogen ) and transferred to a PVDF membrane . The membrane was blocked in 5% milk , washed with 1xPBS-T and incubated in 1% milk with indicated antibodies: Grhl3 ( Andersen Lab ) , MLL2 ( AbCam ) , MLL1 ( Bethyl ) , SETD1A ( AbCam ) , WDR5 ( AbCam ) , HA ( Covance ) followed by incubation in secondary antibodies anti-rabbit HRP or anti-mouse HRP . Signal was detected using ECL per manufacturer's protocol ( Denville ) . Sequencing libraries were generated for the GRHL3 , WDR5 , and Input samples using the NEB Next reagents and Illumina adaptors and oligos , according to the Illumina protocol for ChIP-Seq library preparation , with some modification . After adaptor ligation , PCR amplification was performed prior to size selection of the library [61] . Clustering and 50-cycle single end sequencing were performed on the Illumina Hi-Seq 2000 Genome Analyzer . Resulting reads were aligned using Bowtie [62] , and only uniquely aligning reads were retained . Peaks were called using MACS [63] , and Galaxy [64]–[66] was used for further analysis .
|
Human epidermal keratinocyte differentiation provides a highly suitable system to understand how progenitor cells become specialized . Previous work has implicated resolution of repressive histone modifications in the activation of the terminal differentiation gene expression program . Our work shows that this mechanism only accounts for the regulation of a subset of the differentiation gene expression program and that activating histone modifications by Trithorax chromatin modifiers , acting alone or in combination with the release from repressive chromatin changes , is essential . Furthermore , we show that the Trithorax complex is recruited to a subset of differentiation gene promoters by the transcription factor Grhl3 , an evolutionarily conserved regulator of the epidermal differentiation program . Altered differentiation is characteristic for several skin diseases , including skin cancer and inflammatory diseases such as psoriasis . While genetic abnormalities play a role in these diseases , the cellular and macro-environment may also alter the course of these diseases through chromatin changes ( epigenetics ) . Understanding the epigenetic regulation of keratinocyte differentiation may in the future lead to the development of new drugs for skin diseases .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"microarrays",
"developmental",
"biology",
"gene",
"expression",
"genetics",
"epigenetics",
"biology",
"computational",
"biology",
"chromatin",
"genetics",
"and",
"genomics",
"cell",
"differentiation",
"dna",
"transcription",
"histone",
"modification"
] |
2012
|
GRHL3/GET1 and Trithorax Group Members Collaborate to Activate the Epidermal Progenitor Differentiation Program
|
Treatment for human African trypanosomiasis is dependent on the species of trypanosome causing the disease and the stage of the disease ( stage 1 defined by parasites being present in blood and lymphatics whilst for stage 2 , parasites are found beyond the blood-brain barrier in the cerebrospinal fluid ( CSF ) ) . Currently , staging relies upon detecting the very low number of parasites or elevated white blood cell numbers in CSF . Improved staging is desirable , as is the elimination of the need for lumbar puncture . Here we use metabolomics to probe samples of CSF , plasma and urine from 40 Angolan patients infected with Trypanosoma brucei gambiense , at different disease stages . Urine samples provided no robust markers indicative of infection or stage of infection due to inherent variability in urine concentrations . Biomarkers in CSF were able to distinguish patients at stage 1 or advanced stage 2 with absolute specificity . Eleven metabolites clearly distinguished the stage in most patients and two of these ( neopterin and 5-hydroxytryptophan ) showed 100% specificity and sensitivity between our stage 1 and advanced stage 2 samples . Neopterin is an inflammatory biomarker previously shown in CSF of stage 2 but not stage 1 patients . 5-hydroxytryptophan is an important metabolite in the serotonin synthetic pathway , the key pathway in determining somnolence , thus offering a possible link to the eponymous symptoms of “sleeping sickness” . Plasma also yielded several biomarkers clearly indicative of the presence ( 87% sensitivity and 95% specificity ) and stage of disease ( 92% sensitivity and 81% specificity ) . A logistic regression model including these metabolites showed clear separation of patients being either at stage 1 or advanced stage 2 or indeed diseased ( both stages ) versus control .
In many fields of medicine , investigators are seeking ways to determine which patients will respond to particular drugs in order to guide therapy . For example , some cancers will respond to particular therapies if they carry faulty alleles of oncogenes whose products are the targets of those drugs . Examples where determination of an individual’s likelihood of response based on precise diagnosis of their disease are currently rare . One condition in which patient stratification has been possible based on available diagnosis is human African trypanosomiasis ( HAT ) . HAT , also known as sleeping sickness , is a parasitic disease of sub-Saharan Africa affecting isolated , rural communities . Two sub-species of the parasite infect humans . Trypanosoma brucei gambiense persists in West and Central Africa and is responsible for 90% of cases , while T . b . rhodesiense exists in East and Southern Africa . Uganda is the only country where both sub-species of the parasite exist . Upon infection , through the bite of an infected tsetse fly , parasites multiply in the blood and lymphatic systems of the patient ( stage 1 ) , before invading the central nervous system in stage 2 of the disease . Stage 2 disease leads to progressive neurological dysfunction: anxiety , depression , psychotic episodes , disrupted sleep-wake profile , coma and ultimately death if untreated [1][2] . Vaccines to prevent the disease are unlikely to be developed due to a complex process of antigenic variation [3] . Appropriate medication is therefore crucial to control HAT . Patients in stage 1 disease are treated with pentamidine or suramin , but as these drugs do not cross the blood-brain barrier , the highly toxic melarsoprol is required for stage 2 rhodesiense disease [4] . Nifurtimox-eflornithine combination therapy is commonly used to treat gambiense disease [4] . Two new drugs , fexinidazole and SCYX-7158 , that may treat both stages of the disease , are in development [5 , 6] , but these compounds are years away from the clinic and may still fail in clinical trials . Even when they are in use , a staging test will still be required to evaluate efficacy in both stages and as a measure of post-treatment success . Current screening methods for the disease [7] include a serological blood test ( for the gambiense form of the disease only ) , followed by blood microscopy then microscopy of cerebrospinal fluid ( CSF ) to check for parasites . If stage 2 disease is suspected , but no parasites are observed in the CSF , a white blood cell ( WBC ) count above an arbitrary number ( often 5 WBC/μL ) may be used for staging ( Fig 1 ) . However , the requirement for lumbar puncture , coupled to poor sensitivity , makes staging difficult to perform . Since the turn of the millennium , the number of cases of HAT have dramatically reduced [8] and in 2014 , there were fewer than 4 , 000 reported cases [9] . HAT was included in the London Declaration of 2012 , an agreement between pharmaceutical companies , charities , NGOs and endemic countries that endorsed a roadmap set by the World Health Organisation to build towards elimination of five neglected tropical diseases by 2020 [10] . As a result , there has been a push to develop new drugs , non-pharmaceutical interventions and diagnostic tools for HAT . As the number of cases of HAT decreases , new ways of diagnosing the disease become more important . Time-consuming microscopy with specialised technicians is no longer suitable when large swathes of the population are being screened to search for the few remaining patients . Improved diagnostics are critical for staging HAT and finding patients as the elimination campaign proceeds . Moreover , trials on drugs aimed at bringing therapies to assist in elimination will benefit from tests , preferably not involving lumbar puncture , that are indicative of cure of stage 2 . Although identification of parasites in blood or CSF has remained the gold standard in diagnosing HAT , there has been increasing use of serological markers ( circulating antibodies detected by the card agglutination test for trypanosomiasis ( CATT ) and recently introduced rapid diagnostic tests ( RDTs ) based on lateral flow devices [11 , 12] and molecular tools including loop-mediated isothermal amplification ( LAMP ) of parasite DNA [13] . Elevated IgM and proteins in CSF have also been proposed as possible biomarkers for staging , and a series of CSF and plasma based cytokine markers have all been investigated [14 , 15] . A CSF related metabolite , neopterin , however , has emerged as the most sensitive molecular biomarker for staging discovered to date [14 , 15] . A targeted analysis was carried out , given the finding of elevated neopterin in the CSF of patients before , during and after treatment [16] . The value of neopterin as a biomarker was confirmed , and the marker was also found to be predictive of cure following treatment [16] . However , tests based on neopterin still require collection of CSF in order to identify its elevation in infection , and specificity is relatively low given that neopterin is also found in other neurological infections , notably HIV , tuberculosis and malaria , which are frequently co-existent in HAT patients [17–20] . The fact that neopterin is a metabolite , however , emphasises the potential of metabolic biomarkers in diagnostics . Metabolic biomarkers have been used in the diagnosis of a range of conditions for many years . Diabetes , for example , is diagnosed due to elevated glucose levels in the blood [21] , pregnancy tests detect human chorionic gonadotropins [22] and blood creatinine levels [23] are used to identify failures in kidney function . High resolution mass spectrometry can be used to identify a wide range of metabolic species in a technique known as metabolomics [24] . Differences in the abundance of these small ( <1200 Da ) metabolites in the biofluids of infected individuals may be used to discriminate between different disease states with the aim of developing new diagnostic tools [25] and a recent study made the first tentative investigations into possible biomarkers in HAT patients . This study was limited , however , due to a lack of patient stratification , and was only done in T . b . rhodesiense HAT patients [25] . A simple biomarker-based test for HAT would revolutionise the way that the disease is screened . Microscopes and centrifuges would no longer need to be transported into the field , allowing much greater access to the isolated communities still affected by the parasite . To be successful , the test would need to be stable at a range of temperatures for a long period of time , simple and quick to use without extra equipment , reliable ( with a high specificity and low false positive and false negative rates ) and cheap . A dipstick format where a colour change indicates the presence of infection would be ideal . These types of tests are possible with metabolic biomarkers linked to a colour change reaction to detect their presence [26] . Here , we report a comprehensive , untargeted metabolomic analysis of human blood , urine and CSF from 16 seropositive ( but parasite negative ) subjects , 20 stage 1 and 20 advanced stage 2 HAT patients .
Ethical clearance was obtained from the “Direccao National de Saude Publica , Ministerio da Saude” . Written informed consent was received from these subjects prior to enrolment and/or from their parents or guardians for participants below 18 years of age . Any individual who declined to participate was followed up according to the standard procedures of the national control programme . Metabolite extractions—metabolite extractions were performed as per standard procedures [30] in January-March 2015 ( after between 5 and 7 years in storage ) . Samples were checked for metabolite degradation and all passed . Briefly , 5 μL of sample was extracted in 200 μL of UPLC grade chloroform:methanol:water ( 1:3:1 ) on ice . Samples were centrifuged and stored at -80°C before being run through the LC-MS system . LC-MS—Samples were run on a QExactive mass spectrometer ( Thermo ) after separation on a zic-HILIC column ( Sequant ) according to previously published methods [30 , 31] . A 10μL sample injection was used . Data analysis—Raw data were filtered and aligned using mzMatch [24] then further filtering and putative annotation for metabolic features was conducted using IDEOM [32] version 19 using generous parameters ( 0 . 5 minute retention time window for matching to a standard , 3ppm mass error for identification , minimum number of detections of three per group , a peak height intensity filter of 1000 and a relative standard deviation filter of 0 . 8 ) . Data were exported from IDEOM to MetaboAnalyst [33] PiMP ( http://polyomics . mvls . gla . ac . uk/: PCA plots and TICs ) and Graphpad Prism ( histograms ) . Metabolite identification—Metabolic features in this manuscript are named according to their best match based on exact mass , retention time match to an authentic standard , retention time prediction [34] , fragmentation pattern match to MzCloud database ( https://www . mzcloud . org/home . aspx ) and isotope distribution . If an annotation was not possible based on these parameters , then the metabolite exact mass ( neutral ) is given . The evidence collated for each metabolite discussed in this manuscript is summarised in S1 Table . Classification model—Classification models based on Bayesian logistic regression [35] were built in order to provide a system to distinguish stage 1 from advanced stage 2 and to distinguish control from infected subjects in plasma . Each individual LC-MS peak was placed into its own logistic regression model predicting disease state and the deviance calculated . The fifty peaks with the lowest deviance were then picked for further analysis as follows . A recursive feature elimination algorithm [36] was run 10 times ( Monte-carlo cross validation ) to select the best predictors of disease stage , using a maximum of 2 predictors with a logistic regression model . At each run , 10 sub-runs ( Monte-carlo cross validation ) each calculated the area under the receiver operating characteristic curve ( AUROC ) as the metric to maximise . The results of the feature elimination algorithm were an ordered list of the best predictors . Due to the amount of data ( 20 samples in each condition ) , it was decided to develop a model with a maximum of two predictors . The top predictor was found to be m/z 216 , which had a strong and well-separated LC-MS signal . In examining the next predictor , m/z 133 was found to increase performance , have a strong and well-separated LC-MS signal and be identifiable as ornithine , and was therefore chosen as the second factor . Model performance was calculated using 1000 repetitions of Monte-carlo cross validation ( AUROC = 92% , Sensitivity = 92% , Specificity = 81% ) . This was achieved by taking re-sampled subsets of the data and calculating AUROC and then calculating the mean and standard deviation over the subsets . The given sensitivity and specificity were chosen to be at the operating point closest to the perfect sensitivity and specificity . In developing the model to distinguish control from infected subjects , the same procedure as above was followed . In addition , a model using the same predictors as the first model was developed . There was no significant difference between the two procedures , therefore the model using the same predictors as the first model was shown .
Twenty stage 1 patients , 20 stage 2 and 16 controls ( serological suspects in whom parasites were not found ) were randomly selected from a larger cohort and used for metabolomics analysis . Characteristics of these subjects are summarised in Table 1 . Advanced stage 2 patients were clearly identified clinically with the presence of neurological signs and CSF cell counts above 20 WBC/μL . Although a lumbar puncture was performed to confirm diagnosis , a trained practitioner could have made the diagnoses clinically in all 20 cases . Stage 1 patients are not easy to ascertain clinically . Here , 13/20 presented with symptoms , but only seven with signs more specific to HAT . Signs and symptoms from these patients are presented in S2 Table . Control patients presented general symptoms in six cases as well as having tested CATT positive indicating the possible presence of HAT or other diseases . All included patients were negative for HIV and syphilis . One stage 1 patient was identified microscopically with malaria at diagnosis . None of these patients showed evidence of other co-infections . The control group comprised individuals who were serologically positive using the CATT test at high titres ( 16 or 32 ) , but in whom parasites were not found using microscopy combined with concentration techniques . It is not ethically acceptable to extract CSF from patients in whom the disease is not initially suspected , so seronegative controls from the same region could not be sought . The CATT test is not 100% specific and will produce some false positives , however , it is also possible that some patients in our control group did contain undetected trypanosomes . Among the control subjects , one presented microfilaria in the blood , one had signs of bronchitis and one had rheumatoid pains . Given the high level of clinical suspicion and presence of raised CSF cells , three subjects were treated with eflornithine . Nine control patients were followed without any treatment and five became seronegative during follow up while the remaining four were lost to follow up , and their outcome was unknown . The five seropositive patients that became seronegative either “self-cured” or were never infected . Spectral profiles—After data processing and filtering , 1 , 021 robust metabolite features ( features that were reproducible in terms of mass and retention time , above an intensity cut-off and present in at least three samples of a sample group ) were detected in patient plasma , 512 in CSF and 694 in urine . The high number for plasma reflects the richness of this biofluid compared to CSF . The lower number in urine compared to plasma is likely due to computational filtering based on the concentrations of the metabolites being so variable that the software was unable to include in meaningful analysis . Total ion count variability between samples within groups was below 0 . 06 relative standard deviation in CSF and plasma samples , and the number of metabolite features detected for each patient in a group did not vary significantly in plasma or CSF ( S1 Fig ) . Around half of the plasma peaks could not be annotated for a metabolite identity , which is usual for untargeted metabolomics data , given the large numbers of metabolites that have yet to be formally characterised chemically [38] . A larger proportion of the metabolites in CSF and urine were putatively annotated . The coverage of the metabolome varied somewhat between the three different biofluids , with more fatty acyls detected in CSF compared to urine or plasma , and more glycerolipids discovered in plasma ( Fig 2 ) . We specifically sought the trypanosomatid specific metabolite trypanothione , but could not identify it in samples , probably because the low parasitaemia characteristic of T . b . gambiense infections would assure it remains below the detection limit ( 5 nM ) . It has been shown that trypanosomes also secrete large quantities of keto acid derivatives of the aromatic amino acids , e . g . phenylpyruvate , enol-phenylpyruvate , indole pyruvate and indole lactate [39] , however , these were not detected in blood or CSF , which again is likely to be attributable to the low parasitaemia . Urine metabolites vary in concentration making biomarker discovery challenging—An analysis of the total ion chromatogram for each urine sample revealed a very large variation in the concentration of the samples ( Fig 3a ) . This is a common problem in urine samples as the water levels in the samples are not controlled . There are debates on the best method to normalise urine sample data to account for the wide variation in concentration [40] . Creatinine levels are often used for this type of normalisation , but can cause over fitting of the data , resulting in amplification of irrelevant differences [33] . A principal components analysis of the urine data showed that the samples did not cluster into groups , either before or after creatinine normalisation ( Fig 3b and S2 Fig ) . Univariate analysis of individual metabolites also failed to identify any whose abundance clearly varied between patients at stage 1 or advanced stage 2 and controls . CSF findings—HAT and control patients separated slightly in a principal components analysis plot ( Fig 4a ) . Previous studies have used neopterin in CSF as a marker for stage 2 trypanosome infection; our data confirmed neopterin as a good marker with a sensitivity and specificity of 100% for discrimination between stage 1 and advanced stage 2 infections ( Fig 4b ) . This finding also corroborates the untargeted metabolomics approach as offering unrivalled potential for novel biomarker discovery for HAT . Four control patients ( A010C , A011C , A013C and A020C ) showed high levels of neopterin in their CSF above the cut-off in the ROC ( receiver operating characteristic ) curve of 24 , 342 ( S3 Table ) . Several metabolites were altered following trypanosome invasion of the CNS , including an increase in 5-hydroxytryptophan , and a decrease in tryptophan in advanced stage 2 patients . This change was accompanied by a small increase in kynurenine in advanced stage 2 disease ( S4 Table ) . The ten metabolites ( apart from neopterin ) showing the greatest significant change in abundance in advanced stage 2 infection compared to stage 1 infection are shown in Fig 4b . Notably , there are several patients with much higher levels of m/z 188 , O-acetylcarnitine and m/z 202 . Using a cut-off for each of these 10 metabolites , determined by the greatest combined sensitivity and specificity ( Gain in certainty [41] ) for each metabolite intensity , the individual performance of each biomarker for every patient was determined ( Fig 5 and S3 Table ) . It is common in untargeted metabolomics experiments to detect numerous metabolites which are not yet represented in databases , given the huge diversity of metabolite space [42] . Indeed , some of the metabolites measured here could not be identified based on their mass and retention time . Fragmentation patterns and isotope distributions also failed to provide clues to their identities . All of these masses are singly charged and have normal carbon isotope patterns . Mass m/z 646 . 4512 has a single positive charge , a bimodal peak shape , and a predicted formula of C34H68O5N2PS . Additional analytical approaches would be required to provide robust identities of these metabolites . In addition to 5-hydroxytryptophan , which is a key metabolite associated with somnolence , we also identified metabolites with masses allowing putative annotation ( based on mass and formula ) as linoleamide and oleamide . These metabolites were both increased in stage 1 and advanced stage 2 samples compared to the control group , albeit without reaching statistical significance ( Fig 6 ) . Given the inter-sample variability and low sample size , caution is warranted in interpreting any possible effect . The synthesis of linoleamide has not been studied , but oleamide has been shown to be synthesised from oleic acid and ammonia ( and to a lesser extent , glutamine ) [43] . The likely precursors of linoleamide and oleamide ( linoleic acid and oleic acid ) were not detected . Biomarkers in plasma—Metabolic features in plasma samples were generally very well regulated , and significant deviations from normal levels were minimal due to normal homeostatic regulation . Sample groups were not separated on a PCA plot ( Fig 7a ) . There were , however , 308 metabolic features whose abundance was significantly altered between stage 1 and control patients , 181 between advanced stage 2 and uninfected , and 237 between stage 1 and advanced stage 2 ( S5 Table ) . None of these significant changes were large , however two masses ( Fig 7b and 7c ) were able to produce a good model ( Fig 7d ) that separated stage 1 and advanced stage 2 infections . This model had an area under the ROC curve of 92% and could be marginally improved by adding two more masses ( S3 Fig ) . Confidence intervals for sensitivity and specificity can easily be calculated for any operating point by examining Fig 7d . M/z 133 . 0971 , detected in positive ionisation , was decreased in advanced stage 2 disease compared to stage 1 and was identified as ornithine ( match to standard retention time , fragmentation pattern and expected isotope distribution ( S1 Table ) ) . M/z 216 . 1958 , detected in positive ionisation , was increased in advanced stage 2 disease compared to stage 1 and was annotated as aminododecanoic acid ( based on its mass plus fragmentation and isotope distribution ) . A second model ( Fig 7d ) was built that separated control and infected subjects using the same peaks as the first model . This model had an area under the ROC curve of 94% and showed 87% sensitivity and 95% specificity at the best point on the curve .
Many studies show metabolite differences that may act as biomarkers of infectious diseases in sub-Saharan Africa [18 , 44 , 45] , including a recent paper analysing metabolic biomarkers in T . b . rhodesiense infection [25] where changes in amino acid and lipid metabolism compared to uninfected control patients were reported , although robust markers that would be suitable for diagnostics were not proposed [25] . We were able to find changes in the levels of ornithine and aminododecanoic acid in blood that were predictive of both the presence of disease and disease stage . These metabolites were not detected in the study by Lamour et al . [25] , and so it would be interesting to test samples from T . b . rhodesiense HAT patients to see if they are also altered upon infection in an East African cohort . A greater challenge for HAT diagnostics , however , is not to diagnose an infection , but to use an alternative to CSF to accurately stage the disease once infection has been detected by microscopy or a serological test . This staging would ideally avoid the need for dangerous lumbar punctures in the field allowing correct treatment to be given . Urine could be of great value as a source of biomarkers , since its collection is not invasive , and samples can be used with minimal preparation . Several urine based biomarkers have been proposed e . g . for diabetes , prostate cancer [46] , bladder cancer [47] and possibly for diseases of the brain [48] . Unfortunately the high degree of variation in urine water content and therefore metabolite concentration confounds its utility . This could be ameliorated by providing controlled volumes of water at predetermined times before collecting urine , although this was not done in the current study , and may not be feasible in large HAT screening campaigns in remote field settings . Variability in urine metabolite levels in this dataset therefore made it difficult to identify metabolites that would be predictive of disease . CSF is mainly produced by ependymal cells of the choroid plexus in the brain , turning over around four times per day , washing the central nervous system of metabolic waste [49] . CSF contains much less protein than blood plasma . Moreover , pH and levels of different neurotransmitters and various metabolites must be tightly regulated to avoid damage to the brain . In our analysis there were clear differences between the CSF of patients in stage 1 and late stage 2 HAT . Neopterin has previously been seen in stage 2 HAT patients , but its use as a biomarker is limited by the wide range of inflammatory disorders it is predictive of [18 , 50 , 51] . That , and the fact that our cohort contained only advanced stage 2 samples , is why specificity is much lower in the field [50] . Four control patients had levels of neopterin over our cut-off . It is possible that these individuals had a separate infection that was not diagnosed , emphasising the limitations of neopterin as a biomarker when used in isolation . The increase in 5-hydroxytryptophan and the decrease in tryptophan seen in our CSF analysis could be due to an increase in the activity of tryptophan 5-monooxygenase . The depletion in tryptophan levels has been noted before in blood of rodents infected with trypanosomes [52 , 53] and dietary tryptophan is rapidly metabolised to tryptophol and indole acetic acid [54] . Tryptophan depletion is also common in inflammation due to its conversion into metabolites of the kynurenine pathway ( kynurenine also being increased in advanced stage 2 disease ) [52 , 55] . However , we failed to identify significant changes in metabolites of this pathway in blood , and the majority of these metabolites were not detectable in human CSF ( 5-hydroxytryptophan , kynurenine and tryptophan being the exceptions ) . Previous studies have shown that single nucleotide polymorphisms in tryptophan 5-monooxygenase could be linked to neuropsychiatric disorders [56] . A substantial increase in 5-hydroxytryptophan is seen in aromatic L-amino acid decarboxylase deficiency , where patients are rendered in a vegetative state [57] and treatments with 5-hydroxytryptophan have been used to treat depression caused by serotonin deficiencies [58] . 5-hydroxytryptophan may be a possible marker of stage 2 infection , and could increase the specificity of neopterin for staging using CSF . It would be interesting to investigate whether 5-hydroxytryptophan levels revert to normal after treatment and could therefore be used as a test of cure . The increase in levels of putative linoleamide and oleamide in both stage 1 and stage 2 HAT is interesting . These molecules are known to increase in the brains of sleep deprived animals [59–61] , and oleamide has been shown to induce sleep when injected into rats [59] . It may be that the changes associated with sleep disturbances in HAT start to occur in stage 1 of the disease , as has previously been suggested [62] . Linoleamide and oleamide may be early markers of sleep disturbance , although further work on the identity of these metabolites ( so far based solely on mass ) and experiments to determine roles would be needed , as are larger patient cohorts to determine significance . The CSF metabolites shown in Fig 5 can be used to stage patients with 100% specificity and sensitivity , but will still require a lumbar puncture—something which would be avoided with blood based biomarkers . Whole blood is routinely taken in the field when screening for HAT by microscopy , and a biomarker that could detect the presence of trypanosomes , while also staging the disease would be ideal . Several biomarkers were found that had significantly altered levels in the blood of advanced stage 2 patients compared to stage 1 patients , and two of these were able to produce a robust model able to stratify patients by disease stage . As shown by the ROC curves in Fig 7d , by using the logistic regression classification models , we were able to find changes in the levels of ornithine and aminododecanoic acid in blood that were predictive of both the presence of disease and disease stage . Ornithine levels were decreased in advanced stage 2 disease compared to stage 1 . Ornithine is used by trypanosomes to synthesise polyamines and trypanothione—the main reducing thiol in the cells . In normal adult blood , there is an estimated 50–100 μM of ornithine ( www . HMDB . ca ) and trypanosomes have been shown to import significant amounts of ornithine [30] . The increased ornithine in the blood of stage 1 HAT patients compared to advanced stage 2 may reflect an increased production in the blood to compensate for the uptake by the parasites . However , the very low parasitaemia in gambiense patients would make it doubtful that it is parasite consumption of ornithine that leads to its loss , in which case hitherto unknown roles for ornithine in infection might be at play . The second mass used in the model was putatively annotated as aminododecanoic acid and was increased in advanced stage 2 disease . Aminododecanoic acid is not a naturally occurring metabolite and may be a breakdown product of a nitrogenous lipid . These metabolites warrant further investigation in another cohort to determine whether they can be used to stage HAT in blood . Limitations—This pilot study identified several biomarkers , but had a number of limitations . Firstly , the numbers of patients in the cohort tested were small ( 16–20 per arm ) . It would be very interesting to see whether the metabolites identified could be used for diagnosis and staging in a second , larger cohort . If these markers are able to diagnose and/or stage disease in a second cohort , then their development into prototype tests should be prioritised . Secondly , there is ambiguity as to what constitutes early stage 2 and late stage 1 infection . For this reason , advanced stage 2 patients were compared to stage 1 patients in our study . Analysing the levels of the proposed biomarkers in early stage 2 infections will be vital to validating their use in the field . Thirdly , the identity of some of the biomarkers could not be achieved using our mass spectrometry platform . To understand the mechanisms behind the changes seen , it would be useful to get true identifications of all the metabolites seen . This could be achieved using stepped fragmentation mass spectrometry or , if levels in the tested biofluid are high enough , nuclear magnetic resonance . Conclusions—The work reported here reveals the extraordinary power of the untargeted metabolomics approach to identify biomarkers of disease . The finding of metabolites whose abundance is predictive of infection and able to discriminate between stage 1 and advanced stage 2 disease , even using levels of metabolites found in blood , offers the potential of removing the need for lumbar puncture from HAT staging algorithms . These results would , however need to be confirmed using a large , independent cohort of patients , before they can be developed into a useful test . A rapid diagnostic test developed for HAT screening will need to be cheap , fast and easy to use , used with minimal sample preparation , non-refrigerated and accurate . The sensitivity and specificity of the test will depend on whether active or passive screening is required . For active screening , a more sensitive test may be required to detect more cases , whereas in passive screening more specificity may be required to avoid false positives . The levels of the metabolites used as a cut-off in our model can therefore be altered to achieve a more desirable sensitivity vs specificity trade-off .
|
Human African trypanosomiasis , also known as sleeping sickness , is a parasitic disease that affects people in sub-Saharan Africa . There are two stages of the infection . The first stage involves parasites proliferating in the bloodstream following introduction via the bite of an infected tsetse fly . The second , more serious stage , involves parasite invasion and proliferation within the central nervous system causing characteristic disturbances to the patients’ sleep wake patterns and progressive appearance of other neurological signs , including walking disabilities behaviour changes , abnormal movements , incontinence , then ultimately coma and death . Drugs are available to treat both stages of the disease , but the drugs for stage 2 disease have serious side effects and must be administered in hospital settings . Stage determination is thus a key element for disease management . Currently staging involves microscopic evaluation of CSF following a lumbar puncture . Here , we have analysed the metabolome of CSF , blood and urine of patients to seek biomarkers to stage the disease based on these biofluids . CSF and blood fluids were found to have distinctive metabolic biomarkers and when several of these metabolites are combined , a sensitive and robust discriminatory staging test can be developed . Some CSF metabolic markers relate to brain inflammation , whilst others may be related to somnolence associated with the disease in stage 2 patients , which may also help in understanding disease progression . Interestingly , distinctive biomarkers were also found in plasma , potentially abrogating the need for diagnostic lumbar punctures in the future .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"nervous",
"system",
"african",
"trypanosomiasis",
"tropical",
"diseases",
"biomarkers",
"parasitic",
"diseases",
"urine",
"metabolomics",
"metabolites",
"neglected",
"tropical",
"diseases",
"infectious",
"diseases",
"zoonoses",
"protozoan",
"infections",
"trypanosomiasis",
"hematology",
"biochemistry",
"blood",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"metabolism",
"cerebrospinal",
"fluid"
] |
2016
|
Metabolomics Identifies Multiple Candidate Biomarkers to Diagnose and Stage Human African Trypanosomiasis
|
Domestic dog rabies is an endemic disease in large parts of the developing world and also epidemic in previously free regions . For example , it continues to spread in eastern Indonesia and currently threatens adjacent rabies-free regions with high densities of free-roaming dogs , including remote northern Australia . Mathematical and simulation disease models are useful tools to provide insights on the most effective control strategies and to inform policy decisions . Existing rabies models typically focus on long-term control programs in endemic countries . However , simulation models describing the dog rabies incursion scenario in regions where rabies is still exotic are lacking . We here describe such a stochastic , spatially explicit rabies simulation model that is based on individual dog information collected in two remote regions in northern Australia . Illustrative simulations produced plausible results with epidemic characteristics expected for rabies outbreaks in disease free regions ( mean R0 1 . 7 , epidemic peak 97 days post-incursion , vaccination as the most effective response strategy ) . Systematic sensitivity analysis identified that model outcomes were most sensitive to seven of the 30 model parameters tested . This model is suitable for exploring rabies spread and control before an incursion in populations of largely free-roaming dogs that live close together with their owners . It can be used for ad-hoc contingency or response planning prior to and shortly after incursion of dog rabies in previously free regions . One challenge that remains is model parameterisation , particularly how dogs’ roaming and contacts and biting behaviours change following a rabies incursion in a previously rabies free population .
Rabies is among the most lethal infectious diseases , present on all populated continents except Australia [1] . The domestic dog remains the most important vector worldwide , causing >95% of all human cases [2–4] . Despite availability of an effective vaccine for more than a century and repeated demonstration that vaccinating the domestic dog population is the most effective way to eliminate the disease [5–8] , rabies remains endemic in large areas in Africa and Asia . Moreover , the disease has ( re ) emerged in areas previously free ( such as Bhutan [9 , 10] , Indonesia [11 , 12] , and the Central African Republic [13] ) . Rabies continues to spread through the Indonesian archipelago via human mediated domestic dog movements [11 , 12 , 14] , most recently through the previously rabies-free province of Maluku in eastern Indonesia [11 , 15] . The risk of incursion into rabies-free areas − Timor , Irian Jaya , Papua New Guinea ( PNG ) and northern Australia − is therefore high . Possible incursion scenarios into Australia include yachts or fishing boats hosting latently rabies infected dogs traveling from Indonesian islands to remote areas in northern Australia [16] . Also , close cultural ties between PNG and Torres Strait Island communities exist , increasing the risk of movements of dogs incubation rabies from PNG to Australia , if an incursion in PNG occurs [16] . In these regions large , free-roaming domestic dog populations [17 , 18] increase the risk of rabies establishment , which would subsequently impact human and wildlife populations . Because there are no historical precedents , the spread and final impact of such rabies incursions is difficult to estimate . However , such knowledge is critical to informing preparedness and response plans prior to an incursion , and to design the most effective strategies . Descriptions and applications of several rabies models in wildlife [19–24] , domestic dogs [5 , 7 , 25–28] or a combination of these [8 , 29] have been published . All have been based on empirical field data in rabies endemic regions and typically aim to inform policy on reducing rabies prevalence and thus impacts . However , for a region in which rabies is exotic , predictions of the effectiveness of different interventions following the initial detection of rabies are more relevant . An issue is how rabies behaves when introduced to a previously free population , particular the effect of rabies on contact rates and biting rates . Evidence on these behaviour changes from previous rabies incursion may serve as an approximation but is typically vague and therefore equivocal . To our knowledge , epidemic models simulating rabies invasion in regions never exposed to rabies do not exist , a barrier to rabies preparedness planning . Here , we describe the development of a novel simulation model of rabies epidemics in domestic , free-roaming dog populations in remote Indigenous communities in northern Australia–as an example of the potential scenario in many regions of the world where rabies is absent but where the risk of a rabies incursion is present and its spread is likely due to large populations of free-roaming domestic dogs . Results of a systematic sensitivity analysis are also presented and model application options are discussed .
Data collection required to estimate model parameters has been approved by the Human Ethical Committee of the University of Sydney , grant no . 2013/757 and the Animal Ethical Committee of the University of Sydney , grant no . N00/7-2013/2/6015 . The rabies simulation model development was based on data from two distinct regions in northern Australia , the Northern Peninsula Area ( NPA ) of Cape York , Queensland and Elcho Island , the Top End of the Northern Territory . Characteristics of the two study sites are presented elsewhere [18] . Briefly , in the NPA five Aboriginal and Torres Strait Islander communities are located in close proximity ( 2−4 km ) to each other . On Elcho Island one larger Aboriginal community is present . The dogs are typically owned but unrestrained and build a large population in all communities ( human:dog ratio 2 . 7−8 . 8 per community , Table 1 ) [18] . The dog population in the NPA − which informs the simulation model − is based on the most recent dog census conducted by the NPA Regional Council in 2009 . As such information was not available from Elcho Island , the number of dogs are calculated based on the average human:dog ratio of the NPA communities and official human census data from Elcho Island in 2011 ( http://www . censusdata . abs . gov . au/census_services/getproduct/census/2011/quickstat/SSC30094 ) and similar household sizes as in the NPA are assumed . The model developed is stochastic , spatially explicit , based on individual dog data and assumes a daily simulation time unit . It starts with the introduction of a latently ( non-clinical ) infected dog and ends when no infected dog remains . The exact location of each dog’s home is known and a closed dog population within the region is assumed , but dog movements between regional communities are simulated . In the model , 429 and 410 dogs in 175 and 163 households are included in the NPA and on Elcho Island , respectively ( 137−451 and 68−142 households per km2 , respectively , Table 1 ) . The two regions are simulated separately . The average number of dogs per dog holding household ranges from 1 . 9–3 . 2 per community . Parameter value definition is a critical component in modelling studies , driving the outcome of any simulation or mathematical model . While some parameter values can be taken from the literature ( e . g . disease or vaccine related parameters ) , other parameter depend on the settings in the specific environment in which the model was developed or applied . Seven out of 37 ( 19% ) parameter values of the model presented here are sourced from the literature ( those of rabies virus and vaccine related parameters ) and 23 ( 62% ) are based on assumptions ( S1 Table ) . The latter can further be classified as experimental parameters ( parameters defining control strategy implementation as e . g . delay in starting control strategies or vaccination coverage , 16/37 [43%] ) and parameters for which value information are currently lacking ( e . g . bite probability given a contact , owner compliance to cease dog movements , 7/37 [19%] ) . The remaining seven ( 19% ) parameter values were estimated based on our field collected data , including contact data within and between communities and mean distance between households used for dog confinement strategy ( S1 Table ) . Data used to calculate the distance kernel function applied for contact rates between dogs of different households was derived from a large scale GPS study on 69 domestic dogs in all of the six communities ( S5 and S6 Tables ) [18] . The number of contacts between each pair of dogs within the same community was extracted using the definition of contact being within 20 meters during the same minute . As the model runs on a daily basis , the contact information was converted into a binary variable with two dogs having at least one ( 1 ) or no ( 0 ) contacts within 24 hours . This binary outcome was analysed by a logistic regression model with the known distance between the two dogs’ homes as the explanatory variable . The outcome variables estimated by the logistic regression ( intercept α , coefficient β and standard error of the coefficient βse ) were further used to build the distance kernel ( Eqs 1 and 2 ) . Daily contact probability of two dogs living in the same household was estimated based on the same dataset plus similar data collected during the post-wet season ( monsoon ) in the same communities . One of 31 ( 3% ) pairs of dog living in the same household was not observed to have at least one contact per 24 hours . This within-household contact probability was implemented as a uniform distributed parameter with 97% as the mean . Four parameters defining the dog movements between communities − both short term and permanent − were estimated from questionnaire data collected in the NPA ( S2 Table , approved by the Human Ethical Committee of the University of Sydney , # 2013/757 ) together with observations of short term movements by GPS and of permanent movements of dog owners from one NPA community to another during a year . Twenty-nine dog owners were interviewed in September 2013 and one in September 2014 , including questions on frequency of dog movements to other NPA communities due to pig hunting or other trips ( e . g . visits or work ) . A daily movement probability of 0 . 058 per dog was calculated from these reported data , while from the study we observed that only 8 of 81 ( 10% ) dogs were moved during an observation period of 6 . 2 days , resulting in daily movement probability of 0 . 016 per dog . Combining these reported and observed data and giving twice the weight to observed data , 0 . 03 was defined as the beta-pert distribution mode for daily short term movement probability per dog; 2- and 0 . 5-fold values were used for the minimum and maximum limits of the distribution , respectively . The duration of the short term movements were derived from the questionnaire in which all hunters reported trips of one to two days and observations from the GPS study in which all 8 dogs stayed less than one day in the community visited . The frequency of permanent movements was estimated from questionnaire data and observations of permanent movements during September 2013 and September 2014 . Dog owners reported that 18% ( 6/33 ) of the NPA dogs originated from a different community within the NPA , which resulted in an estimated probability of permanent movements of 1 . 64*10−4 per dog per day assuming an average dog life of three years . In addition , owners of 6% ( 3/49 ) of the dogs were observed to have permanently moved between NPA communities during the year , resulting in a probability of daily movements of 1 . 64*10−4 per dog . The sum ( 3 . 3*10−4 ) was selected as the mode of the beta-pert distributed daily probability of permanent movements per dog , with 0 . 5 and 2-fold values for minimum and maximum . Finally , the destination community for both permanent and short term movements was observed to be more frequently a neighbouring community than any other; consequently a neighbouring community as the destination was assumed to be twice as likely as for any other community in the NPA . The median distance between each household and its closest neighbour–used in the model to truncate the distance kernel for the dog confinement control strategy–were estimated for each community separately using the coordinates of all households per community . Household coordinates were derived from Google Earth ( http://earth . google . com/ ) where placemarks were set on all private dwellings , identified with the help of community maps . The distances between each household and its closest neighbour were calculated and the mean per community implemented in the model as a fixed value parameter . In the first step of the sensitivity analysis ( SA ) , all model parameters used for six different control strategies were tested using the strategy’s default values: a ) vaccination with 70% coverage either pre-emptively ( PV ) or reactive ( RV ) ; b ) culling of dogs contacted by a rabid dog ( CC ) or reactive culling ( RC ) with culling levels of 80 and 50% , respectively; c ) dog confinement plus movement bans between communities ( MB ) with 80% and 90% compliance , respectively; and d ) a non-intervention strategy ( NI ) . For all of these strategies , including NI , culling of dogs detected rabid ( DC ) was applied . The number of index dogs was set to 1 and randomly selected in the region . For each of the 12 combinations of the two regions ( NPA and Elcho Island ) and six control strategies , 1000 model repetitions were simulated . For stochastic parameters , the mode or mean ( for beta-pert distributed and uniform distributed parameters , respectively ) was allowed to range between ±25% around the default value while the difference between the minimum and maximum values was kept fixed ( no variation of the distributions’ shape; S3 Table , S1 Fig ) . Deterministic parameters were allowed to vary ±25% around the default value . Variation of 25% has been chosen to allow enough variation for parameters with wide distributions ( e . g . the infectious period ) and avoid too large distinction between the lower and upper limit of narrow distributed parameters ( e . g . the rabies transmission probability given a bite ) . The distance kernel was tested using three different shapes representing a minimal kernel and increased probabilities for short and long distance contacts , respectively ( S2 Fig ) . The influence of the index community within the NPA region was investigated by defining one of the five communities hosting the index dog . To explore the sensitivity of the model on the weighting matrix to choose the destination community for between-community movements , an alternative matrix was tested beside the default with equal chance for all communities to be selected as the destination . The values of all parameters tested were randomly selected from the described ranges so that for each simulation , an individual set of parameter value combination was chosen . For each of the 12 region-strategy combinations , linear multivariable regression analysis was modelled with the outbreak duration and–where applicable–outbreak size as the response variable and the parameter values as explanatory variables . For the stochastic parameters the mode ( beta-pert distribution ) and mean ( uniform distribution ) values were modelled as explanatory variable values . Correlations between the parameter values were explored using Kendall’s tau correlation , Chi-Square and Wilcoxon Rank Sum test for two continuous , two categorical and a continuous and categorical parameter , respectively . Because the assumption of a normally distributed response variable was not always met ( S3 Fig ) , logistic regression following the same principle was modelled defining an outbreak with a duration or size above the median as 1 and as 0 otherwise . Based on both the linear and logistic regression analyses , parameters were defined to be highly ( statistically significant p-values < 0 . 05 in ≥ ¾ of all tested regression models ) , low ( statistically significant p-values < 0 . 05 in < ¼ of all tested regression models ) or moderate ( otherwise ) sensitive to the outbreak duration and size . Additionally , scatter plots of the outbreak duration and size over the range of each parameter value were visually analysed and correlations were calculated between outbreak duration and size and parameter values with continuous scale . A correlation of >|0 . 1| was considered as a threshold to distinguish between sensitive and non-sensitive parameters . Parameters found to be highly sensitive in either of the regression analyses during the first step of the SA were further explored in a second step to identify the influence of their mode or mean ( default , large , small ) and shape ( default , narrow , wide ) on the model’s outcome . For the beta-pert and uniform distributed parameters nine combinations per parameter with the three values of mode or mean ( default and ±10% ) , and three values of difference to the minimum and maximum ( default and ±10% ) of the distribution were defined ( S4A–S4C Fig ) . For the vaccine efficacy parameter , the variation of 10% had to be reduced to 4% , which was the highest variability still ensuring a maximal value < 1 , a requirement for probabilities ( S4D Fig ) . For each parameter , 1000 repetitions were simulated for the same 12 region-strategy combinations described in step 1 of the SA ( in case of the vaccine efficacy only the scenarios for RV and PV ) , where one of the nine options was randomly chosen while all other parameters in the model were kept at their default value . The distance kernel , defined by the three variables α , β and βse , was explored by varying the three variables around a default value ±50% , resulting in 27 combinations ( S5 Fig ) . Six thousand repetitions were simulated for all 12 region-strategy combinations with a randomly selected distance kernel out of the 27 options while all other model parameters were kept at their default values . The outcome was analysed visually comparing boxplots of the outbreak duration and number of rabid dogs . A critical question for stochastic models is always , how many repetitions are required to sufficiently reflect the variability of the model ? The coefficient of variation ( CV = standard deviation/mean ) of model outputs’ mean has been proposed as a measurement to determine the critical number of repetitions required [30 , 31] . The CV of the estimated mean of outcome of interest ( e . g . outbreak size or duration ) over n model simulations is expected to approach 0 for infinite sample sizes n and a threshold of the CV of 15% has proposed to predict outputs with acceptable precision [30] . We used the same approach , but reduced the CV threshold value to 3% . To demonstrate the model’s functionality for the different control strategies , 1000 outbreaks were simulated for the default strategies: 1 . reactive vaccination with 70% coverage at the household level ( RV ) ; 2 . reactive culling with 50% of the dogs culled in affected communities ( RC ) ; and 3 . dog confinement between and within communities with 90 and 80% compliance , respectively ( MB ) . Culling of dogs detected rabid ( DC ) was applied for all of these strategies . The model was simulated in both regions , NPA and Elcho Island , separately . As outcome measures , the epidemic duration and size , i . e . the number of rabid dog and the number of dead dogs ( including rabid and culled dogs ) were calculated and visually compared between the different scenarios .
Two measures of outbreak size are the cumulative number of rabid dogs and cumulative number of dead dogs ( due to rabies plus culled ) . The outbreak duration is defined as the number of days from the introduction of the latently infected index dog until the death of the last infectious dog . The simulation model resulted in plausible results comparing outputs for the three different default control strategies ( Fig 4 ) . Rabies spreads through the communities in a wave pattern and , depending on the control strategy , can kill the entire dog population ( S6 Fig illustrates an example epidemic curve ) . The reactive culling ( RC ) strategy reduces the number of rabid dogs; however the number of dead dogs is only slightly less than the total dog population . For the reactive vaccination ( RV ) strategy the number of rabid dogs ( equal to the number of dead dogs ) showed higher variability among the 1000 model simulations compared to RC , but in both regions , the median of RV was lower than for RC . Obviously , vaccination saves the dogs from death in contrast to culling strategies . The movement ban ( MB ) strategy showed a slight decrease of the outbreak size in the region of Elcho Island whereas no effect was observed in the NPA . However , it was found to be the strategy with the longest durations of outbreaks , demonstrating that movement bans ( if not 100% compliant ) only slow the speed of spread rather than reducing its size . The reduction in the number of movements between communities for MB was obvious , decreasing from a median of 52 ( RV ) and 46 ( RC ) to 19 ( S7 Fig ) . Overall , outbreak duration ranged from 1−20 months ( median 6 . 7 months ) and was more homogenous between interventions than the outbreak size ( Fig 4 ) . Outbreaks lasting for one month did not spread beyond the index dog . For the RV strategy , the vaccination coverage was set at 70% of the households , producing dog level vaccination coverage of 59−75% and 56−76% for the NPA and Elcho Island region , respectively . The control strategy with the largest number of simulations required ( n = 490 ) to capture the variability in the model’s output with the defined CV threshold of 3% was found to be the MB strategy in the NPA ( S8 Fig ) . The number of secondary cases was reported for every rabid dog over the duration of the outbreak . From these records , the basic reproductive ratio R0 was calculated and defined as the mean number of secondary cases for dogs becoming infectious within the first phase–i . e . up to its peak–of an epidemic . The peak of the epidemic is defined as the day with the highest number of newly infectious animals over the entire outbreak . R0 ranged from 0−6 . 1 ( median 1 . 8 ) for RV , 0−6 . 1 ( median 1 . 7 ) for RC and 0−5 . 7 ( median 1 . 7 ) for MB ( S9 Fig ) with an overall median of 1 . 7 ( Fig 5A ) . The epidemic peak was reached on average after 93 days ( Fig 5B ) with a mean of 17 newly infected dogs ( Fig 5C ) . The number of secondary cases derived from each index dog was highly variable and ranged from 0 to 79 ( median of 25 ) for NPA and 4 to 106 ( 40 ) for Elcho Island . Over the duration of the outbreak , the effective reproductive ratio Rt and the number of dogs in the susceptible population decreased in a wave pattern ( S10 Fig ) . The value of 1 for mean Rt is reached during the second or third wave . This reflects that Rt depends on the dogs remaining in the population and finally the outbreak dies out because there are no susceptible dogs left that are close enough to the infectious dogs . The simulation model outputs were highly sensitive to seven parameters: incubation period ( G1 in S4 Table ) , transmission probability given a bite ( G8 ) , distance kernel ( G5 ) , bite probability given a contact between dogs of different households ( G7 ) , vaccine efficacy ( V5 ) , index community ( G12 ) and delay in starting the control strategy of movement restrictions between communities ( B2 ) . The same sensitive parameters , in addition to the detection delay of the first clinical case , were also identified via correlation tests , with the exception of B2 . These outcomes were also confirmed by scatterplots , which express particularly dependencies between the outbreak duration and the incubation period , distance kernel and index community ( S11 Fig ) . Significant correlations between parameters included in the regression analyses were only observed between categorical and continuous parameters where 2 . 1−12 . 1% ( mean 5 . 4% ) of all parameter combinations resulted in Wilcoxon Rank Sum test p-values <0 . 05 . The influence of this subset of parameters was further explored in step 2 of the SA , with the exception of the index community and the delay in commencing movement restrictions because these two parameters directly relate to incursion and intervention scenarios . For all parameters , except the distance kernel , it was found that both the mode and mean ( for beta-pert and uniform distributions , respectively ) and the shapes influence outbreak duration and number of rabid dogs ( S12 Fig ) . The mean and mode were found to have a greater impact , particularly for the incubation period and rabies transmission probability . For the distance kernel , the regression coefficient β was most influential on both the number of rabid dogs and the outbreak duration , followed by βse ( standard error of β ) , particularly for the Elcho Island region ( S13 Fig ) . Outputs were less sensitive to the intercept α .
The model described herein provides insights into short-term rabies epidemics occurring within a small spatial extent in previously rabies-free regions . This is of crucial value for contingency planning in areas where rabies is exotic and the model fills a gap in the published literature on rabies models . The example of Bali , Indonesia demonstrates the impact of a rabies incursion on an under-prepared region [12] . Late detection of the disease , lack of surveillance strategies and an unsuccessful initial response ( focused on dog culling ) resulted in island-wide disease spread and high impacts on both dog and human populations [7 , 12 , 32] . Another example in Indonesia–comparable to communities in northern Australia with a high density of free-roaming dogs and limited veterinary health services − is the island of Flores in East Nusa Tengarra province [14 , 33 , 34] . There , one or more latently infected dogs were introduced , developed clinical rabies and transmitted the disease to local dogs . The disease consequently spread throughout island with a considerable impact on dogs and humans . This is another example of a rabies invasion in a new area with very severe impact . Another novel aspect of the current model is the inclusion of individual susceptible and rabid dogs modelled within a continuous spatial dimension , an approach previously used to simulate highly infectious diseases of livestock ( e . g . foot-and-mouth-disease [35 , 36] ) but not rabies . To date , published dog rabies models have been based on mathematical differential equations [5 , 24 , 25 , 27] or spatially explicit models simulating the spread of rabies within grids [7] . Our approach has several advantages , including stochasticity to capture epidemic variability , incorporation of detailed population structure to better represent real target regions of interest , and detailed spatio-temporal model outputs . By simulating three default control measures–vaccination , culling and dog confinement–our model produced plausible results , suggesting it adequately captures how an epidemic of an infectious disease with a relatively long incubation period would develop in a previously uninfected population . The disease spread temporally in wave-like patterns , peaking on average about three months after an incursion . The average R0 that was estimated from the model outputs was 1 . 7 , consistent with previous estimates of rabies spread in endemic countries [16 , 37] and estimated from the Bali outbreak [7] . For the calculation of R0 we considered rabid dogs up to the peak of the epidemic , while peak has been defined as the day during the epidemic with the highest number of newly rabies infectious animals . It has been demonstrated that both , clustering within the susceptible population and high number of repetitive contacts among individuals in the population–thus a non-random mixing situation–affects the dynamics of disease spread via depletion of susceptible population within a cluster and therefore also affects R0 [38] . A re-evaluation of R0 of a H1N1 infection revealed that it was overestimated during the early stage of the outbreak when only cases within a cluster has been considered rather than the population-wide epidemic [39] . In our model , random mixing of infectious and susceptible animals has to be rejected as rabies transmission depends on the distance between infectious and susceptible animals . However , neither fixed repetitive contacts ( as in network based models ) nor significant clusters within communities are included in the model . The only functionally relevant cluster structure is present in the region of the NPA with the five communities building distinct clusters . We respected the cluster structure in the R0 calculation by not only considering cases to the peak within the first cluster , i . e . the community of the index dog , but all cases occurring before to the epidemic peak within the total population at risk in the respective region . As illustrative examples we simulated the most often discussed and applied control strategies for dog rabies , namely vaccination and culling of dogs as a reactive action , as well as dog movement restrictions . According to the model , the only beneficial measure ( based on outbreak size ) is vaccination . This is consistent with a range of studies in regions where a rabies incursion has been observed: culling as a single control measure was unsuccessful [7 , 12 , 14 , 33] , whereas vaccination was demonstrated to be a successful strategy to control recent rabies invasions [7 , 10 , 32 , 40] . However , success of vaccination campaigns obviously depends on the vaccination coverage , as the example of a unsuccessful rabies control via low level vaccination coverage demonstrated in Flores [34] . Also , culling can lead to an eradication of timely detected outbreaks , as for example in region in Bhutan [9] , however might be impractical because of non-acceptance , depending on culture and religion of the community . In Australian Indigenous communities , culling of dogs is unlikely to be culturally acceptable . According to our model , movement bans as a single strategy does not seem to be sufficient to reduce rabies spread from one community to another nor within a community–at least for the simulated dog owner compliance ( 80–90% ) that was simulated here . Movement bans would culturally also be difficult to implement as travel with dogs between communities and regions is common . These results are similar for both regions , the NPA and Elcho Island . Within the NPA , rabies epidemics were able to sustain after incursion in each the five communities , as it does for the one community on Elcho Island , identifying the here considered regions and communities equally susceptible . Targeting surveillance should therefore be based on information revealed by risk assessment pathways exploring high risk regions for a rabies incursion . Further detailed simulations exploring combinations of response measures and their threshold values for effectiveness–including the effect of surveillance intensity [40]–is warranted as future research . The non-intervention strategy was implemented in the model and applied during the sensitivity analysis as a “baseline” to quantify the benefit of other control strategies , although it will most certainly never be observed in the field . Keeping in mind the assumption of a closed dog population with no influx of new dogs ( neither birth nor immigration ) , the high densities of dogs that roam freely in the modelled community and the assumption of an infectious period of up to 12 days combined with a fully naïve population , it is expected that the epidemic will eventually kill the modelled dog population , in the case of no intervention applied . Model outputs were sensitive to the assumed distance kernel for rabies transmission between dogs from different households . This highlights that the distance kernels should be empirically developed for the particular regions in which the model is to be applied . In our case , the kernel was estimated from roaming dog data collected within the actual study regions [18] . Model outputs ( size and duration of the epidemics ) were also sensitive to the assumed rabies incubation period and probability of rabies transmission given a bite . These parameters were derived from previously published field observations in Africa ( S1 Table ) , and are likely applicable to many situations , as is the vaccine efficacy parameter . For the probability of bite given a contact between dogs , no specific published data could be found . We assumed this parameter to be >50% due to the aggressiveness that rabies can cause [41] , but we also included a large range of uncertainty ( 60–80% ) . Critical parameter values ( as for contact rates and the population structure ) in this model are informed by data collected in the field . This guarantees the fit of the model to the intended environment of application , although limitations still do occur . Data informing the distance kernel were based on the roaming behaviour of healthy dogs [18] . A rabid dog might change its normal roaming behaviour , as for example reported for rabid racoons in New Jersey ( USA ) which moved over significantly larger distances than healthy racoons [42] . Apparently the effect of rabies on the roaming behaviour of domestic dogs has not been reported , but considering the observed changes in racoon behaviour , our model might underestimate the spread of rabies . When comparing contact distances observed in our study ( healthy dogs , mean distance = 103 meters ) that inform this model with the published spatial infection kernel of contacts between rabid dogs and resulting infection ( mean = 880 meters , [37] ) , rabid dogs might roam up to 8 . 5 times further than healthy dogs . This again indicates that in our model , although long distance movements have been included otherwise , transmission events might be underestimated . Studies on the nature of roaming , contact rates and biting rates of dogs in rabies-endemic studies–comparing these between infected and apparently uninfected communities in the same environment–are needed to address this gap in our knowledge and to more realistically parameterise models of rabies incursion . In the model , we have focused on the domestic dog population and ignored possible spread to wildlife ( in this region , wild dogs and dingoes ) because we were interested in the initial epidemic behaviour of a rabies incursion − and hence its impact on domestic dog health and by implication , human health . We assumed a closed dog population without introductions , births and natural deaths , again because our focus was on exploring initial disease response actions rather than the design of long-term rabies control programs . Although a critical issue for simulation model development , we have not yet validated the model using real outbreak data . Since Australia is historically rabies-free , a direct validation of the model in the region where it is intended for use is impossible , until an actual rabies incursion occurs . However , outbreak data from regions with similar dog population characteristics and recent rabies epidemics might be used , for example the island of Bali [12] or parts of Bhutan [9 , 10] . Validation of the model using data from a rabies outbreak in domestic dogs in Bhutan in 2008 is currently being planned . Global control of rabies is a declared goal of the World Health Organization and the Global Alliance for Rabies Control [43] . Within endemic , mostly developing countries , knowledge of effective control strategies is advanced and the challenge is to transfer this knowledge into successful actions [44] . Effectiveness of control strategies to prevent rabies from establishing in previously free regions–as is currently occurring in several areas with large free-roaming dog populations and limited public health services–has not received the same attention . The model described here is a tool to generate such information for remote northern Australia; and it is flexible enough to be adapted to other regions .
|
Rabies in domestic dog populations still causes >50 , 000 human deaths worldwide each year . While its eradication by vaccination of the reservoir population ( dogs and wildlife ) was successful in many parts of the world , it is still present in the developing world and continues to spread to new regions . Theoretical rabies models supporting control plans do exist for rabies endemic regions; however these models usually provide information for long-term programs . Here , we describe a novel rabies simulation model for application in rabies-free regions experiencing an incursion . The model simulates a rabies outbreak in the free-ranging dog population in remote indigenous communities in northern Australia . Vaccination , dog density reduction and dog confinement are implemented as control strategies . Model outputs suggest that the outbreak lasts for an average of 7 months and typically spreads through all communities of the region . Dog vaccination was found to be the most effective response strategy . The model produces plausible results and can be used to provide information for ad-hoc response planning before and shortly after rabies incursion .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Development of a Novel Rabies Simulation Model for Application in a Non-endemic Environment
|
Genome-wide gene expression profiles accumulate at an alarming rate , how to integrate these expression profiles generated by different laboratories to reverse engineer the cellular regulatory network has been a major challenge . To automatically infer gene regulatory pathways from genome-wide mRNA expression profiles before and after genetic perturbations , we introduced a new Bayesian network algorithm: Deletion Mutant Bayesian Network ( DM_BN ) . We applied DM_BN to the expression profiles of 544 yeast single or double deletion mutants of transcription factors , chromatin remodeling machinery components , protein kinases and phosphatases in S . cerevisiae . The network inferred by this method identified causal regulatory and non-causal concurrent interactions among these regulators ( genetically perturbed genes ) that are strongly supported by the experimental evidence , and generated many new testable hypotheses . Compared to networks reconstructed by routine similarity measures or by alternative Bayesian network algorithms , the network inferred by DM_BN excels in both precision and recall . To facilitate its application in other systems , we packaged the algorithm into a user-friendly analysis tool that can be downloaded at http://www . picb . ac . cn/hanlab/DM_BN . html .
The complex functions in eukaryotic cells are implemented through a highly organized regulatory network composed of concerted activities of many genes and gene products . Gene expression can be directly regulated by transcription factors ( TFs ) [1] , the states of chromatin structures [2] , [3] and small RNAs , and interactions among them [4]–[6] . In other words , the mRNA expression level of a gene is the output synthesized from the information of several input signals . Gene knockout is a classic approach to studying gene functions and the collection of yeast knockout strains has enabled systematic genome-wide functional analysis [7] . Transcriptional profiles of mutant strains have been used as molecular phenotypes for functional analysis and genetic epistasis analysis [8] , [9] . In addition , the expression profiles of single , double and triple deletion mutants of chromatin machinery components , protein kinases and phosphatases were used to analyze the functional overlaps among these proteins [3] , [10] . Dion et al . constructed 15 mutants of lysines 5 , 8 , 12 , and 16 to arginine in the histone H4 tail and characterized the resulting genome-wide gene expression changes [11] . Transcriptional regulatory networks in different cellular contexts have been constructed through the DNA microarray analysis of transcription factor deletion mutants and over expression strains in S . cerevisiae [1] , [12] by directly linking the genetically perturbed transcription factors ( TFs ) with the genes that change expression in response to the perturbations . As none of the regulators works alone , probably more important than constructing such regulator-target networks is to understand how the regulators cooperate to form regulatory pathways to specifically regulate a transcriptional program or biological processes [3] . Here we use the transcriptional profiles of deletion mutants as the molecular phenotypes of the mutants to determine how the regulators interact genetically or cooperate functionally with each other to modulate gene expression . We propose a Bayesian network ( BN ) approach to reverse engineer regulator networks from these gene expression profiles . The approach excels previous methods such as context-dependent regulation and correlation coefficient analysis [12]–[14] in that it can easily integrate different datasets and infer causalities in the regulatory program . Nodes in the network are the genes deleted in the mutants and the algorithm greedily searches over all possible Bayesian network structures for the one that best summarizes the relationships among the global differential expression change profiles upon deleting these genes . Thus by exploring the relationships among the global differential gene expression profiles for the deletion mutant genes , we can obtain valuable causal or non-causal relationships among these regulatory deletion-mutant genes through the inferred BN structure . Then , we used the above approach to analyze the global differential gene expression profiles of 544 single or double deletion mutants of transcription factors , chromatin machinery components , protein kinases and phosphatases in S . cerevisiae . The BN inferred identified with high precision and recall causal regulatory and non-causal interaction relationships among these regulators in different cellular contexts .
The deletion mutants of transcriptional regulators used in this study are nonessential genes in yeast under rich medium growth conditions , yeast extract peptone dextrose medium ( YPD ) or synthetic complete medium ( SC ) . We compiled expression profiles of sequence-specific DNA binding transcription factors ( STFs ) deletion strains grown in SC and YPD mediums [12] , [15] . We also collected the expression profiles for deletion mutants of protein kinases , phosphatases [10] and chromatin machinery components [3] ( See Methods for more detailed data descriptions ) . To confirm that regulators belonging to the same protein complex or regulatory pathway tend to share common targets [15] , we used Jaccard similarity index ( JI ) to examine the similarities between targets profiles of the perturbed regulators ( see Methods , Figure S1A , Table S1 ) and we observed that a STF is more likely to connect with another STF than with a general transcription regulator ( GTFs e . g . chromatin modifiers and remodelers ) whether the regulators are derived from the same data set or different data sets . Indeed , the percentage of known physical interactions or genetic interactions ( downloaded form SGD ) present among predicted gene pairs increases as the threshold of pair-wise Jaccard similarity index ( JI ) used in prediction increases ( Figure S1B ) , suggesting that the similarities of gene expression profile changes after genetic perturbation of transcriptional regulators can be used to infer relationships among these regulators . However , JI is only a crude measure that is subject to different cutoffs and cannot infer directionality or causality of regulatory relationships . In contrast , Bayesian network is a solid statistical inference method that can infer directions or causality of regulatory relationships and is more appropriate for this task . A Bayesian network [16] is a directed probabilistic graphical model which represents conditional independency relationships between variables . The BN learning approach has been extensively used in previous works to analyze gene expression and other high throughput data sets [14] , [17] , [18] . Suppose that the expression of a deletion mutant gene ( denoted by G ) is fully determined by its three intermediate regulator genes ( denoted by A , B , C ) , if the expression of genes A , B , C can be controlled precisely , we can find a specific expression configuration of A , B , C ( e . g . , A is up-regulated and B , C are down-regulated ) so that the expression of G is as small as possible just like being deleted . As such , we can anticipate that the global differential gene expression profile of deleting G versus the wild type strain can be well predicted from the global differential expression profiles of deleting B , deleting C and over-expressing A , respectively . Although the datasets contain only genetic deletion strains , no over-expression strains , the global differential expression profile of the profile of over-expressing A is often opposite to that of deleting A , we can thus well predict the differential gene expression pattern of deleting G from the three differential gene expression profiles of deleting genes A , B and C , respectively . In general , if one gene is combinatorially regulated by a set of other genes , usually we can approximate its deletion-mutant differential expression ‘phenotype’ fairly well by the deletion-mutant differential expression ‘phenotypes’ of its regulator genes . However , in deletion mutant experiments , it is typical that most genes have small expression changes in deletion mutant strains compared to their WT . For instance , 80% yeast genes have similar expressions to the WT strain in protein kinase or phosphatases deletions under the same growth condition [19] . Thus , the differential expression profiles of these regulators are sparse . The majority of ‘neutral’ gene expression changes ( represented by ‘0's ) in the differential expression profiles will artificially induce a high similarity between the deletion mutant genes ( regulators ) in classic BN learning methods . To this end , we developed a new Bayesian network structure-learning algorithm called Deletion Mutant BN ( DM_BN ) ( Figure 1 ) , which is specifically designed for reverse engineering regulatory networks of deletion mutant genes from differential gene expression profiles in the corresponding deletion mutant strains . Note that , the input of this algorithm is a matrix of discrete values: 1 , −1 , 0 , which denote the differential gene expression of the mutant strain versus the WT . Each column of the matrix records the differential gene expression profile for one deletion mutant gene . As described above , the training data for Bayesian network is skewed towards 0 , it is not viable to exploit classical Bayesian network learning approaches based on discrete data [20] . Indeed , in our extensive comparison of the proposed DM_BN algorithm with state-of-the-art BN learning algorithms with three other scoring metrics [20]–[23] , a well-known software package for BN learning [24] and two widely used non-Bayesian approaches to building regulatory networks [25] , [26] on the yeast deletion mutant datasets , the significantly improved network inference quality fully confirmed the advantage of the DM_BN algorithm ( See below ) . The main technical contribution of the DM_BN algorithm is to employ the kernel based approach to Bayesian network inference [27] and the introduction of a novel kernel for discrete data that is specifically designed for characterizing the deletion mutant data sets . Specifically , suppose and are two discrete variables which could take values 1 , −1 , 0 , the trivial kernel for discrete data in [27] is defined as: , i . e . , when ; and when . This is not viable for dealing with the deletion mutant data sets since the dominant value in such data is 0 , the trivial kernel for discrete data will induce a large similarity output ( 1 . 0 ) for all most all gene pairs which are neither up-regulated nor down-regulated . To prevent this biased effects , we modified the trivial kernel to the DM kernel below:The implication from the new DM kernel is clear: the differential expression changes of two genes in a deletion mutant experiment are considered similar ( with kernel output 1 . 0 ) if they are either up-regulated or down-regulated simultaneously . The similarity between genes that are not responsive to the deletion mutant experiment is abandoned ( with kernel output 0 . 0 ) . In this way , only the information of the co-regulation activity is fed into the Bayesian network-learning algorithm ( Methods ) . Another contribution of the DM_BN algorithm is the incorporation of the a priori knowledge from deletion mutant experiments into Bayesian network learning . For this purpose , we employ a network template to constrain the space of graph search in Bayesian network learning and to provide additional causal information in the learning and interpretation of Bayesian network structure . The basic idea of constructing the adjacency matrix of the network template ( template matrix for short ) is as follows: First , we start with an empty template matrix of zeros . Then , we define the list of target genes of a deletion mutant gene to be the genes whose mRNA levels either up- or down-regulated compared to the WT strain . If both deletion mutant genes A and B ( with indices , respectively ) are not in the target gene list of each other , the two genes do not seems to have a direct regulatory relationship , but they could cooperate to regulate other genes . So , if the target gene list of A and B overlap ( i . e . , at least one gene appear in both of the two target gene lists ) , the elements of the template matrix are set to 1 , which means that either one of the two edges might appear in the final BN . Finally , if gene B appears in the target gene list of A , but A is not in the target gene list of B , we set , which means that could appear in the final BN while the reversed edge is forbidden . In rare occasions , when both A , B appear in the target gene list of each other , we set , since we do not know which direction of the interaction represents the dominant regulatory effect while the other represents the secondary feedback effect ( Figure 1 ) . To identify potential causal interactions from Bayesian network structure , we have to determine whether the directionality of each edge in the network is reversible or not [28] . In this step , the template matrix again provides a priori causal information to guide the algorithm to disambiguate more edge directionalities . More details of the algorithm are presented in Methods . To quantitatively compare the performance of the DM_BN learning algorithm with other approaches to infer regulatory networks , we curate a database of ground-truths protein-protein interactions , regulatory interactions , genetic/epistatic interactions and protein complexes from the KEGG and SGD databases . Here , methods being compared include alternative Bayesian network learning algorithms ( the WinMine Toolkit [24] , the BDeu scoring approach [22] with optimized prior [23] and the BIC scoring approach [20] , [21] and non-Bayesian network approaches ( the ARACNE [25] software , the Disruption Network [26] and the Jaccard similarity index ( JI ) approach ) . Details of these algorithms and the strategies used in the testing are described in Methods and Supplemental Note 1 ( Text S1 ) . Basically , two key performance indicators are important for comparing the above algorithms: 1 ) Precision-recall curve , which quantifies the ability of an algorithm to correctly predict bona fide interactions between these regulators; 2 ) The precision of orientation , which measures the ability of an algorithm to predict correct directionality for each causal interaction . We first calculated the precision and recalls of all the predicted yeast regulator networks ( Methods ) . In this computation , directionality is not considered in matching a predicted edge and a known interaction in the database , which is partly because we only have very limited knowledge about the causality of these ground-truth interactions . By plotting the corresponding precision-recall points ( or point , if an algorithm predicts only one network ) for each algorithm , we found that DM_BN algorithm outperforms all the alternative network construction approaches in both precision and recall ( Figure 2 , Table S2 ) . In other words , regardless of causality , DM_BN algorithm has the highest precision of de novo network predictions over the whole range of recall rates . A close examination of the BN inferred by the DM_BN algorithm suggested it indeed recapitulated many interactions in protein complexes or pathways . Specifically , the BN structure visualized in Figure 3 with precision 0 . 4704 and recall rate 0 . 0323 ( Figure 2 ) includes both causal ( represented by directed edges ) and non-causal ( by undirected edges ) relationships among these regulators , which are known to take place in diverse biological processes to combinatorially regulate the expression of target genes . Moreover , we also computed the functional enrichment of these regulators based on their target genes ( Methods ) . The result suggests that regulators that tightly interconnected in the BN more significantly share common functions than other regulator pairs ( Figure 3 , Table S3 ) . The network learned by DM_BN algorithm further predicts how these regulators interact with each other in different cellular processes ( Figure 3 ) . For example , the predicted network module among subunits of the chromatin remodeling machinery complex ( Figure 3 , shown by purple nodes ) has a high precision of 0 . 85217 ( Table S4 ) ; the network module consisting of protein kinases Vps15 , Ark1 , Prk1 , Cdk8 , Cka2 and protein phosphatase Ptc1 , Ptc2 , Pph3 , Ptp3 is involved in three interrelated cell processes: cell wall organization or biogenesis , amino acid metabolism and carbohydrate metabolism , which is consistent with biological knowledge; and the predicted network suggests that Rpd3 complex , Sir complex and Ste11 mediated MAPK kinase cascades pathway cooperate with each other in mating process ( Figure 3 , Table S3 ) . We also observed that a STF is more likely to connect with another STF than with a GTF , which is similarly observed in the densely connected network inferred by the Jaccard index ( JI ) similarity measure ( Figure S1A ) . Our results are also consistent with the E-MAP results , which are quantitative genetic interactions between phosphorylation related genes in S . cerevisiae [19] . For instance , it is known that histone variant H2A . Z ( encoded by the Htz1 gene ) exchange with histone H2A in nucleosomes through the SWR1 complex [29] , [30] and that Htz1 displays positive genetic interactions with SWR1 ( +3 . 5 ) , Vps71 ( +3 . 9 ) and Vps72 ( +3 . 5 ) [19] . These interactions are all predicted by the network inferred by DM_BN ( Figure 3 ) . Indeed , the target sets of Htz1 , SWR1 , Vps71 and Vps72 deletion mutants have high similarity , with Jaccard indices ( Methods ) ( Figure 3 ( blue circle ) , Table S5 ) . Moreover , the functional enrichment of the predicted SWR1 complex target gene sets for vesicle organization ( Table S3 ) is consistent with the fact that SWR1 complex is required for vacuolar protein sorting [31] . More examples of the inferred pathway relationships are listed in Table S6 , S7 . Inferring correct directionalities for causal interactions or epistasis is an important aspect for regulatory network predictions . However , most non-Bayesian network algorithms are unable to do so . For example , the ARACNE [25] software and the Jaccard similarity index ( JI ) approach could only predict undirected interactions . Although the Disruption Network [26] could predict the direct causal relationships between deletion mutant genes and differentially expressed genes , such knowledge is derived from the deletion mutant experiments without performing causality inference . It is of special interest to see if an algorithm can make de novo predictions about causality among deletion mutant genes from the similarity of their genome-wide differential expression profiles . In principle , Bayesian network learning algorithms hold this promise and thus we compare the performance of the four BN learning algorithms ( DM_BN , Winmine toolkit [24] , the BDeu [22] and BIC scoring approaches [20] , [21] ) in predicting causal relationships . Since the ground-truth causal relationships derived from existing databases for the 378 regulator genes is very limited , and also because we do not know the exact cellular contexts in which those causal relationships hold true , to quantify the performance in predicting causal relationships , we calculated the recall and the precision of all these network inference approaches except BDeu using the four MAPK ( mitogen-activated protein kinase ) cascades where clear causal relationships are well described among these kinases . The exclusion of BDeu here is simply because it does not have a tunable parameter to generate a relative sparse network that is comparable to the size of networks generated by the other three approaches . However , using a different evaluation approach , BDeu's causality prediction apparently does not perform as well as DM_BN and BIC ( see below ) . Yeast contains at least four MAPK ( mitogen-activated protein kinase ) cascades that convert extracellular stimuli into intracellular signals during a variety of cellular processes , such as mating , cell wall remodeling and high osmolarity adaption [32] . We found that when all tools predicted roughly the same number of edges , the DM_BN algorithm with prior information can predict more interactions with correct orientations than other tools among kinases involved in the same signal transduction pathway ( Table S8 , S9 , S10 , S11 ) . To test whether the correct inference of edge direction is solely the result of applying a template , we examined the directionality of edges in BNs inferred by DM_BN without any template . At various parameters , 73 . 3–89 . 2% of the edges have the same direction as the regulator-DEG relationships identified in the deletion mutants experiments . These proportions are significantly higher than that expected by chance ( random coin tossing p = 0 . 5 , Binominal test p = 0 . 0625∼5 . 42e-07 , Table 1 ) . This indicates that the correct inference of edge directions by DM_BN is largely not attributed to using the template . However , the BN inferred by DM_BN with the template , did correct a small number edges incorrectly predicted when not using the template ( 1/34∼4/15 edges , Table 1 ) . This is because the network template not only corrects edge orientation errors inconsistent , but also improves the global causal structure in the BN through the cascading interactions between edges . Therefore , a template is included in the actual implementation of the DM_BN algorithm as the default setting . Using a similar approach , we also compared with other BN inference algorithms , the performance of DM_BN in de novo predicting causal relationships without using the a priori information encoded by the template . The DM_BN algorithm and the BIC scoring approach [20] , [21] generally predict non-compelled directed edges remarkably more precise than the BDeu scoring method [22] or the WinMine toolkit [24] ( Supplemental Note 2 and 3 in Text S1 , Figure S2 ) . In particular , the causal relationships inferred by the DM_BN algorithm ( with the network template ) correctly recapitulated the linear cascade structure for regulators in the HOG signaling pathway involved in the osmotic stress response ( Figure 4A ) . For instance , Ste11 MAPK kinase ( MAPKKK ) phosphorylates Pbs2 MAPK kinase ( MAPKK ) . Then , the activated Pbs2 phosphorylates Hog1 in the MAPK kinase cascade pathway for osmostress adaptation [33] . In the mating process , DM_BN not only accurately grouped the SIR complex and the Ste11 mediated MAPK cascade pathways , but also correctly predicted the connectivity among components of the complex or the pathway ( Figure 4B ) . The results correctly recovered the role of Ste7 and Ste11 protein kinases in two different MAPK Fus3 and Kss1 cascade pathways that controls mating , respectively [33] , [34] . The inferred causal relationships or non-causal interactions between these gene expression regulators not only confirmed known relationships , such as physical interactions and genetic epistasis relationships among these regulators , but also predicted many novel relationships that could be important in gene regulation . For example , the DM_BN algorithm not only correctly predicted the connection between components in the SIR complex or in the MAPK pathway , but also predicted the dense connection between the SIR complex and Ste11-mediated MAPK cascades ( Figure 4B , Table S3 and Table S12 ) . Clustering of the expression profiles of the genes in these network modules shows that the genes up-regulated in the deletion mutants of Sir2 , Sir3 and Sir4 ( Figure 5A , right panel ) are all within 10 kb to their nearest telomere . Meanwhile , the predicted functions of the genes down-regulated by all deletion mutants in Figure 5A are enriched for mating process ( see also Table S3 ) . All these findings are consistent with the knowledge that SIR complex plays roles in silencing at HML , HMR loci which carry unexpressed copies of mating-type genes and telomeres [35] and that SIR complex is comprised of two structural proteins Sir3 and Sir4 , deletion of which will cause reduced mating rate at different levels [36] . The mRNA expression levels of Fus3 and Fus1 are very low in the deletion mutants of Sir2 , Sir3 and Sir4 in all the three data sets ( Figure 5A , left panel ) . However , the mRNA levels of other genes in the model are not changed compared to WT ( except the expression levels of the deletion mutant genes themselves ) ( Figure 5A , left panel ) . From the causal , non-causal relationships predicted by DM_BN ( Figure 4B ) and the expression profiles of deletion mutants experiments ( Figure 5A ) , we can infer a novel model implying that the Ste11 mediated MAPK cascades pathway may have overlapping functions with the SIR complex ( Figure 5B ) . Thus , SIR complex could indirectly influence the mRNA expression of kinase Fus3 , which is involved in the MAPK cascades pathway in mating process . Although a STF is more likely to have a similar gene expression pattern with another STF than with a GTF generally ( Figure 3A ) , the network suggests that STFs Cst6 , Sfp1 , Bas1 , Mac1 , Gsm1 , Ixr1 , haa1 , Ume6 and Cad1 connect densely to GTFs ( some subunits of SIR complex , SWI/SNF complex and SAGA complex ) ( Figure 5C ) . Clusters ( Figure S3 ) of the expression profiles of these regulators revealed high similarities between the target profiles of the STFs , Sfp1 and Cst6 , and the GTFs SWI/SNF complex and SAGA complex . Another example is the high similarity between the targets profiles of STFs: Ixr1 , Cad1 , Bas1 and Stp4; and GTFs: SIR complex , SAGA complex . Although no physical interactions or binding relationships between them have been reported in the literature , SAGA subunit Spt3 has been reported to have negative genetic interactions with Stp4 and Ixr1 [10] . These novel predictions by the DM_BN algorithm may serve as blueprints for further experimental explorations .
Uncovering complex regulatory networks is an important and challenging task [1] , [12] , [19] , [37] . Here , we introduced a new Bayesian network inference algorithm “DM_BN” , specifically designed to infer regulatory networks from gene expression profiles generated by gene perturbations , such as gene deletions . DM_BN can work with both small and large datasets and infer causal and non-causal relationships among the perturbed genes . To address the sparsity of gene expression changes in the perturbation experiments , we developed a kernel-based BN learning algorithm DM_BN , which is appropriate for modeling such gene expression data sets . Comparing with known biological interactions , both the recall and the precision of the network inferred by the proposed DM_BN algorithm are significantly higher than that inferred by WinMine and by the Jaccard Index ( JI ) similarity measure ( Figure 2 ) . The DM_BN network model not only successfully recapitulated known interactions among the yeast transcriptional regulators , but also predicted many novel interactions among these regulators and regulatory protein complexes , offering new insights into the yeast transcriptional regulatory network . Our results show that the improved performance of the DM_BN algorithm can be mainly ascribed to the new kernel . Since an edge between two regulator genes is allowed in the network template if they share at least one target gene , the template matrix actually allowed all the possible interactions between these genes , hence has a very little predictive value by itself . However , the DM_BN algorithm still benefits from using the network template in two aspects . First , by eliminating all impossible edges , the template effectively reduced the search space to speed up the DM_BN algorithm . Second , by encoding the a priori regulator-target causal knowledge in deletion mutant experiments , the network template not only corrects edge orientations that are inconsistent with such information ( Table 1 ) , but also improves the global causal structure predicted by DM_BN through edge-edge interactions , as we demonstrated in the inference of MAPK pathways ( Table S11 ) . Although the DM_BN approach has achieved big success in inferring yeast regulatory network from perturbation-based gene expression data sets , there are still a few limitations to its applications . For example , the mRNA expression levels of target genes are not fully representative of the activities and interactions of the regulators in modulating gene expression . This is because post-transcriptional changes and the regulators' context-specific transient activity were not measured in the experiments . Due to the intrinsic limitation of mRNA expression data , our method failed to identify certain relationships among the regulators under certain conditions , especially when the activity of the regulators is not screened in the microarray experiments . Nevertheless , these problems are not the fault of the proposed BN inference algorithm but rather inherent limitations of current experimental systems , which are expected to overcome by introducing other types of high-throughput datasets . In this sense , the application of the DM_BN algorithm is not limited to microarray expression profiles of genetic perturbations , it can actually be extended to work on many kinds of high-throughput data , such as epigenomic , transciptomic , proteomic data sets , and even quantitative phenotype data .
All the gene expression profiles are downloaded from the Gene Expression Omnibus ( GEO ) database , including 269 transcription factors knockout strains grown in yeast extract peptone dextrose medium ( YPD ) [12] , 150 deletion mutants of protein kinases and phosphatases [10] , 165 mutants of chromatin machinery components [3] and 52 sequence-specific DNA binding transcription factors ( STFs ) deletion strains grown in synthetic complete medium ( SC ) [15] . Altogether the four data sets above contain gene expression profiles of 544 yeast deletion mutants . The series accession numbers of these data sets are GSE4654 [12] , GSE25644 [10] , GSE25909 [3] and GSE2324 [15] . The detailed DNA microarray normalization and statistical analysis procedures see described in Supplemental Methods ( Text S1 ) . After processing , the gene expression changes are represented by discrete values: 1 ( significant up-regulation ) , −1 ( significant down-regulation ) and 0 ( no significant expression change ) . We employ the kernel-based Bayesian network learning algorithm [27] with three modifications . First , we use the ‘DM kernel’ instead of the trivial kernel to handle the yeast deletion mutant datasets . Second , we use a template matrix to constrain the space of all possible Bayesian network structures . Details of the ‘DM kernel’ and the construction of the template matrix are described in the Results and will not be repeated here . Finally , we modified the BIC scoring function by increasing the weight of the complexity term for penalizing the Kernel Generalized Variance [38] measure . This is necessary for removing biological noise and increasing the precision and sparsity of the finally obtained network structure . Formally , the Bayesian network scoring function is modified as follows ( cf . eqn . 4 in ref [27] for details ) :Here , is the BIC score for node and its parents , and the overall score for a full Bayesian network is: . and are the Kernel Generalized Variance [38] for node sets and . is the multiplicative weight that we impose on the second term of the scoring function . With the DM kernel inside the KGV measure , the template matrix as a structural constraint and the modified scoring function , we can search for the Bayesian network structure that optimally fit the yeast deletion mutant datasets . Specifically , in each step , we consider 1 ) adding an edge that is consistent with the template; 2 ) deleting an edge from the current BN structure; 3 ) reversing the direction of an edge that will not violate the causal constraints embodied the template . In accordance with previous studies , we use the greedy ascent TABU search method [39] to find the ideal Bayesian network structure . Here , ‘TABU’ denotes a Meta searching strategy that prohibits the algorithm from ‘undoing’ a recent operation . It helps the search procedure from being getting stuck in the local optima regions [39] . Finally , we adopt an efficient dynamic graph acyclicity checking method [40] in the Bayesian network structure search , since the most computational intensive task in this study involves inferring a Bayesian network of up to ∼400 nodes , using the conventional static graph acyclicity checking method would be fairly slow . Interpreting the causalities in the Bayesian network structure is not a straight forward task . This is because there are equivalence classes of Bayesian network structures . All BNs in the equivalence class are semantically equivalent . They share the same set of skeletons ( edge connections regardless of arrows ) , but differ in the directionalities of some edges [41] . As such , there are two types of edges in a Bayesian network: compelled edges , whose directionalities are fixed among BNs in the equivalence class and non-compelled edges , whose directionalities are not consistent in the equivalence class [42] . The authors also proposed an efficient algorithm to dissect a Bayesian network into compelled ( directed ) edges and non-compelled ( undirected ) edges ( the results are collectively represented as a partially directed acyclic graph , a . k . a . , PDAG ) [42] . However , the approach is not well suited for our study because the template matrix contains much a priori causal information , which was not used by Chickering's algorithm [42] . To overcome this problem , we employ Meek's rules [28] to convert the Bayesian network structure into a PDAG . The merit of this algorithm is that prior causal knowledge could be fully exploited in making causal interpretations of the Bayesian network structure . As such , we extract asymmetries in the template matrix and impose those constraints in the causal interpretation algorithm [28] . In this way , the causal information conveyed in deletion mutant experiments is used maximally . Jaccard's similarity index [43] quantifies the similarity between two sets of elements . In this work , we use Jaccard index to evaluate the similarity between the deletion mutant expression profiles of two regulators , which is calculated as follows:Here: : No . of common targets for regulator1 and regulator2 . : No . of genes whose expression significantly changed in the deletion mutant of regulator 1 . : No . of genes whose expression significantly changed in the deletion mutant of regulator 2 . We use known relationships between the 165 mutants in 30 chromatin modification complexes [3] , ground-truths protein-protein interactions , regulatory interactions and genetic/epistatic interactions curated from the KEGG and SGD databases to evaluate the precision/recall for the predicted networks . For this purpose , we calculated the recall and the precision of a network using the formula below:: the number of edges that are correctly predicted in the network ( true positives , i . e . , predicted edges that are also consistent with known protein-protein interactions ) . : the total number of known pair-wise interactions between nodes in the network ( e . g . , between subunits in known protein complexes ) . : the number of edges in a predicted network . In rare cases , a deletion-mutant regulator gene appears in more than one data set . Removing any copy of that gene will typically cause many of its interactors in that data set undetected . So , multiple copies of all these overlapping genes are retained in our analysis . In this scenario , no matter how many edges between two same genes from two different datasets are predicted in the network , we only count this interaction once . Moreover , in the analysis of causal relationships , if the directionalities of these edges are consistent , we retain the directionality of that interaction; otherwise , we treat this interaction as an undirected ( non-compelled ) edge . The significance of the functional enrichment of a gene list is computed by performing the hypergeometric test . In this work , all the GO annotation , phenotype and pathway data sets were downloaded from the SGD , KEGG database . The P-value is calculated as follows:: The total number of genes with significant expression change in any deletion mutant experiment . : Number of genes in one particular functional category . : Number of genes with significant expression change in the current deletion mutant experiment . : Number of overlapped genes between the n genes and the S genes . To systematically compare the performance of the DM_BN algorithm with existing network-inference methods , we tested the ARACNE [25] software , the Disruption Network approach [26] and the Jaccard Index ( JI ) approach , which are not based on Bayesian networks . We also tested the WinMine Toolkit [24] and Bayesian network learning with two other scoring metrics ( BIC [20] , [21] and BDeu [22] ) ( Detailed description of each method is listed in Supplemental Note 1 in Text S1 ) . To ensure the comprehensiveness of the evaluation , for each method and whenever possible , we use a wide range of parameter settings to infer a number of networks to best reveal the tradeoffs between precision and recall . Specifically , for the DM_BN algorithm , we run it with a wide range of values for the parameter ( See Methods “Inference of the optimal Bayesian network structure” for details ) to infer a set of BNs with different number of edges . Similarly , for the WinMine toolkit [24] , we run it at different values of the kappa parameter to infer BNs; for ARACNE [25] , various mutual information thresholds are used to infer a set of regulatory networks; for the Jaccard index ( JI ) approach , a wide range of the JI similarity cutoffs are used for network construction and for the BIC score approach , different weights ( like the parameter in DM_BN ) are multiplied to the penalty term ( , see Eqn . 42 in [20] ) in the Bayesian information criterion to infer BNs . Note that we could only generate a single best network for the BDeu scoring approach [22] using the optimal ESS value [23] since there is no way to tune the precision-recall tradeoffs . Similarly , only a single network can be inferred for the disruption network approach [26] . This is because to determine whether a gene is significantly differentially expressed in a deletion mutant strain , we employed the default statistical tests and significance thresholds used in the original experimental study ( See “Statistical analysis of expression profiles” in Supplemental Methods ) . It is not clear how to adjust the thresholds synchronously for the four data sets .
|
The complex functions of a living cell are carried out through hierarchically organized regulatory pathways composed of complex interactions between regulators themselves and between regulators and their targets . Here we developed a Bayesian network inference algorithm , Deletion Mutant Bayesian Network ( DM_BN ) to reverse engineer the yeast regulatory network based on the hypothesis that components of the same protein complexes or the same regulatory pathways share common target genes . We used this approach to analyze expression profiles of 544 single or double deletion mutants of transcription factors , chromatin remodeling machinery components , protein kinases and phosphatases in S . cerevisiae . The Bayesian network inferred by this method identified causal regulatory relationships and non-causal concurrent interactions among these regulators in different cellular processes , strongly supported by the experimental evidence and generated many testable hypotheses . Compared to networks reconstructed by routine similarity measures or by alternative Bayesian network algorithms , the network inferred by DM_BN excels in both precision and recall . To facilitate its application in other systems , we packaged the algorithm into a user-friendly analysis tool that can be downloaded at http://www . picb . ac . cn/hanlab/DM_BN . html .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2013
|
Functional Dissection of Regulatory Models Using Gene Expression Data of Deletion Mutants
|
African Americans have a disproportionate risk for developing nephropathy . This disparity has been attributed to coding variants ( G1 and G2 ) in apolipoprotein L1 ( APOL1 ) ; however , there is little functional evidence supporting the role of this protein in renal function . Here , we combined genetics and in vivo modeling to examine the role of apol1 in glomerular development and pronephric filtration and to test the pathogenic potential of APOL1 G1 and G2 . Translational suppression or CRISPR/Cas9 genome editing of apol1 in zebrafish embryos results in podocyte loss and glomerular filtration defects . Complementation of apol1 morphants with wild-type human APOL1 mRNA rescues these defects . However , the APOL1 G1 risk allele does not ameliorate defects caused by apol1 suppression and the pathogenicity is conferred by the cis effect of both individual variants of the G1 risk haplotype ( I384M/S342G ) . In vivo complementation studies of the G2 risk allele also indicate that the variant is deleterious to protein function . Moreover , APOL1 G2 , but not G1 , expression alone promotes developmental kidney defects , suggesting a possible dominant-negative effect of the altered protein . In sickle cell disease ( SCD ) patients , we reported previously a genetic interaction between APOL1 and MYH9 . Testing this interaction in vivo by co-suppressing both transcripts yielded no additive effects . However , upon genetic or chemical induction of anemia , we observed a significantly exacerbated nephropathy phenotype . Furthermore , concordant with the genetic interaction observed in SCD patients , APOL1 G2 reduces myh9 expression in vivo , suggesting a possible interaction between the altered APOL1 and myh9 . Our data indicate a critical role for APOL1 in renal function that is compromised by nephropathy-risk encoding variants . Moreover , our interaction studies indicate that the MYH9 locus is also relevant to the phenotype in a stressed microenvironment and suggest that consideration of the context-dependent functions of both proteins will be required to develop therapeutic paradigms .
Chronic kidney disease ( CKD ) is an acute public health problem world-wide . Within the United States alone , it affects up to 14% of the adult population and is associated with both high costs and poor clinical outcomes[1] . Compared with European Americans , African Americans have a disproportionate risk for several forms of CKD , including human immunodeficiency virus ( HIV ) -associated nephropathy , focal segmental glomerulosclerosis ( FSGS ) , hypertension-attributed CKD , and sickle cell disease nephropathy ( SCDN ) , all of which contribute to a four-fold increased risk of the most severe stage of CKD , end-stage renal disease ( ESRD ) [1–5] . A genomic region on chromosome 22q12 likely accounts for almost all of this racial disparity . This region contains two genes , non-muscle myosin heavy chain IIA ( MYH9; Entrez , 4627 ) and apolipoprotein L1 ( APOL1; Entrez , 8542 ) , both of which have been associated with increased risk among African American patients with nondiabetic nephropathy[5–11] . Initial admixture mapping and subsequent fine mapping studies focused on MYH9[8 , 9 , 11] . However , due to the inability to identify variants in MYH9 that alter protein sequence , the major source of genetic association has been attributed to APOL1 , located 14 kb downstream of MYH9[6] . Two APOL1 alleles , G1 ( encoding p . S342G and p . I384M in cis ) and G2 ( encoding p . N388del:Y389del ) , comprise one of the strongest genetic signals ever reported in complex human disease ( odds ratios ranging from 10 . 5 to 16 . 9 ) [6 , 7] . Additionally , these alleles alter the protein to confer resistance to Trypanosoma brucei rhodesiense , offering a potential evolutionary explanation for the increased occurrence observed among individuals of African ancestry[6] . Despite these genetic findings and the association of this locus with increased risk of multiple forms of CKD , there is a dearth of functional data to inform directly whether MYH9 or APOL1 is the driver of this genetic association . In mice , homozygous Myh9 knockouts die at an early embryonic stage[12] , and heterozygotes appear viable without any detected abnormalities[13] . However , subsequent studies have demonstrated that knock-in mutants display renal glomerulosclerosis , while podocyte-specific deletion of Myh9 may predispose mice to glomerulopathy[14–16] . In zebrafish , myh9 is required for the normal development of the glomerulus; morpholino ( MO ) -induced myh9 suppression results in non-uniform podocyte foot processes and glomerular basement membrane thickening[17] . In contrast , the possible relevance of APOL1 to CKD is derived primarily from in vitro work: cellular localization studies of APOL1 in nondiabetic kidney disease patient biopsies suggest an implication in arteriopathy[18 , 19] , while overexpression of APOL1 and its risk alleles enhance podocyte necrosis in vitro [20] . Nephropathy is a major contributor to early mortality in patients with sickle cell disease ( SCD ) [21 , 22] . SCDN is a clinically well-characterized pathology that includes glomerular hypertrophy , hyposthenuria , tubular dysfunction , proteinuria , and overall progressive renal failure[23] . We reported previously an association of both MYH9 and APOL1 variants as independent risk factors for proteinuria in a SCD study population[5] . Additionally , when glomerular filtration rate ( GFR ) in SCD patients was modeled as a function of the previously reported MYH9 risk haplotype and the APOL1 recessive model , we observed a significant interaction between the two genes , suggesting that APOL1 and MYH9 may act together to induce SCDN[5] . However , as with other forms of CKD , well-characterized in vivo model systems are needed to understand both the individual effects of APOL1 relevant to disease , and also the potential interaction of APOL1 with MYH9 in the context of anemic stress as observed in SCD . Here , we used zebrafish as an in vivo model to study the consequences of gene perturbation and potential synergistic effects of APOL1 and MYH9 in kidney disease . Although the zebrafish pronephros is a simplified kidney , the structure and function of the larval glomerulus is similar to that of humans and represents a tractable model in which to study apol1 ( RefSeq: NM_001030138 ) and myh9 ( RefSeq: NM_001098177 . 2 ) [24 , 25] . In this report , we provide insight into the role of apol1 in glomerular development and pronephric filtration in zebrafish embryos , as well as the effects of APOL1 G1 and G2 allelic expression . Moreover , we provide functional evidence for an interaction between myh9 and apol1 under anemic stress conditions . Overall , these data implicate both MYH9 and APOL1 as significant biological contributors to non-diabetic nephropathy and intimate context-dependent roles in disease pathology .
The apolipoprotein L family of genes evolved rapidly in humans and some non-human primates[26 , 27] . However , using BLAST and reciprocal BLAST searches against the D . rerio and H . sapiens genomes , we identified a single D . rerio locus encoding a protein of unknown function ( chr2:37 , 674 , 122–37 , 676 , 731 Zv9; NCBI Ref: NP_001025309 . 1; 38% identity , 46% similarity on the amino acid level ) as a possible unique functional ancestral ortholog to the human apolipoprotein L family ( Fig 1A–1D ) . To explore the function of this transcript in developing zebrafish , we first asked whether the candidate apol1 ortholog is expressed in a temporal manner amenable to transient assays of renal development and function . RT-PCR analysis of cDNA generated from wild-type ( WT ) whole-larval total RNA collected at three days post-fertilization ( dpf ) and 5 dpf showed expression at time points corresponding to the formation of the pronephros . Additionally , we detected apol1 expression in flow-sorted podocyte fractions harvested from glomeruli of pod::NTR-mCherry adult zebrafish ( Fig 1E ) [28] . To test the effects of apol1 suppression , we designed a translation-blocking morpholino ( MO; Gene Tools , LLC ) targeting the candidate zebrafish apol1 locus ( apol1-MO ) and we injected increasing doses into embryos at the one to four cell stage ( n = 49–65 embryos/injection; repeated three times ) . Masked scoring for morphological defects at 5 dpf revealed a dose-dependent increase of the percent of larvae displaying pericardial and yolk sac edema , a phenotype that has been implicated previously in glomerular filtration defects[24 , 30] ( Fig 2A–2C ) . Co-injection of WT APOL1 human mRNA ( GenBank Accession: BC112943 . 1; 100 pg/nl ) rescued significantly the edema caused by apol1 suppression ( p<0 . 0001; Fig 2D ) , arguing not only that the phenotype was unlikely to be a non-specific toxic effect of the MO , but also that the zebrafish locus we targeted is the ortholog of the human transcript . Importantly , co-injection of human mRNA encoding other human apolipoprotein L members ( APOL2 , APOL3 , APOL4 , APOL5 , and APOL6 ) with apol1 MO did not rescue the edema formation of apol1 morphants ( S1 Fig ) . Additionally , we observed a significant decrease in endogenous APOL1 protein expression in apol1-MO injected zebrafish embryos ( p = 0 . 026 ) , which is restored to normal levels upon co-injection with wild-type human APOL1 mRNA ( S2 Fig ) . Furthermore , as an additional test of the specificity of apol1 perturbation to edema formation , we induced microdeletions in exon 3 of apol1 using the CRISPR/Cas9 system[31 , 32] ( Fig 3A–3C ) . Injection of guide RNA and CAS9 protein into one-cell stage embryos reproduced the edema phenotype ( scored in founders , F0 ) seen in apol1 morphants ( n = 26–38 embryos/injection , repeated three times; p<0 . 001; Fig 3D ) . To test whether the generalized edema phenotype was relevant to nephropathy , we assessed the integrity of the glomerular filtration barrier in apol1 morphants and F0 mutants as described[30] . First , we injected 70-kDa FITC-labeled dextran into the cardiac venous sinus of larvae at 48 hours post-fertilization ( hpf ) . After injection , the eye vasculature was imaged at 24 and 48 hours post-injection ( hpi; Fig 2E and 2F ) . We quantified the average fluorescence intensity ( ImageJ ) and calculated changes in intensity at 48 hpi relative to the 24 hpi measurements . apol1 morphant larvae display a significant reduction in circulating 70-kDa dextran compared to controls ( n = 26; p = 4 . 44x10-4; MO vs . control; Fig 2E and 2F ) , consistent with the occurrence of proteinuria . Importantly , this phenotype was also reproduced in apol1 CRISPR/Cas9 larvae ( Fig 3E ) . Upon co-injection of WT APOL1 human mRNA , the increased dextran clearance in apol1-MO larvae was rescued significantly and fluorescence intensity returned to levels indistinguishable from controls ( n = 28; p = 7 . 75x10-4 , MO vs . MO + mRNA; Fig 2E and 2F ) . Next , we evaluated the cellular organization and patterning of the developing glomerulus in the context of apol1 suppression . We performed transmission electron microscopy ( TEM ) of ultrathin sections of zebrafish larvae at 5 dpf in WT and apol1 morphants and mutants , with myh9 morphants as a positive phenotypic control . In agreement with previous studies[17] , myh9 morphant larvae exhibit focal bulges and glomerular basement membrane ( GBM ) thickening in comparison to controls , as well as the presence of microvillus protrusions , a defining characteristic of proteinuria ( S3 and S4 Figs ) . Notably , apol1-MO injected larvae display a similar glomerular ultrastructure compared with myh9 morphants . Naked patches of GBM are apparent throughout the glomerulus , indicative of extensive podocyte effacement ( Figs 2G , 2H , and S4 ) . However , we did not observe GBM thickening as evident in myh9-MO injected larvae ( S3 Fig ) . In areas in which we did observe foot process formation , podocyte protrusions were irregular and inhibited slit diaphragm development ( Figs 2G , 2H , and S4 ) . We also noted the formation of microvillus protrusion in the urinary space of apol1 morphants . Similarly , apol1-CRISPR/CAS9 injected embryos display an aberrant glomerular ultrastructure , as evident by podocyte foot process effacement ( Fig 3F ) . Co-injection of orthologous WT human mRNA in apol1 morphants rescued these glomerular ultrastructure defects ( Fig 2I ) . Together , these data represent compelling in vivo evidence implicating APOL1 in renal function . Initial reports associating APOL1 variants with kidney disease in African Americans identified two independent sequence variants , termed G1 and G2 , which reside in a 10-kb region in the last exon of the gene[5–7 , 10] . The G1 allele consists of two nonsynonymous coding variants in perfect LD , rs73885319 and rs60910145 , while the G2 variant consists of a six base pair deletion that removes amino acids N388 and Y389 ( ~21% and ~13% allele frequency in African Americans , G1 and G2 respectively; Fig 1D ) . Therefore , we evaluated the ability of each of the G1 and G2 alleles to rescue apol1-MO injected zebrafish larvae . APOL1 G1 ( I384M/S342G ) and G2 allelic constructs were generated from a WT APOL1 human cDNA clone , transcribed , and co-injected with apol1-MO in zebrafish embryos ( 100pg/nl ) . Importantly , each APOL1 allelic construct produces a stable protein detectable by immunoblotting when co-injected with apol1-MO ( S2 Fig ) . apol1 morphants co-injected with either APOL1 G1 ( I384M/S342G ) or G2 human mRNA did not display significant rescue of edema formation in developing embryos compared to apol1-MO injected embryos alone ( Fig 4A and 4B ) . In addition , we also co-injected each individual G1 variant ( I384M and S342G ) into apol1 morphant embryos . APOL1 message encoding either p . I384M or p . S342G were individually able to rescue significantly the edema caused by apol1 suppression ( Fig 4C and 4D ) suggesting that the cis effect of both variants in the same haplotype is required to confer pathogenicity . When APOL1 G2 mRNA was injected alone , a significant number of embryos developed edema in comparison to sham-injected controls ( n = 52–63 embryos/injection; repeated three times; p = 0 . 012; Fig 4B ) ; no edema was observed with injection of 100pg APOL1 G1 mRNA alone ( Fig 4A ) . Additionally , dextran clearance assays demonstrated that neither APOL1 G1 or G2 mRNA were able to rescue glomerular filtration defects caused by apol1 suppression , while APOL1 G2 mRNA injected alone caused significant filtration defects compared to controls ( n = 12–21; p = 0 . 003 , Control vs . G2 mRNA; Fig 4E and 4F ) . Finally , when we injected embryos with APOL1 G2 titrated with increasing concentrations of APOL1 WT mRNA , we observed a significant reduction of edema formation in developing embryos ( Fig 4G ) suggesting that this allele is conferring a dominant negative effect on protein function . We also examined the glomerular ultrastructure of apol1 morphants co-injected with either APOL1 G1 or G2 human mRNA using TEM . However , we did not observe any noticeable improvement in glomerular ultrastructure abnormalities at 5 dpf ( S5 Fig ) . In concurrence with our observations of gross morphological defects , embryos injected with G2 mRNA alone also display glomerular aberrations and microvillus protrusions ( Fig 4H ) similar to myh9 and apol1 morphants ( Figs 2H and S4 ) ; no abnormalities were seen in larvae injected with G1 mRNA alone ( Fig 4I ) . These data provide direct evidence for a functional consequence of the human APOL1 G1 and G2 risk alleles , and suggest that they confer loss-of-function and dominant negative effects , respectively . Although recent studies have provided statistical evidence implicating APOL1 variation in nondiabetic nephropathies[7 , 33 , 34] , MYH9 risk variants are still associated with chronic kidney disease ( CKD ) in non-African American populations[35] and in sickle cell disease nephropathy[5] . As such , our group and others have hypothesized that these genes may be co-regulated to induce nephropathy risk; in fact , when we modeled glomerular filtration rate in sickle cell patients as a function of the previously reported MYH9 risk haplotype and an APOL1 recessive model , we observed a significant interaction between the two genes[5] . Therefore , we tested for functional interaction effects between apol1 and myh9 in zebrafish , an experimentally tractable model for investigating additive and synergistic effects[36–40] . First , we co-injected both apol1-MO and myh9-MO into embryos and we scored for gross morphological defects at 5 dpf . Under this co-suppression model , we observed no significant differences in edema formation when compared to batches injected with either MO alone ( Fig 5A ) , even when individual MO concentrations were reduced to subeffective doses ( Fig 5B ) . Next , we tested the possibility that suppression of either apol1 or myh9 in zebrafish could be rescued significantly by the co-injection of the reciprocal human mRNA . myh9-MO was co-injected with human APOL1 WT mRNA ( 100pg/nl ) and apol1-MO was co-injected with human MYH9 WT mRNA ( 100pg/nl ) . However , we were unable to rescue the suppression phenotypes of either apol1 or myh9 with the human mRNA of the reciprocal gene ( S6 Fig ) . Our hypothesis for an interaction between APOL1 and MYH9 was based on data derived from SCD patients . Thus , we posited that myh9 and apol1 may only interact under additional biologic stress , such as anemia or hemolysis . Accumulating evidence suggests that both anemia and hemolysis , which are key features of SCD pathophysiology , impact renal function; in particular , hemolysis appears to be associated with both microalbuminuria and hyperfiltration[41 , 42] . While a zebrafish model of SCD does not exist currently , suppression of ATPase inhibitory factor 1 ( atpif1α ) , a mitochondrial protein , produces profound anemia in zebrafish by interfering with heme synthesis through decreased catalytic efficiency of ferrochelatase[43] . The resultant effect of low hemoglobin and hematocrit stresses the kidney because of the organ’s high oxygen consumption . Consistent with the original report[43] , we observed a dose-dependent reduction in hemoglobin with increasing concentrations of the atpif1a MO ( atpif1α-MO ) , as measured by o-dianisidine staining of whole MO-injected larvae at 4 dpf . Strikingly , we found a significantly more severe nephropathy phenotype in an anemic context as indicated by accelerated dextran clearance , with co-suppression of apol1 and myh9 under atpif1α-MO induced anemia ( n = 12–19 embryos/injection; p<0 . 001 for myh9/apol1 MOs vs . myh9/apol1/atpif1a MOs; Fig 5C and 5E ) . Importantly , neither morphant alone resulted in a more severe phenotype under atpif1α-MO induced anemia ( e . g . myh9-MO vs . myh9-atpif1α-MO; p = 0 . 78; or apol1-MO vs . apol1-atpif1α-MO; p = 0 . 90; Fig 5E ) . Furthermore , these observations were reproducible using an independent and non-genetic induction of anemia . Butafenacil , an inhibitor of protoporphyrinogen oxidase , causes loss of hemoglobin following exposure during early zebrafish development[44] . In a butafenacil-induced anemic context ( 0 . 195 μM treatment at 48 hpf ) , we observed a similar effect upon co-suppression of apol1 and myh9 ( n = 17–23 embryos/injection; p<0 . 001 for myh9/apol1 MOs vs . myh9/apol1 + 0 . 195 μM butafenacil; Fig 5D and 5F ) . To dissect further the possible genetic interactions between myh9 and apol1 , we tested whether suppression of endogenous apol1 or ectopic expression of mutant human APOL1 could alter expression of myh9 in zebrafish embryos . We monitored myh9 expression in zebrafish larvae using quantitative real-time PCR in the context of apol1 suppression , and G1 or G2 expression , as well as apol1/APOL1 modulation in conditions of anemia induced by atpif1α-MO injection at 5 dpf ( Fig 6A ) and 3 dpf ( Fig 6B ) . We observed a significant decrease in myh9 expression when zebrafish embryos were injected with the proposed dominant-negative APOL1 G2 allele alone ( 21% reduction; p = 0 . 043; Fig 6B ) , suggesting that the mutant protein may be suppressing myh9 , either directly or indirectly , to induce nephropathy . Furthermore , zebrafish embryos co-injected with APOL1 G2 mRNA and atpif1α-MO display an even greater reduction in myh9 expression compared to controls ( 46% reduction; p = 0 . 0013; Fig 6B ) , and a significant reduction of myh9 expression compared to APOL1 G2 mRNA alone ( p = 0 . 0297; Fig 6B ) , suggesting that the altered APOL1 ( p . Asn388_Tyr389del ) protein has a more pronounced effect on myh9 expression in the context of anemic stress . We also observed a significant increase in myh9 expression in APOL1 G1/atpif1α-MO vs . APOL1 G1 injected embryos ( Fig 6A ) , however , neither of these conditions induced nephropathy . To determine whether this effect was specific to myh9 or was a general effect on transcripts expressed in the glomerulus , we also assessed expression levels of other nephropathy-associated genes during apol1/APOL1 modulation and atpif1α induced anemia . We observed no significant differences in expression of genes implicated in familial focal segmented glomerulosclerosis , including anln[45] , trpc6b[46] , and wt1a[47] upon apol1/APOL1 modulation ( S7 Fig ) , suggesting that APOL1 G2 regulation may be specific to myh9 . Based on the observations that APOL1 G2 expression has the ability to decrease myh9 expression in vivo , we next attempted to rescue APOL1 G2 defects by co-injecting human WT MYH9 mRNA . We injected a constant amount of APOL1 G2-encoding message ( 100pg ) with increasing amounts of human MYH9 mRNA ( 100pg , 150pg , and 200pg ) and scored larvae live for generalized edema at 5dpf . However , we did not observe a significant reduction of edema in APOL1 G2/MYH9 co-injected embryos ( Fig 5C ) , suggesting that compensation with MYH9 message alone is not sufficient to account for the deleterious effects of the G2 variant , possibly because APOL1 G2 has a trans effect on other loci in the genome or is acting to perturb cellular pathways[20] .
In recent years , multiple lines of statistical evidence have implicated the MYH9/APOL1 locus on chromosome 22q12 . 3 with nondiabetic end-stage renal disease , focal segmental glomerulosclerosis , HIV-associated nephropathy , lupus nephritis , SCDN , and diabetic nephropathy in patients of recent African ancestry and European Americans[5–10 , 33 , 35 , 48–50] . Additionally , APOL1 has been associated with an increased burden of cardiovascular disease in African Americans participating in the Jackson Heart Study[51] . Compelling statistical evidence in human cohorts points to the G1 and G2 alleles of APOL1 , rather than MYH9 variation , as the most likely contributors to nephropathy risk . Nonetheless , functional studies of the MYH9 locus provide biological evidence for its role in the kidney , including perturbed glomerular development in myh9 morphant zebrafish[14–17] . Here , we have identified a functional ortholog of human APOL1 in zebrafish and , using transient genetic manipulation , provide functional evidence demonstrating apol1 involvement in both kidney development and filtration . Although the human APOL gene cluster has undergone recent natural selection in primates[26 , 27] , we report the identification of a functional APOL1 ortholog in the zebrafish genome and its implication in renal function . Specific detection of the zebrafish apol1 protein product with the human APOL1 antibody , rescue of kidney defects in apol1 morphant embryos with human APOL1 mRNA , as well as recapitulation of renal phenotypes with an apol1-CRISPR/CAS9 F0 mutant , provide evidence that zebrafish apol1 is indeed functionally relevant to its human ortholog with respect to its role in the glomerulus . Furthermore , no other human mRNA in the human apolipoprotein L family ameliorated kidney defects induced by apol1 knockdown , supporting further its functional orthology to human APOL1 . Nonetheless , it is unclear whether the zebrafish APOL1 protein serves all functions of its human counterpart , especially given the lack of a secretory domain in the zebrafish APOL1 peptide ( Fig 1A ) . Suppression and genome-editing of apol1 in zebrafish and three independent phenotypic scoring paradigms support a role for apol1 in nephropathy; we observed severe edema formation with concomitant glomerular filtration defects and severe podocyte loss . Complementation of apol1 suppression with APOL1 CKD risk alleles ( G1 and G2 ) failed to ameliorate these observed defects . Notably , complementation of each individual variant of the G1 haplotype ( I384M and S342G ) rescued significantly nephropathy phenotypes caused by apol1 suppression , suggesting that both variants must be present in cis to confer risk . This is concordant with initial reports on the lytic potential of APOL1 recombinant proteins on T . b . rhodesiense , in which APOL1 variants with either S342G or I384M alone were less lytic than if both were present together[6] . Strikingly , injection of human APOL1 G2 mRNA alone resulted in significant edema formation in 5dpf zebrafish larvae as well as perturbed glomerular filtration and ultrastructural defects . Our expression data suggest that this could arise from myh9 suppression induced by the altered APOL1 protein harboring the G2 variant . The G2 deletion lies in the SRA-binding domain of APOL1 ( Fig 1B and 1D ) . Therefore it is plausible that disruption of this region of the protein may either prohibit proper binding of APOL1 to its usual partners , or perhaps permit new interactions that induce nephropathy . Further studies are needed to elucidate the functional impacts of the altered APOL1 protein to nephropathy . We also report for the first time functional evidence of a genetic interaction between myh9 and apol1 . Intriguingly , this interaction was only observed in the presence of anemic stress , consistent with our previous genetic association findings in human SCD patients[5] . An immediate question remains regarding the mechanism by which apol1 suppression is inducing kidney injury . Early studies revealed APOL1 mRNA expression in the placenta , lung , and liver , with specific cell-type expression found in endothelial cells and possibly macrophages[26] . More recent studies , however , have characterized the cellular localization of APOL1 in human kidney sections to podocytes , proximal tubules , and arteriolar endothelial cells[18] . These data are consistent with our observation of apol1 morphants and mutants exhibiting extensive podocyte loss and suggest that apol1 is necessary for the development and/or maintenance of glomerular podocytes . Interestingly , it has been shown that APOL1 may cause toxic renal effects through programmed cell death pathways leading to glomerulosclerosis[52 , 53] . Thus , apol1 suppression could dysregulate autophagic pathways , causing podocyte malformation , thereby promoting the susceptibility of the pronephros to glomerular injury . Initial studies implicating MYH9 in nondiabetic nephropathy failed to identify coding variants associated with renal outcome[8 , 9] , and since the nearby nonsynonymous variants identified in APOL1 provided stronger statistical association[5–7] , it was hypothesized that APOL1 variation represents the true attribution to renal disease risk . In fact , it has been shown in multiple studies that controlling for the APOL1 risk alleles ( G1-G2 ) attenuates significantly the effect of MYH9 SNPs[6 , 33] . However , recent reports still demonstrate statistical association of MYH9 in nondiabetic nephropathy[5 , 35] and previous in vivo modeling studies provide further evidence for the role of Myh9 in glomerular development and glomerulosclerosis[14–17] . As such , our group and others have postulated that complex genetic models may exist in this region , including the possibility of MYH9-APOL1 gene interaction[5 , 10] . Our observation of exacerbated glomerular filtration in the context of anemic stress provides biological evidence in support of this hypothesis . Because knockdown of each of myh9 and apol1 independently impairs proper pronephric development and filtration , it is plausible that their encoded proteins are functioning in separate pathways to induce kidney dysfunction . However , these effects only appear to become additive under an additional stress ( anemia ) . The associated variants alone may not be sufficient to induce nephropathy progression , while under low hemoglobin and hematocrit levels , additive effects between MYH9 and APOL1 may become apparent and result in a more drastic reduction in renal function , along with the observed significantly high early mortality rates among SCD nephropathy patients[21 , 22 , 41 , 54] . Furthermore , we provide evidence suggesting that the functional consequences of APOL1 variation may not be acting in a strictly recessive manner as had been previously suggested[5–7 , 55] . Our data demonstrate that APOL1 G1 ( I384M/S342G ) confers loss of proper APOL1 function in the developing zebrafish kidney , while APOL1 G2 is acting in a dominant-negative manner to induce nephropathy , possibly through suppression of myh9 . These data indicate that the risk conferred by the APOL1/MYH9 locus is likely to be governed by a more complex model than recessive patterning as suggested previously . In summary , our study demonstrates the essential role of both apol1 and myh9 in the development of the pronephric glomerulus and proper renal filtration in zebrafish . We report comprehensive in vivo causal evidence of apol1 involvement in kidney decline , and we provide the first in vivo evidence of a potential dominant-negative effect of the APOL1 G2 allele . Further , we have shown that the presence of the G2 allele decreases significantly the expression of myh9 . Similar to the common haplotype on 10q26 that influences age-related macular degeneration underscored by complex regulatory events of neighboring genes ARMS2 and HTRA1 , our data highlight further the importance of comprehensive evaluation of functional consequences at a susceptibility locus[56] . Taken together , these data provide essential biological insight into the mechanisms by which MYH9 and APOL1 confer disease risk and progression in human nondiabetic nephropathies .
We maintained WT zebrafish stocks ( Ekkwill , Ekkwill x AB F1 outcross , or pod::NTR-mCherry[28] according to standard zebrafish husbandry procedures . Embryos were obtained from natural matings of adult fish . Complementation assays were designed essentially as described[57] . Briefly , a MO was designed by Gene Tools , LLC ( Philomath , OR ) to target the translation initiation site of zebrafish apol1 ( NM_001030138 ) ( apol1-MO ) , ( 5’-AGTCGTCCAGCCATTCCATGAGGGT-3’ ) . A translation-blocking morpholino ( MO ) targeting zebrafish myh9 and a splice-blocking MO targeting zebrafish atpif1a were described previously[17 , 43] . APOL1 G1 and G2 allelic constructs were synthesized from a WT APOL1 human ORF clone ( GenBank: BC112943 ) using site-directed mutagenesis ( Stratagene , QuikChange II ) , subsequently transcribed ( mMESSAGE mMACHINE , Life Technologies , Ambion ) into capped mRNA and co-injected with apol1-MO into zebrafish embryos at the one-to-four cell stage ( WT , 100pg/ nl; G1 , 100pg/nl; G2 , 100pg/nl ) . Controls were injected with phenol red . A WPI pneumatic pico pump microinjector was used for MO and mRNA injection to deliver 1 nl/embryo . After injection , embryos were maintained at 28°C in embryo medium . 48 h . p . f . larvae were anesthetized in 1 . 0% tricaine and placed laterally in agarose wells . 70 kDa FITC-conjugated dextran ( LifeTechnologies , 3 . 0nl/embryo ) was injected into the cardiac venous sinus and larvae were transferred to embryo medium for recovery after injection . The eye vasculature of individual fish was imaged at 24 , and 48 hours after dextran injection using a Nikon AZ100 fluorescent microscope and Nikon NIS Elements AR software . The average fluorescence intensity was measured across the eye ( ImageJ ) and changes in intensity relative to the 24 h . p . i measurements were calculated for comparison . GraphPad Prism version 6 . 03 ( GraphPad Software , San Diego , CA ) was used for statistical analysis of relative intensity . Glomeruli from pod::NTR-mCherry adult zebrafish were manually dissected and dissociated in 0 . 5% trypsin/collagenase . Dissociated cells were then filtered through a 70μm strainer and filtered again through a 30μm strainer . Cell-sorting was done on a Beckman Coulter Astrios instrument for mCherry ( 610nm ) . Sorted cells were placed in RLT Buffer ( Qiagen ) and RNA was extracted using the RNeasy Micro Kit ( Qiagen ) . Total RNA from zebrafish embryos was extracted with TRIzol Reagent ( Life Technologies ) and cDNA was reverse transcribed using QuantiTect Reverse Transcription Kit ( Qiagen ) . The following primers were used for amplification: actb1 , Fwd: TTGTTGGACGACCCAGACAT , Rev: TGAGGGTCAGGATACCTCTCTT; nphs2 , Fwd: CCTTCGCTAGCATTCCAGAC , Rev: GCAGCTCTGGAGGAAGATTG; wdr81 , Fwd: ATGGAGAGAAAAACATGGAGGA , Rev: AAGGAGAAAACCTGGAAGAACC; apol1 , Fwd: GACTTTCGATTAAGTGAAACTCAGAGAGA , Rev: GTTATGGTAGCTACACCTCCCACAGCGCTG; myh9 ( qRT ) , Fwd: GGAAAAACCGAAAACACCAA , Rev: CAATATTGGCTCCAACGATGT; anln ( qRT ) , Fwd: TTTGACCTTCACCACCACATT , Rev: TTTGGTGTGATTGCCTTTGA; wt1a ( qRT ) , Fwd: ATGGCCAAACTGTCAGAAGAA , Rev: TTATTTCCTGCCGTTTCTGTG; trpc6b ( qRT ) , Fwd: GGCACCATGAGCCAGAGCCCGGCGTTCGGG , Rev: CTAAGGTGGGCCCATTGGCACTTAAGAAAA . qRT-PCR was performed on a ABI Prism 7900HT instrument and cycle threshold values were computed using SDS 2 . 3 software ( Applied Biosystems ) . Relative expression was calculated against actb1 in each sample and compared against sham-injected controls to determine significant differences in expression . 5 dpf embryos were anesthetized in 1 . 0% tricaine and then fixed in 4 . 0% gluteraldehyde in 0 . 1M Na2PO4 buffer containing 0 . 12mM CaCl2 at 4°C overnight . Fixed larvae were washed in 1X PBS , washed in 1X phosphate buffer , postfixed in 2% osmium tetroxide for 2 hours , and dehydrated through a graded acetone series . Embedding was performed with Epoxy 812 . Sections were cut on a Leica-Reichert Ultracut E ultramicrotome and semithin sections ( 1 . 0μm ) were collected and stained with toluidine blue . 90nm ultrathin sections were placed on copper grids and contrasted with 4 . 0% uranyl acetate for 10 minutes . Grids were incubated in lead citrate ( Reynolds Lead ) for 3 minutes and then examined on a Phillips CM12 electron microscope . Images were taken with an AMT XR61 camera . apol1 gRNA was produced by synthesizing and annealing two oligonucleotides , gRNA F: TAGGGTTGCAGGCCAACCAGTCCT and gRNA R: AAACAGGACTGGTTGGCCTGCAAC . The annealed oligos were then ligated to a T7cas9sgRNA2 vector by performing the ligation and digestion in a single step in a thermal cycler as described [31] . 2 μL of the reaction was used for transformation . Prior to transcription , the gRNA vector was linearized with BamHI . gRNA was transcribed using the MEGAshortscript T7 kit ( Life Technologies , AM1354 ) and purified using alcohol precipitation . A total of 100pg of apol1 gRNA and 200pg of CAS9 protein ( PNA Bio ) was co-injected into individual cells of one-cell stage embryos . For T7 endonuclease I assay , genomic DNA was prepared from 1 dpf embryos as described [58] . A short stretch of the genomic region ( ~270–280 bp ) flanking the apol1 gRNA target site was PCR amplified from the genomic DNA ( Fwd: TGTGTGAAGGATGCATTTGTT , Rev: TGGGATAATGTATGGGAGAATG ) . The PCR amplicon was then denatured slowly and reannealed to facilitate heteroduplex formation . The reannealed amplicon was then digested with 5 units of T7 endonuclease I ( New England Biolabs ) at 37°C for 45 minutes . The samples were resolved by electrophoresis through a 3 . 0% agarose gel and visualized by ethidium bromide staining . Whole embryo protein lysates were collected at 2 dpf by homogenizing anesthetized embryos immersed in RIPA Buffer ( 50 mM Tris , 150 mM NaCl , 0 . 1% SDS , 0 . 5% sodium deoxycholate , 1% Triton X 100 , protease inhibitor ( Roche , cat . no . 11697498001 ) ) . 100 mg protein was loaded into individual wells of a Mini-PROTEAN TGX Precast Gel ( Bio-Rad ) and a western blot was performed as described [59] . Blots were incubated overnight at 4°C with anti-APOL1 antibody ( 1:1000; Abcam , EPR2907 , ab108315 ) . The membranes were subsequently washed in PBST ( 0 . 1% Tween 20 ) and incubated for 1 hour at room temperature with anti-rabbit IgG conjugated to horseradish peroxidase ( 1:20 , 000; GE Healthcare , NA934V ) . ACTIN antibody ( 1:1000 , Santa-Cruz , cat . no . sc-8432 ) was used as a loading control . All animal protocols were reviewed and approved by the Duke University Institutional Animal Care & Use Committee ( IACUC; protocol A229-12-08 ) .
|
African Americans have a disproportionate risk for developing chronic kidney disease compared to European Americans . Previous studies have identified a region on chromosome 22 containing two genes , MYH9 and APOL1 , which likely accounts for nearly all of this difference . Previous reports provided strong statistical evidence implicating APOL1 as the major contributor to nephropathy risk in African Americans , driven by two coding variants , termed G1 and G2 . However , other groups still report statistical evidence for MYH9 association in kidney disease , and animal models have demonstrated biological relevance for MYH9 function in the kidney . Here , we show that suppressing apol1 in zebrafish embryos results in perturbed kidney function . Importantly , using this in vivo assay , we show that the G1 variant appears to cause a loss of APOL1 function , while the G2 variant results in an altered protein that may be acting antagonistically in the presence of normal APOL1 . We also report a genetic interaction between apol1 and myh9 under anemic stress , which is consistent with our previous findings in sickle cell disease ( SCD ) nephropathy patients . Finally , we provide functional evidence in vivo that the G2-altered APOL1 may be interacting with MYH9 to confer nephropathy risk .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
In vivo Modeling Implicates APOL1 in Nephropathy: Evidence for Dominant Negative Effects and Epistasis under Anemic Stress
|
Preventive vaccination is a highly promising strategy for interrupting leishmaniasis transmission that can , additionally , contribute to elimination . A vaccine formulation based on naturally excreted secreted ( ES ) antigens was prepared from L . infantum promastigote culture supernatant . This vaccine achieved successful results in Phase III trials and was licensed and marketed as CaniLeish . We recently showed that newly identified ES promastigote surface antigen ( PSA ) , from both viable promastigotes and axenically-grown amastigotes , represented the major constituent and the highly immunogenic antigen of L . infantum and L . amazonensis ES products . We report here that three immunizations with either the recombinant ES LaPSA-38S ( rPSA ) or its carboxy terminal part LaPSA-12S ( Cter-rPSA ) , combined with QA-21 as adjuvant , confer high levels of protection in naive L . infantum-infected Beagle dogs , as checked by bone marrow parasite absence in respectively 78 . 8% and 80% of vaccinated dogs at 6 months post-challenge . The parasite burden in infected vaccinated dogs was significantly reduced compared to placebo group , as measured by q-PCR . Moreover , our results reveal humoral and cellular immune response clear-cut differences between vaccinated and control dogs . An early increase in specific IgG2 antibodies was observed in rPSA/QA-21- and Cter-rPSA/QA-21-immunized dogs only . They were found functionally active in vitro and were highly correlated with vaccine protection . In vaccinated protected dogs , IFN-γ and NO productions , as well as anti-leishmanial macrophage activity , were increased . These data strongly suggest that ES PSA or its carboxy-terminal part , in recombinant forms , induce protection in a canine model of zoonotic visceral leishmaniasis by inducing a Th1-dominant immune response and an appropriate specific antibody response . These data suggest that they could be considered as important active components in vaccine candidates .
Leishmaniasis is among the most severe parasitic infections affecting humans and dogs in the world . It is the second-highest number of deaths caused by parasites worldwide . Leishmaniasis is remarkably associated with poverty and is an important part of neglected and uncontrolled tropical diseases [1] . Infection is provoked by protozoans of the genus Leishmania , transmitted by the bite of different species of phlebotomine sandflies . They replicate within host mononuclear phagocytes [2–4] . In humans , Leishmania parasites cause a wide spectrum of human diseases ranging from asymptomatic disease , self-healing cutaneous ( CL ) , to disfiguring diffuse cutaneous leishmaniasis ( DCL ) or mutilating mucosal infections ( MCL ) , and from subclinical to acute visceral disease ( VL ) that results in death in susceptible people , causing more than 59 , 000 deaths annually [1 , 5] . Anthroponotic visceral leishmaniasis in India and in Central Africa is caused by L . donovani and is characterized by the absence of animal reservoirs , making man the disease reservoir . Canids ( wild and domestic ) are unequivocally recognized as the main reservoir , continuously supplying the transmission cycle of Leishmania infantum in Old World and L . chagasi ( its synonym in the New World ) . Zoonotic VL is found in the Mediterranean region , several Middle East , African and Asian countries , in South and Central America and probably in southern US [6–8] . VL increased incidence and severity are linked to parasite reservoir migrations , new insect vector localization due to environmental changes , co-infection with immunosuppressive diseases , disease urbanization , deforestation , and poverty [9–13] . In southwestern Europe , at least 2 . 5 million dogs are probably infected [14] and involved in the transmission cycle of L . infantum in humans . In the absence of effective and low-cost drugs for mass administration , priority control measures for elimination aim at reducing the transmission of VL . They include clinical , serological and parasitological diagnosis , treatment by chemotherapy , reduction of the vector population and preventive infectious sandfly bites by using several topical insecticides [15] . These actions are difficult to improve , expensive and weakly effective . Therefore , preventive canine vaccination is a highly promising strategy for interrupting VL transmission and could contribute to elimination [11 , 16] . Furthermore , vaccine development is a promising perspective for human leishmaniasis , as a successful chemotherapy restores impaired cellular immune responses [17] and individuals recovering from leishmaniasis are usually protected against a further infection [18–21] . The elaboration of a safe , efficient and inexpensive anti-leishmanial vaccine provides a substantial goal for global public health for animals and humans . Our interest in Leishmania ES molecules was supported by previous studies showing their immunological properties on macrophage functions and in protection against challenge [22–29] . In recent years , a vaccine based on L . infantum promastigote ES ( LiESAp ) achieved successful results in Phase III trials [23 , 28] . This vaccine was commercialized in Europe as CaniLeish since 2011 [23 , 28 , 30] . A prototype of this vaccine ( LiESAp/MDP ) conferred 93% protection under field condition in France [28] . CaniLeish ( LiESAp/QA-21 ) vaccine provided a significant reduction in the risk of progressing to active infection or overt disease with a clinical efficacy of 68% [30] in a context of a cumulative incidence of 80% of Leishmania infection [31] . More recently , we identified and characterized LaPSA-38S and LiPSA-50S as major immunodominant ES components of L . amazonensis and L . infantum promastigotes , respectively [32] . These proteins were selectively recognized by vaccinated and protected dogs and human cells from immune individuals [24 , 32] . Members of this multi-gene family and the existence of membrane-bound members of the PSA family , such as the PSA-2 gene , have been well documented [33–37] . Leishmania PSA proteins are involved in resistance to complement lysis and in macrophage adhesion/invasion via the complement receptor 3 ( CR3 ) [38 , 39] . PSA-2 complex proteins induce a potent Th1 response in humans [40] , and confer protection against a virulent challenge in mice [41] . In L . infantum and L . major-protected humans , we also clearly demonstrated that the native ES LaPSA-38S protein in recombinant form ( rPSA ) evidenced a Th1-dominant response within an overall mixed Th1/Th2 profile and a cytotoxic response [24] . We hypothesize that this molecule might represent a potential candidate for a next second generation vaccine design . Current recombinant technology leading to a highly purified/defined vaccine formulation increased vaccine safety , stability and reproducibility by decreasing batch to batch variation , compared to the already licensed first generation vaccines . This strategy should decrease the risk of adverse reactions associated with complex vaccine . Moreover , administering just the most immunogenic protein produces a more targeted immune response focused on identified protective antigens . We report here a vaccination trial on naive dogs using the recombinant ES LaPSA-38S antigen ( rPSA ) expressed in L . tarentolae cultured in defined serum-free medium or its carboxy terminal part ( Cter-rPSA ) expressed in E . coli . Recombinant LaPSA-12S antigen ( Cter-rPSA ) is a truncated polypeptide corresponding to the protease-resistant immunodominant carboxyl-terminal domain of the protein [42] . In this first investigation using these recombinant proteins and including a limited number of animals in a short term survey , dogs were randomly included into three experimental groups that received three subcutaneous injections at 4-week interval of either buffer saline or adjuvanted rPSA or adjuvanted Cter-rPSA . Vaccinated dogs had significantly higher levels of specific and efficient IgG2 antibodies and developed an early and specific Th1-dominant cellular immune response ( increased NO-mediated anti-leishmanial macrophage activity in response to higher levels of IFN-γ ) . This dominant Th1 immune profile was found to be correlated with a good protection against an intravenous L . infantum promastigote challenge .
No local and/or general adverse reactions , no hyperthermia , no body weight loss were seen in all immunized dogs upon vaccination . The candidate vaccine tolerance was satisfactory . Immunized dogs were checked monthly for the appearance of external clinical manifestations until 6 months after administration of the parasites . No obvious clinical signs of leishmaniasis were noted in any of the dogs from the vaccinated and placebo groups until 6 months after challenge . The occurrence of parasites in sub-cultures ( Fig 1A ) , the presence of parasite DNA ( Fig 1B ) and the parasite loads using q-PCR ( Fig 1C ) were determined in bone marrow aspirates for the three groups of dogs at 2 , 4 and 6 months post-challenge . A major difference in the number of infected animals between placebo group and the groups of vaccinated dogs during the post-challenge period was noted ( Fig 1 ) . In the control group ( n = 5 ) , three and five dogs were culture positive at 4 months and 6 months respectively , whereas all the animals were PCR-positive at every time point . By contrast , only two ( culture-positive ) and one ( PCR-positive ) out of five dogs vaccinated with Cter-rPSA were found parasite-positive at 4 months and only one remained infected ( PCR- and culture- positive ) at 6 months . Similarly , one ( 11 . 1% ) and two ( 22 . 2% ) out of 9 dogs vaccinated with rPSA were found culture- and PCR- positive at 4 months and 6 months , respectively . Parasite loads are shown in Fig 1C . Mean values were significantly lower in vaccinated dogs ( rPSA: p = 0 . 019 and 0 . 004; Cter-rPSA: p = 0 . 018 and 0 . 031 ) than in placebo group at 4 and 6 months post-infection , respectively . As shown in Fig 2 , specific IgG2 antibody responses against rPSA ( A ) , Cter-PSA ( B ) and LiESAp ( C ) were assessed by ELISA in all serum samples . A weak specific IgG2 response was measured in all dogs immediately before immunization and in dogs from the placebo group at different times post-immunization: 1 month after the second dose of vaccine and 2 months after the third dose . Dogs vaccinated with rPSA/QA-21 and Cter-rPSA/QA-21 had significantly higher levels of anti-LiESAp [except for Cter-rPSA group: p = 0 . 0420 and p = 0 . 4206 , respectively ( Fig 2C ) ] , anti-rPSA [p = 0 . 0033 and p = 0 . 0010 , respectively ( Fig 2A ) ] and anti-Cter-rPSA [p = 0 . 0119 and p = 0 . 0079 , respectively ( Fig 2B ) ] antibodies . IgG2 antibodies were detected as early as one month after the second vaccine candidate injection compared to placebo group for the same period . Specific IgG2 antibody titers remained significantly higher in vaccinated dogs compared to those of placebo animals two months after the third injection . Overall , IgG2 anti-leishmanial response was significantly higher in vaccinated groups compared to placebo group . Viability of L . infantum promastigotes untreated or treated for 30 min with placebo dog sera collected two months after the third injection was nearly identical . More than 90% of promastigotes remained healthy and viable ( data in S1 Fig ) . In contrast , a 30 min exposure of promastigotes with vaccinated dog sera at the same time point revealed around 45% of viability . The anti-proliferative effects on L . infantum promastigote growth were evaluated by exposing parasites 30 min to serum samples from placebo and vaccinated dogs [rPSA ( n = 9 ) or Cter-rPSA ( n = 5 ) ] , washing and culturing for 72 h under standard culture conditions . As shown in Fig 3 , promastigote treatment with placebo dog sera did not affect parasite’s growth during the whole culture time . However , an inhibitory effect was evidenced on the growth of promastigotes previously incubated with vaccinated dog sera collected two months after the third injection ( rPSA: p = 0 . 0010 , Cter-rPSA: p = 0 . 0079 , respectively ) . Macrophage ability to kill L . infantum when pre-infected macrophages were cultured in presence of autologous lymphocytes , expressed as percentage of parasite inhibition index , is presented in Fig 4 . Anti-leishmanial activities were evaluated immediately before immunization and two months after the third injection . At starting point , the assay did not reveal any significant anti-leishmanial activity in any of the dogs , either placebo or vaccinated groups ( Fig 4 ) . By contrast , statistical differences were obtained between vaccinated and placebo groups two months after the completion of vaccine administration . As shown in Fig 4 , higher anti-leishmanial activities were evidenced by infected macrophages from dogs vaccinated with rPSA/QA-21 or Cter-rPSA/QA-21 after co-culture with autologous lymphocytes , as demonstrated by a significant inhibition ( 86 . 4% ( p = 0 . 0033 ) and 63 . 8% ( p = 0 . 0119 ) , respectively ) compared to placebo group ( 13 , 5% ) . NO3-/NO2- ( NO derivatives ) measurements in supernatants of co-cultured cells were directly correlated with NO pathway activation and parasite intracellular killing ( Fig 5A ) . NO production in supernatants of all co-cultures was measured prior to immunization and two months after the third injection . Placebo dogs-infected macrophages synthesized low NO levels ( around 0 . 6 nmol/105 cells/72 hours ) at the different time points analyzed prior to immunization and two months after the third injection . As shown in Fig 5A , the NO levels produced by macrophages from vaccinated dogs ( rPSA/QA-21 and Cter-rPSA/QA-21 ) were significantly higher ( 20 . 75 +/- 4 . 88 nmol/105 cells/72 hours , p = 0 . 0009 and 12 . 24 +/- 4 . 48 nmol/105 cells/72 hours , p = 0 . 0069 , respectively ) than those evaluated before vaccination ( 0 . 52 +/- 0 . 06 nmol/105 cells/72 hours ) and in cell supernatants from placebo dogs 2 months after the vaccine course ( 0 . 50 +/- 0 . 07 nmol/105 cells/72 hours ) . Moreover , the mean value was significantly increased in the group of dogs vaccinated with rPSA/QA-21 compared to the group of dogs vaccinated with Cter-rPSA/QA-21 ( p = 0 . 012 ) . Cytokine contents in culture supernatants from placebo and vaccinated dogs were assessed prior to immunization and at two months after the third injection . As indicated in Fig 5B , prior to immunization , cell culture supernatants from all dogs expressed low IFN-γ levels ( around 0 . 05 ng per mL ) . No increase in IFN-γ level was measured in supernatants of co-cultured pre-infected macrophages from placebo group two months after the third injection . By contrast , in supernatants of pre-infected macrophages co-cultured with autologous lymphocytes from rPSA/QA-21- or Cter-rPSA/QA-21-vaccinated dogs , significant higher IFN-γ levels were measured ( 1 . 98 +/- 0 . 66 ng/mL , p = 0 . 0010 and 1 . 99 +/- 0 . 59 ng/mL , p = 0 . 0079 , respectively ) than in those from placebo ( 0 . 060 +/- 0 . 007 ) or pre-immune dogs ( 0 . 052 +/- 0 . 006 ) ( Fig 5B ) . IL-4 and IL-10 levels in co-culture supernatants from all groups were low and not significantly different prior to immunization , and 2 months after completion of the vaccination protocol ( data in S2 Fig ) .
In this study , we report that naive dogs immunized thrice with either the recombinant ES LaPSA-38S ( rPSA ) or its recombinant C terminal part LaPSA-12S , ( Cter-rPSA ) , and combined with QA-21 as adjuvant , conferred a marked protection against L . infantum promastigote infection . A negative bone marrow parasite load was evidenced in 77 . 8% and 80 . 0% vaccinated dogs respectively at 6 months post-challenge follow-ups . The parasite burden was significantly reduced in three infected vaccinated dogs , but the short term survey did not allow the observation of any delayed disease progression or a role as reservoirs . Investigations on potential Leishmania antigen screening and best adjuvant for vaccines have required substantial effort [11 , 43] . First vaccine attempts based on killed or attenuated parasites , conceptually simple to produce in endemic areas at low cost , gave disappointing results in field trials [9 , 44] . Second-generation vaccines are now available . Only two dog vaccines achieved successful results in Phase III trials: 1- Leishmune ( Fort Dodge Animal Health ) , based on a glycoproteic complex ( fucose mannose ligand ) from L . donovani adjuvanted with QS-21 and deacylated saponins of Quillaja saponaria , licensed and used in Brazil [11 , 45 , 46] and 2- CaniLeish ( Virbac Animal Health ) a formulation related to the LiESAp vaccine using L . infantum promastigote ES antigens from culture supernatant combined to QA-21 as adjuvant [22 , 25 , 27 , 28 , 47 , 48] , licensed and marketed in Europe since 2011 . A primary vaccination course of three injections at three week intervals , followed by annual booster vaccinations , is required with CaniLeish . The same protocol including two booster injections was scheduled in this study , which represent an important drawback in endemic areas . Various Leishmania molecules have already been reported as promising vaccine candidates ( reviewed by [49 , 50] ) , however very few recombinant vaccines have succeeded beyond the laboratory rodent stage . This in part is linked to laboratory animal models which do not mimic natural infection . Only LeishTec vaccine ( recombinant A2 antigen plus adjuvant ) led partial protection ( 40% ) versus L . chagasi promastigote challenge and has been marketed as a canine vaccine ( Hertape Calier ) in Brazil . Other vaccine candidates , such as the multicomponent Leish-111f fusion protein including TSA , LmSTI1 , and LeIF antigens [31 , 51] , gave disappointing results in dogs . Similarly , trials with recombinant cysteine peptidases plus canine IL-12 , or recombinant Histone H1 , hydrophilic acylated protein B1 , or Leishmania activated C kinase receptor ( LACK ) analog , did not protect dogs from Leishmania infection [52] . We report here a successful vaccination study with recombinant Leishmania proteins in dogs . Efficacy of vaccination was corroborated by both qPCR and culture methods and clear differences in humoral and cellular immunity were evidenced between placebo and vaccinated groups . Our results demonstrated clear-cut immune response differences between vaccinated and placebo dogs . An early increase in specific IgG2 antibodies , correlated with protection , was only evidenced in rPSA/QA-21- and Cter-rPSA/QA-21-immunized dogs . The IgG1 antibodies are associated with susceptibility , disease severity and correlates with a Th2 response [53–55] , while IgG2 are predominant in naturally resistant or vaccinated dogs and are associated with an appropriate Th1-dominant response [56–58] . More recently , we provided direct evidence that cooperation of both humoral and cellular immune response might be essential for protection in LiESAp/muramyl dipeptides immunized dogs [22] . We demonstrated here that inactivated sera from rPSA/QA-21- or Cter-rPSA/QA-21-vaccinated dogs supported promastigote killing and a significant parasite growth culture inhibition . Our data evidenced that an appropriate antibody response , such as mediated by anti-rPSA and anti-Cter-rPSA antibodies , might play a major role in canine VL protection . Recent technological advances have allowed a global analysis of Leishmania secretome [59 , 60] . However , secretome functions are poorly known since few of these molecules have been extensively characterized [59 , 61] . Analysis of recombinant rPSA and Cter-rPSA amino acid sequences reveal that Cter part contains a Threonine/Serine-rich domain and Proline/Cysteine-rich regions . These regions are relatively conserved between Leishmania species but their lengths vary considerably and they can be totally absent in some PSAs [33] . The immunodominant humoral and cellular responses to this region might be linked to its proteolytic stability and hydrophobicity [36 , 42 , 62] . This suggests that ES PSA might have an important immunoregulatory role between parasites and their target cells and on the immune responses . Promastigotes might secrete PSA to modify macrophage functions even before parasite engulfment . Analogously , amastigotes secrete PSA , which might be transported across the parasitophorous vacuole membrane , to interfere with macrophage signaling pathways , thereby preventing macrophage activation . So , future candidates might exist in Leishmania ES proteins and be used for new generation vaccine design . Intracellular Leishmania parasite killing by macrophages is essential for cure . We previously developed an ex vivo canine macrophage-autologous lymphocyte co-culture system to investigate protective cellular response . This co-culture system was used to analyze NO pathway involved in macrophage parasite killing . Increased IFN-γ and NO production , as well as anti-leishmanial macrophage activity , were verified only in vaccinated dogs , and lasted two months after the full vaccination course . This further argues that Leishmania killing is mediated via an L-arginine NO pathway , induced by Th1 cytokines , mainly IFN-γ . All these points support the hypothesis that ES PSA and its carboxy terminal part , in recombinant forms , induce protection in canine VL by inducing a Th1-dominant immune response and an appropriate specific antibody response . In addition , the nature and magnitude of immune responses revealed by PSA formulations is very similar to those induced by the different LiESAp formulations . This suggests that ES PSA is one of the main active compounds of the already licensed CaniLeish . As NO and IFN-γ are also involved in human leishmaniasis [63] , monitoring these parameters might represent markers of a protective response for recombinant PSA or its carboxy-terminal part-based vaccine development and for large scale field studies . Altogether , these results , obtained in this experiment with a limited number of animals and with a short term survey , deserve further investigations to evaluate these vaccine candidates against a severe disease for both humans and dogs in natural conditions of infection .
Nineteen young adult dogs , 10 males and 9 females , between 2 and 4 years old , were selected on clinical and serological criteria from a colony of naive Beagles from the kennel CEDS ( Domaine des Souches , Mezilles , France ) . Dogs were housed at the animal facility of the National Veterinary School of Lyon ( ENVL , France ) in the “Unité d’Etudes PréCliniques” ( UEPC ) during the time course of the experiment under conditions designed to exclude any possible natural leishmanial infections . They were well-fed animals under constant scrutiny of health problems by a veterinarian and had all received their yearly routine vaccinations . Care and management of dogs were carried out according to ethical guidelines laid down in the National Veterinary School of Lyon ( ENVL ) . Protocols were submitted to and approved by ethics committee of the ENVL ( N° ICLB 135/08 ) and performed according to recommendations to limit the number of animals and long-term experimentations . All dogs had a specific code/ID throughout the experiment . The animals were maintained in quarantine for a period of 30 days before the initiation of the experiment . Prior to vaccination , blood was collected and then sera and genomic DNA of all dogs were separated and extracted in order to exclude any infected dog . Beagles were randomized by sex and age into three experimental groups and the study was performed in a double–blind randomized fashion . Dogs of each group received three subcutaneous injections at a 4-week interval of either freeze-dried dose of 1 mL buffer saline ( Placebo group , n = 5 ) , 25 μg recombinant LaPSA-38S adjuvanted with 60 μg QA-21 ( rPSA group , n = 9 ) or 25 μg recombinant LaPSA-12S formulated with 60 μg QA-21 ( Cter-rPSA group , n = 5 ) . Two months post-immunization , all dogs were challenged by intravenous injection of 108 infective promastigotes of L . infantum ( MHOM/MA/67/ITMAP-263 strain , clone 2 ) . Primary cultures of virulent promastigotes , differentiated from amastigotes isolated from the spleen of heavily infected mice ( BALB/c ) , were used for the virulent challenge . Specific IgG2 antibody responses against rPSA , Cter-rPSA and LiESAp were measured in the serum samples of control and vaccinated dogs by a standard ELISA procedure . Briefly , sera from immune or control dogs were added in triplicate at 1/50 dilution in PBS containing 0 . 05% Tween-20 to 96-well plates previously coated with rPSA ( 0 . 1 μg per well , batch #A50054 ) , Cter-rPSA ( 0 . 1 μg per well , batch #070425 ) or LiESAp ( 1 μg per well , batch #0011 ) . After 1 h incubation at 37°C , plates were washed extensively with PBS-0 . 05% Tween-20 and incubated for 30 min at 37°C with secondary antibody ( horseradish peroxidase-conjugated sheep anti-dog IgG2 , 1/5000 ) . After three washes in PBS-0 . 05% Tween-20 , plates were developed with OPD substrate ( with H2O2 in citrate buffer ) and absorbance was read using microplate reader at 492 nm wavelength . For analysis , a threshold of positivity was estimated by calculating a cut-off using the following formula: mean OD in sera collected from all dogs at the starting point ( before immunization ) + 3 standard deviations . Promastigotes of Leishmania infantum were collected by centrifugation and washed three times in PBS . A total of 5x106 parasites were incubated ( or not ) with 100 μL of complement heat-inactivated serum of dog for 30 min , at dilution ¼ in culture medium . These sera were collected from all dogs 2 months after vaccination . Cell’s viability was assessed by trypan blue staining , the parasites were washed three times in PBS , and then cultivated at 25°C in 5 mL of RPMI medium supplemented with 20% heat-inactivated foetal calf serum ( FCS ) . Parasites were counted daily for three days by flow cytometry ( FACSCanto , Becton Dickinson ) . Results are expressed in percentage of promastigote growth inhibition at day 3 . Peripheral Blood Mononuclear Cells ( PBMC ) were obtained from heparinized peripheral blood by density centrifugation through Ficoll-Hypaque ( GE Healthcare Life Sciences ) . Canine monocyte-derived macrophages ( CM-DM ) and non-adherent cells ( i . e . lymphocytes ) were prepared by differential adherence of PBMC as previously described [25 , 27] . CM-DM separated from lymphocytes were cultured for 5 days at 37°C and 5% CO2 in RPMI 1640 medium ( BioWhittaker ) , supplemented with 2 mM glutamine , 10% heat-inactivated FCS , 100 μg/mL streptomycin and 100 IU/mL penicillin . They were infected with stationary-phase promastigotes of L . infantum ( MHOM/MA/67/ITMAP-263 strain , clone 2 ) at a parasite: macrophage ratio of 5:1 for 150 min in LabTek 16-well glass chamber slides . Non-internalised parasites were removed by gentle washing . The cells were checked to verify that greater than 40% were infected . Infected macrophages were then incubated alone or in the presence of autologous lymphocytes at a lymphocyte: macrophage ratio of 2:1 . After a 72 h co-culture , supernatants were collected for further analyses and the lymphocytes were removed by gentle washings . Macrophages were fixed with methanol and stained with Giemsa in order to determine the parasitic index . For assessment of anti-leishmanial activity , the percentages of infected cells and the number of amastigotes per macrophage were estimated in duplicate experiments by microscopic examination of Giemsa-stained preparation and were used to calculate the parasite index ( PI ) inhibition . Percentage of PI inhibition = 100 –[ ( mean number of amastigotes per macrophage × percentage of infected macrophages when macrophages were incubated with autologous lymphocytes ) / ( mean number of amastigotes per macrophage × percentage of infected macrophages in untreated macrophages ) ] × 100 . NO3-/NO2- accumulation in supernatants from 72 h co-cultured cells ( pre-infected macrophages exposed to autologous lymphocytes ) was used as an indicator of NO production by activated macrophages and was assayed by the Griess reaction using the nitrate/nitrite colorimetric assay of Alexis biochemicals ( Enzo Life Sciences , France ) . The Griess reagent was modified according to Pinelli et al . [64] . IFN-γ , IL-4 and IL-10 levels were determined as previously described [25] in cell culture supernatants by a two-site sandwich ELISA using specific anti-dog IFN-γ ( 2 μg/mL ) , anti-dog IL-4 ( 1 μg/mL ) and anti-dog IL-10 ( 1 μg/mL ) antibodies ( R&D Systems , Minneapolis , USA ) , biotinylated anti-dog IFN-γ ( 100 ng/mL ) , anti-dog IL-4 ( 50 ng/mL ) and anti-dog IL-10 ( 50 ng/mL ) antibodies ( R&D Systems , Minneapolis , USA ) and streptavidin conjugated to horseradish peroxidase ( 1/200 ) ( R&D Systems , Minneapolis , USA ) . Absorbance values were read at 490 nm wavelength in an automatic microplate reader ( Wallac Victor21420 Multilabel counter , Perkin-Elmer life sciences ) . Standard curves for IFN-γ , IL-4 and IL-10 , respectively , were performed by the use of recombinant canine proteins ( R&D Systems , Minneapolis , USA ) . The health status of the animals was routinely followed by veterinarians ( including appetite , physical examination and physical activity ) . The dogs were monitored for 3 weeks after each injection . Local tolerance was investigated by direct visual examination and any lesions were scored daily over a period of 14 days after each injection . General tolerance was investigated by means of a weekly general clinical examination and a daily general health evaluation with rectal temperature measurement . Body weights were measured once a week throughout the trial period . Dogs were monitored for parasite establishment and subsequent development of the disease by routine screening for classical clinical signs and parasite isolation . Infection was assessed at 2 , 4 and 6 months post-challenge . For that , dogs were anesthetized and bone marrow aspirates were collected by sternal puncture into citrate tubes to assess parasite load . The presence of Leishmania parasites was determined by culturing parasites in NNN biphasic medium at 25°C . Bone marrow samples ( about 500 μl ) were cultured in NNN biphasic medium ( containing 2 mL of RPMI-20% heat-inactivated FCS ) for 1 week . Subcultures were weekly realized by adding 0 . 5 mL or 1 mL of culture sample in NNN medium containing 3 mL of RPMI-20% FCS ( 4 subcultures ) . The presence of parasites was determined by regular microscopic observation for 20 min in an inverted microscope at 400x magnification . When parasites were observed , the sample was considered as parasite positive . The presence of Leishmania DNA was also assayed in bone marrow samples of all the enrolled dogs by real-time quantitative PCR ( qPCR ) as previously described for kinetoplast DNA amplification [65] . After lysis , the DNA of each bone marrow sample was extracted using a silica column ( QIAamp DNA mini kit ) . The Stratagene ( La Jolla , California , USA ) MX 4000 system was used for amplification and detection . Optimization experiments led us to use the Stratagene qPCR master mix , 15 pmol of forward primer ( CTTTTCTGGTCCTCCGGGTAGG ) , 15 pmol of reverse primer ( CCACCCGGCCCTATTTTACACCAA ) , and 50 pmol of TaqMan probe ( FAM-TTTTCGCAGAACGCCCCTACCCGC-TAMRA ) . Assays were performed with a 25 μL final volume with 1 μL of DNA sample . The standard curve was established from Leishmania DNA extracted from 5 × 106 parasites; 1 μL of each serial dilution , ranging from 50 , 000 to 0 . 0001 parasites , was introduced into reaction tubes . TaqMan chemistry allowed two-step temperature ( 94 and 55°C ) cycling over 45 cycles . Comparative quantification was performed by using a single copy gene , the DNA polymerase gene , as a normalizer . Primers and a probe described previously by Bretagne et al . [66] ( TGTCGCTTGCAGACCAGATG [200 pmol] , GCATCGCAGGTGTGAGCAC [200 pmol] , and VIC-CCAGGCTCGAAGTTGTTGCTGCCC-TAMRA [200 pmol] ) and the same working conditions as previously described for kinetoplast DNA amplification were used . Results were expressed as the number of parasites per mL of bone marrow aspirate . A sample was considered as positive when the established parasite concentration was superior to 40 parasites per mL . Data analysis was performed with GraphPad Prism version 5 . 03 for Windows , GraphPad Software ( San Diego , California USA ) . Statistical significance of differences between groups was determined by Mann-Whitney-Wilcoxon test . A p-value ≤ 0 . 05 was considered significant . The ENVL Ethical Committee approval confirms that this study was carried out in accordance with the G . R . I . C . E . “Ethical Committee Regulation applied to animal experimentation” guidelines ( implemented in France in 2008 ) under project number 135 . 08 .
|
Visceral leishmaniasis ( VL ) , a potentially fatal disease caused by L . infantum , represents perfectly the need for a “One Health” approach for disease control , since it affects both humans and dogs , with similar clinical outcome and T-cell mediated immunity commitment . The dog vaccine development is highly required as our present resources for VL treatment and control have a limited effectiveness . It would represent the most convenient and efficient control way to decrease the dog-sandfly-dog transmission cycle , essential for human incidence reduction . The results indicate that recombinant forms of soluble promastigote surface antigen ( PSA ) are very promising effective vaccine candidates against canine VL . The elicited immune responses effectively reduced parasite load in in vitro pre-infected macrophages and in experimentally infected dogs . Through this approach , we aim to reduce the number of infected animals developing progressive infections thereby positively influencing human public health .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immunology",
"microbiology",
"vertebrates",
"parasitic",
"diseases",
"dogs",
"animals",
"mammals",
"protozoan",
"life",
"cycles",
"parasitic",
"protozoans",
"vaccines",
"preventive",
"medicine",
"developmental",
"biology",
"protozoans",
"leishmania",
"vaccination",
"and",
"immunization",
"promastigotes",
"public",
"and",
"occupational",
"health",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"life",
"cycles",
"recombinant",
"proteins",
"biochemistry",
"leishmania",
"infantum",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"protozoology",
"macrophages",
"amniotes",
"organisms"
] |
2016
|
Recombinant Forms of Leishmania amazonensis Excreted/Secreted Promastigote Surface Antigen (PSA) Induce Protective Immune Responses in Dogs
|
The large polymerase subunit ( L ) of non-segmented negative strand RNA viruses transcribes viral mRNAs and replicates the viral genome . Studies with VSV have shown that conserved region V ( CRV ) of the L protein is part of the capping domain . However , CRV folds over and protrudes into the polymerization domain , suggesting that it might also have a role in RNA synthesis . In this study , the role of respiratory syncytial virus ( RSV ) CRV was evaluated using single amino acid substitutions and a small molecule inhibitor called BI-D . Effects were analyzed using cell-based minigenome and in vitro biochemical assays . Several amino acid substitutions inhibited production of capped , full-length mRNA and instead resulted in accumulation of short transcripts of approximately 40 nucleotides in length , confirming that RSV CRV has a role in capping . In addition , all six variants tested were either partially or completely defective in RNA replication . This was due to an inability of the polymerase to efficiently elongate the RNA within the promoter region . BI-D also inhibited transcription and replication . In this case , polymerase elongation activity within the promoter region was enhanced , such that the small RNA transcribed from the promoter was not released and instead was elongated past the first gene start signal . This was accompanied by a decrease in mRNA initiation at the first gene start signal and accumulation of aberrant RNAs of varying length . Thus , in addition to its function in mRNA capping , conserved region V modulates the elongation properties of the polymerase to enable productive transcription and replication to occur .
RSV is a major cause of respiratory disease in infants , immunosuppressed individuals and the elderly . Currently there is no vaccine and the only means to mitigate the infection is by prophylactic administration of antibodies or ribavirin treatment in select cases [1] . Thus , there is a need for novel antivirals against the virus . RSV is a member of the Family Pneumoviridae in the Order Mononegavirales , the non-segmented , negative strand RNA viruses ( nsNSVs ) [2] . Like other viruses in the order , the RSV genome is transcribed and replicated by the viral RNA-dependent RNA polymerase . Because it is essential for viral multiplication , and possesses enzymatic properties , the polymerase is considered a highly promising target for antiviral drug development [3 , 4] . The core RSV polymerase is comprised of a complex of two proteins: the 250 kDa large polymerase subunit ( L ) , which contains the enzymatic domains involved in transcription and replication , and the 27 kDa phosphoprotein ( P ) , which is an essential co-factor [5 , 6] . To perform mRNA transcription , the polymerase initiates RNA synthesis from position 3 ( +3 ) of the leader ( le ) promoter at the 3' end of the genome . This results in synthesis of a small le RNA [7] . This RNA is shorter than the 44 nt le promoter region , but heterogeneous in length , ranging from approximately 20–30 nt , with the dominant length being approximately 25 nt . This heterogeneity suggests that the polymerase does not release the RNA in response to a specific signal , but is relatively non-processive in the promoter region and releases the RNA after it has moved away from the core promoter . The polymerase then scans to the first gene at position 45 and transcribes the genome by responding to cis-acting signals that flank each of the genes [8] . A gene start ( gs ) signal directs the polymerase to initiate mRNA synthesis , and a gene end ( ge ) signal causes the polymerase to polyadenylate and release the mRNA [9–11] . During transcription , an elongation factor , M2-1 , is required for synthesis of full-length mRNAs , particularly from the longer genes [12–14] . Replication of the genome involves production of uncapped , encapsidated antigenome and genome RNAs . Like mRNA transcription , RNA replication is initiated at a promoter at the 3' end of the le region [15–17] . In this case , the polymerase initiates RNA synthesis opposite position 1 of the promoter ( +1 ) and synthesizes a full-length antigenome RNA [18–20] . The antigenome contains a trailer ( tr ) promoter region at its 3' end ( the complement of the 5' terminal trailer region ) , which signals the polymerase to synthesize genome RNA from position 1 and a small tr RNA from position 3 [19 , 21] . Replicative RNAs are encapsidated with nucleoprotein ( N ) as they are synthesized , which is thought to allow the polymerase to efficiently elongate the RNA to the end of the template [7 , 8 , 17 , 22 , 23] . Concurrent encapsidation might also be required to enable the polymerase to avoid releasing the RNA after ~ 25 nt in the promoter regions [8] . In addition to their polymerization activity , nsNSV polymerases are responsible for adding a methylated cap to the mRNA [24–26] . Studies with members of the Rhabdoviridae have shown that their polymerases cap viral mRNAs by an unusual RNA:GDP polyribonucleotidyltransferase ( PRNTase ) activity , rather than by a guanylyltransferase reaction [26–29] . This activity has been mapped to a region of the polymerase called conserved region V ( CRV; Fig 1A ) . Key residues within CRV are conserved across the nsNSVs ( Fig 1B ) , indicating that this capping mechanism is likely conserved throughout the order [24 , 26 , 27 , 30] . Capping occurs co-transcriptionally , presumably when the mRNA has reached a sufficient length to be extruded from the polymerization active site; in the case of vesicular stomatitis virus ( VSV ) , this happens when the mRNA has reached 31 nucleotides [31] . Mutagenesis studies of the VSV capping domain have also shown that cap addition is required for efficient polymerase elongation during mRNA transcription , with capping failure correlating with production of transcripts of ~40 nt in length [24 , 32 , 33] . These findings suggest there is a checkpoint during mRNA synthesis to only allow elongation of successfully capped mRNA transcripts . The VSV polymerase structure suggests that in addition to its defined role in capping , CRV might also affect RNA synthesis . In a pre-initiation form of the polymerase , CRV resides in a lobe that is folded over the polymerization domain , and a putative priming loop protrudes into the active site [34] ( Fig 1C ) . A hint that CRV might play a role in RNA synthesis also comes from analysis of an RSV small molecule polymerase inhibitor called BI-D . Resistance to BI-D maps to CRV suggesting that it binds this domain [25] ( Fig 1C ) . Experiments using an in vitro polymerization assay showed that BI-D enabled the RSV polymerase to reach the end of an oligonucleotide template more efficiently than in the absence of compound [35] . Based on these observations , we hypothesized that CRV might have other roles in transcription and replication beyond mRNA capping . In this study , we perturbed CRV using alanine substitutions or BI-D to determine 1 ) if CRV is involved in RSV mRNA capping , and 2 ) if CRV plays an additional role in RNA synthesis , unrelated to its role in mRNA capping . The data obtained show that CRV does affect capping , as expected , but that in addition , functions to regulate the elongation step of RNA synthesis . These two properties could be separated by single amino acid substitutions in CRV indicating that they are distinct . Moreover , the data indicate that the elongation properties of the polymerase must be appropriately calibrated for it to be able to perform transcription and replication .
To tease apart the role of CRV during RSV transcription and genome replication , single amino acid alanine substitutions were introduced into T7-expressed L protein so that their effects could be monitored in a cell-based minigenome system . The substituted residues included G1264 , T1267 , E1269 , S1333 , N1335 and R1339 , which lie within , or proximal to , the pocket defined by the polymerase variants that inhibited capping in VSV , and where BI-D resistance variations could be mapped ( E1269 is one of the BI-D resistance sites; Fig 1C ) . As a control , a substitution was also introduced into conserved region III of L ( D811 ) , which is within the RNA synthesis catalytic domain [6 , 36 , 37] . Prior to performing functional analysis of the variant L proteins , we confirmed that the substitutions did not inhibit expression of the L protein . Because there was no effective antibody available against L protein , hexahistidine tagged versions of wt and variant L proteins were constructed . L requires co-expression of P to be efficiently expressed [6 , 38] and so the RSV L and P proteins were co-expressed from T7 plasmids in HEp-2 cells . Western blot analysis gave only a very weak and inconsistent signal for L , likely due to poor transfer of the protein because of its large size . Therefore , as an alternative approach , L-P complexes were isolated by affinity binding to Ni-NTA agarose beads and examined by SDS-PAGE and colloidal blue staining ( Fig 1D ) . This analysis showed that all L protein variants were expressed at a similar level as the wt L protein and retained their ability to bind P . The L variants were then analyzed in a minigenome assay [39] to examine the effects of the substitutions on RSV transcription and replication . The minigenome contained the le promoter at its 3' end and trailer sequence at its 5' end . The ten naturally occurring RSV genes were replaced with two genes containing different fragments of chloramphenicol acetyl transferase ( CAT ) sequence , each flanked with RSV gs and ge sequences ( Fig 2A ) . To uncouple minigenome transcription from replication and allow the two processes to be examined as independent events , the 5' terminal trailer region was modified with a substitution at position 2 relative to its 5' end . This inactivated the tr promoter at the 3' end of the antigenome , preventing it from acting as a template for genome synthesis and thus limiting minigenome replication to the antigenome synthesis step , as described previously [16] . RSV RNA synthesis was reconstituted intracellularly by transfecting BSR-T7 cells with plasmids expressing the minigenome RNA and RSV N , P , M2-1 , and L proteins . For this , and all following experiments using the minigenome system , BSR-T7 cells were used to supply T7 RNA polymerase , instead of MVA-T7 , to avoid the possibility of artifacts arising from the capping enzyme activity of vaccinia virus . For these experiments , untagged L constructs were used , as the tag impaired polymerase activity to some extent . The products generated by the L variants were detected by Northern blot analysis using a negative sense riboprobe that would detect both mRNAs and the antigenome ( Fig 2A ) . In cells transfected with wt L protein , three major RNA species accumulated , representing encapsidated antigenome , and polyadenylated CAT 1 and CAT 2 mRNAs ( Fig 2B , lanes 2 and 3 ) [12] . The identities of these bands have been determined previously by manipulating gs and ge signals , and using specific probes ( e . g . [13 , 17] ) . Analysis of the L variants showed that the S1333A variant generated a relatively high level of mRNA , and the T1267A and E1269A variants showed moderate levels of mRNA production , as compared to the wt L protein ( Fig 2B and 2C ) . The mRNAs produced by these variant polymerases were diffuse in nature , similar to those produced by the wt polymerase , indicating that they were polyadenylated . In contrast , the G1264A , N1335A , and R1339A polymerases yielded significantly reduced levels of mRNA compared to wt ( p <0 . 05 ) . Thus , several amino acid substitutions inhibited transcription . Given that CRV is known to be involved in mRNA capping , the block to transcription was likely to be at the capping step . We have been unable to reconstitute efficient PRNTase activity in vitro using purified polymerase , as has been described for VSV , Chandipura and rabies viruses [24 , 26 , 28 , 40] to confirm this directly . This might be because the four guanosines that lie at the 5' end of RSV mRNAs have the potential to form G-quadruplexes , preventing oligonucleotides containing this sequence from being bound as a substrate in a trans-capping assay . Therefore , we used primer extension analysis to examine if the mRNAs produced by the variant polymerases were capped . As described in the Introduction , the cap is added co-transcriptionally after the mRNA has reached a given length . If the substitutions inhibiting transcription were specifically targeting the PRNTase activity of the polymerase , it would be expected that the polymerase would retain its ability to initiate mRNA synthesis at the first gs signal at the beginning of the CAT 1 gene . To determine if the variant L proteins were capable of mRNA initiation , the RNA samples were subjected to primer extension analysis using probe 56–75 , which corresponds to nucleotides 56–75 of the minigenome and so hybridizes to nucleotides 12–31 at the 5' end of the CAT 1 mRNA transcript ( Fig 3A ) . If RSV were similar to VSV , in which the cap is added when the transcript reaches 31 nucleotides in length [31] , it would be expected that this primer would be able to bind to transcripts irrespective of whether they were capped or uncapped . Primer extension analysis was first performed using a reverse transcriptase that does not recognize the cap , which allowed accurate determination of initiation site selection . Analysis of RNA produced by the wt L protein revealed two bands , corresponding to antigenome initiated at position 1 of the genome , and RNA generated from the gs signal ( position 45 ) ( Fig 3B , compare lanes 3 and 4 with lanes 2 and 5 ) . RNA initiated at the gs signal was also readily detected in reactions containing the L variants ( Fig 3B , lanes 6–11 ) . There was experimental variation in the levels of this RNA ( Fig 3C ) , likely because some of these products were relatively unstable . Nonetheless , the data clearly showed that all variants were capable of initiating mRNA synthesis from the gs signal . Comparison of transcript length with a molecular weight marker that indicated the length of transcripts initiated at the beginning of the gs signal ( position 45 ) also showed that the transcripts were initiated at the correct position ( Fig 3B , compare lane 1 with lanes 3–11 ) . Thus , the variant polymerases were able to initiate mRNA synthesis from the gs signal , with no significant difference from wt . To determine if the mRNAs were capped , experiments were performed using a different reverse transcriptase capable of reverse transcribing the guanosine cap ( Fig 3D–3F ) . To establish this assay , an additional control was used , in which a double alanine substitution was introduced into the HR motif ( H1338/ R1339 ) . In VSV , both residues are essential for PRNTase activity , and the histidine forms a covalent linkage to the mRNA substrate [29] . Reverse transcriptase products of RNA produced by the wt polymerase migrated as a doublet: a faint lower band migrated alongside the marker ( representing uncapped RNA ) , and a dominant upper band migrated slightly more slowly ( Fig 3D , lane 3 ) . Treatment of the RNAs with pyrophosphatase to cleave the linkage between the guanosine cap and RNA resulted in a reduction in the level of the upper band , and an increase in the signal from the lower band ( Fig 3D , lanes 3 and 4 ) . This finding showed that the upper band corresponded to capped RNA ( cap removal might have been incomplete because the minigenome mRNA was within a pool of total cellular mRNA , which would have competed for enzyme ) . In the case of the H1338A/ R1339A variant , the lower band was the only dominant band detected ( Fig 3D , lane 7 ) . Faint traces of slower migrating RNA could be detected , but this was unaffected by pyrophosphate cleavage ( Fig 3D , lane 8 ) and might be due to low level stuttering of the reverse transcriptase on the repetitive sequence that lies at the 5' end of RSV mRNAs ( 5' GGGG ) . The R1339A variant had a similar phenotype as the H1338A/R1339A variant , demonstrating that this was defective in capping ( Fig 3D , lanes 5 and 6 ) . Analysis of the RNAs produced by the other variant polymerases showed that S1333A variant was efficient in capping , the T1267A and E1269A variants were somewhat less efficient , and the G1264A and N1335A variants capped with poor efficiency ( Fig 3E ) . The RNAs were also examined with a primer that hybridized further downstream on the mRNA transcripts ( primer 91–113 ) , which binds at positions 47–69 from the 5' end of the mRNA . Although this primer detected products from the wt , T1267A , E1269A and S1333A variants , it could only detect very low levels of RNA from the G1264A , N1335A and R1339A variants ( Fig 3F ) . This result shows a correlation between the ability of the polymerase to cap the RNA and to elongate it far enough to be detected by the 91–113 primer . To examine the lengths of the RNA produced by the G1264A , N1335A and R1339A variants , the RNA samples were subjected to Northern blot analysis using gel electrophoresis conditions optimized for examining short RNAs and probed with the 56–75 oligonucleotide used for the primer extension analysis ( Fig 3A and 3G ) . For wt L protein , there were only faint bands apparent under these electrophoresis conditions , suggesting that this protein generates predominantly full-length mRNA ( Fig 3G , lane 2 ) . A similar result was obtained for the S1333A variant , which was highly efficient for capping and full-length mRNA synthesis ( Fig 3G , lane 7 ) . In contrast , the other variants produced RNAs that were ~30–50 nt in length , with the dominant signal at ~40 nt . The quantities of the ~40 nt RNAs detected for the G1264A , T1267A , E1269A , N1335A and R1339A variants were somewhat inconsistent between experiments , likely because the small size of this RNA meant it was not efficiently retained on the membrane . However , it was consistently observed that the variants most defective in transcription produced a ~ 40 nt RNA , which was not produced by the wt or S1333A variants . Together the data in Figs 2 and 3 show that substitutions in CRV cause a defect in transcription that correlates with inhibition of capping . The data also suggest that capping is required for RNA to be elongated beyond ~ 40 nt , consistent with what has been shown for VSV . The data presented in Fig 2 show that , in addition to inhibiting transcription , some CRV variants were deficient in RNA replication , producing low to undetectable levels of antigenome . Given that replication products of nsNSVs are not capped , this finding shows that CRV has another role besides capping . To examine the effect of CRV substitutions on RNA replication specifically , the effects of the substitutions were analyzed using a replication-specific minigenome , in which the le promoter was replaced with nucleotides 1–36 of the tr promoter and the first gs signal was deleted ( Fig 4A ) . The tr is a somewhat stronger replication promoter than the le [41] , and the absence of a gs signal adjacent to the promoter prevents mRNA synthesis [11] , allowing replication products to be measured without contamination with signal from polyadenylated monocistronic and readthrough mRNAs . Northern blot analysis showed that all the variants generated significantly lower levels of replication product ( genome ) than wt L protein ( p <0 . 005 for each variant ) , with variants E1269A , S1333A , and N1335A generating replication product at a reduced level compared to wt ( Fig 4B , lanes 7–9; 4C ) , and no product being detected for variants G1264A , T1267A , or R1339A ( Fig 4B , lanes 5 , 6 , and 10; 4C ) . Thus , substitutions in CRV caused defects that affected replication in addition to transcription and the amino acid requirements for the two processes were not identical , with substitution of N1335 inhibiting transcription to a greater extent than replication , and the opposite being the case for T1267 ( compare Figs 2 and 4 ) . RNA synthesis can be divided into different stages , including initiation and RNA elongation . Primer extension analysis was performed to narrow down which of these steps in RNA replication was affected by the substitutions . The RNA was analyzed using two primers that corresponded in sequence to nt 13–35 and 24–48 relative to the 3' end of the minigenome ( Fig 5A ) . Analysis of the RNA generated by the wt L protein using primer 13–35 showed RNA initiated at the +1 and +3 sites of the tr promoter ( Fig 5B , lanes 4 and 5 ) , consistent with previous results [6 , 21] . This primer also detected relatively high levels of RNA from +1 and +3 for each of the variants , except G1264A , for which only a very weak signal was obtained ( Fig 5B , lanes 7–12 , 5D ) . This shows that each of the variants , except G1264A , was capable of efficiently initiating RNA replication at +1 of the promoter and elongating the RNA far enough to be detected by the 13–35 primer . When RNA generated by the wt L protein was analyzed with primer 24–48 , RNA initiated from +1 of the promoter could be efficiently detected , but RNA initiated from +3 was only barely detectable because most of this RNA is terminated at approximately 25 nt ( Fig 5C , lanes 4 and 5 ) . Analysis of the RNA generated by the variant L proteins showed that in contrast to what could be detected by the 13–35 primer , replication product initiated from +1 could only be detected for the three variants capable of generating full-length RNA replication products: E1269A , S1333A and N1335A , but not the other variants ( Fig 5C and 5E ) . The levels of RNA detected for the E1269A , S1333A and N1335A variants were approximately 30% of wt levels , consistent with the levels of full-length RNA replication product detected by Northern blot analysis shown in Fig 4 . One possible explanation why the 13–35 primer could bind RNA when the 24–48 primer could not , could be that the replicative RNA was degraded . However , primer extension analysis showed no evidence of replicative RNAs with heterogeneous termini , as would be expected if it were degraded by an endonuclease ( see also Fig 3B ) , and it is difficult to rationalize why the 5' terminal fragments detected by the 13–35 primer would be protected if the RNA were degraded by a 3' to 5' exonuclease , given that there is no obvious RNA secondary structure in this region . Therefore , the fact that for each of the variants ( except G1264A ) , the replication products could be detected by the 13–35 primer at a higher level relative to wt than by the 24–48 primer , indicates that substitutions in CRV inhibited elongation of replicative RNA within the promoter region . As a complementary approach for examining the functions of CRV , we investigated how transcription and replication were affected by BI-D . As described in the Introduction section , BI-D is a small molecule inhibitor of the RSV polymerase . Resistance to BI-D maps to CRV ( Fig 1C ) , implying that CRV is either the site where the compound binds the polymerase , or is a domain that it alters allosterically , making BI-D a useful tool for studying the activities of CRV . Cells were transfected with the transcription-competent minigenome used in Fig 2 ( Fig 6A ) and support plasmids , and incubated with varying concentrations of BI-D from the time of transfection until harvest . Northern blot analysis showed that the level of input minigenome template produced by T7 RNA polymerase was similar for each reaction , confirming that BI-D did not cause significant cytotoxicity at the concentrations used ( Fig 6B ) . Northern blot analysis of the positive sense RNA products using the negative sense riboprobe showed that as the concentration of BI-D was increased , there was a decrease in the levels of full-length mRNA and antigenome , and an accumulation of smaller RNAs of heterogeneous size ( Fig 6C , lanes 4–9 , 6D ) . Thus , similarly to substitutions in CRV , BI-D inhibited transcription and replication . Primer extension analysis was performed using the 56–75 primer ( Fig 7A ) to examine how BI-D was inhibiting transcription . This analysis showed that as the concentration of BI-D was increased , there was a decrease in the level of RNA initiated at the gs signal ( +45 ) ( Fig 7B and 7F , gray bars ) . In parallel with this decrease , there was an appearance and increasing level of RNA initiated close to the antigenome initiation site ( Fig 7B ) . The most likely explanation for the appearance of this additional band was that it represented RNA initiated at +3 , but elongated beyond the end of the le region rather than being released after ~25 nt . To determine if this was the case , the RNA was analyzed using a primer that corresponded to nt 15–39 of the le promoter ( Fig 7A and 7C ) . This analysis showed that the only initiation sites near the 3' end of the le region were at +1 and +3 ( Fig 7C ) , confirming that the abnormal band detected with the 56–75 primer was initiated at position 3 . At lower concentrations of BI-D ( ≤ 50 nM ) there was no significant increase in +3 initiated RNA that could be detected by the 15–39 primer , however , there was a significant change in the level of +3 RNA that could be detected by the 56–75 primer , being completely undetectable in the absence of BI-D and clearly detectable , albeit as a faint band , at 25 and 50 nM BI-D ( Fig 7 , compare panel B , lanes 3–6 with panel C , lanes 4–7; panels F and G , white bars ) . This suggested that BI-D was causing an increase in elongation of the RNA initiated at +3 , allowing it to become detectable by the downstream 56–75 primer . As the concentration of BI-D was increased , there was an increase in the level of +3 RNA that could be detected by the 15–39 primer . This could indicate an increase in initiation at the +3 site , which could contribute to the increased levels of this RNA that could be detected by the 56–75 primer , or it could be due to increased elongation of the RNA allowing it to bind more efficiently to the 15–39 primer . To investigate these possibilities further , RNA samples were subjected to Northern blot analysis using gel electrophoresis conditions optimized for examining short RNAs and probed with an oligonucleotide that correlated in sequence to nucleotides 7–37 relative to the 3' end of the minigenome ( Fig 7A ) so that the lengths of RNAs initiated within the promoter region could be determined . In the absence of BI-D , the ~ 25 nt band , representing RNA initiated at +3 and released within the le promoter region was the only clearly detectable band . As the concentration of BI-D was increased , the levels of this band remained similar , but longer heterogeneous RNAs were detected ( Fig 7H ) . The bands ≥ 70 nt could also be detected with the 56–75 probe , consistent with these being le-CAT 1 read through RNAs ( Fig 7I ) . This analysis suggested that at least some of the RNA initiated at +3 was being elongated into the CAT 1 gene . Together , the primer extension and Northern blot data indicate that BI-D caused the polymerase to elongate the RNA beyond the promoter region , preventing efficient initiation at the first gs signal . Previous studies with compound BI-A , an inhibitor in the same class as BI-D , had suggested that it could inhibit mRNA capping as it resulted in inhibition of full-length mRNA synthesis and instead caused production of abortive transcripts of < 50 nt [25] . To determine if BI-D inhibited capping , the RNA was also analyzed by primer extension analysis using a reverse transcriptase capable of detecting a cap , and primers 56–75 and 91–113 , as described for Fig 3 . This analysis showed that , compared to the untreated control , 200 nM BI-D caused no detectable difference in the ratio between the bands representing capped and uncapped RNA , indicating that this concentration did not inhibit capping activity . However , the ratio of bands was altered in reactions that had been treated with 800 nM BI-D , with the upper band of capped RNA being much less intense relative to the lower band of uncapped RNA ( Fig 7D and 7E ) . These data show that BI-D could inhibit capping , in addition to affecting elongation within the promoter region . A direct comparison of RNA produced by the cap-defective R1339A variant L protein and by the wt L protein in the presence of 800 nM BI-D showed that the abortive RNAs that were produced were different in size ( Fig 7J ) . Thus , these data show that BI-D inhibited capping , as expected , however , they also suggest that an additional effect was to cause abnormal elongation of the RNA . To more clearly detect the effects of BI-D on replication , the inhibitor was tested using the replication-specific minigenome that lacked a gs signal after the promoter , as described in Fig 4 ( Fig 8A ) . This analysis confirmed that BI-D did not affect synthesis of the input minigenome ( Fig 8B ) and showed that BI-D inhibited replication , and instead caused production of abortive RNAs ( Fig 8C and 8D ) . Primer extension analysis was performed to examine RNAs initiated at positions 1 and 3 ( Fig 9A and 9B ) . As was the case with the le promoter , there was an increase in the level of RNA initiated from +3 , but in experiments with the tr promoter , there was a more pronounced decrease in the level of RNA that could be detected from +1 ( Fig 9C ) . Analysis of the small tr promoter specific RNA by Northern blotting with a probe corresponding to nt 5–35 of tr showed a similar pattern of RNA products as those synthesized from the le promoter ( Fig 9D ) . In this case , as the concentration of BI-D was increased , there was a clear decrease in the ~25 nt RNA and an increase in the levels of longer RNAs , again suggesting that BI-D was causing an increase in elongation of the RNA initiated at +3 . The data described above are consistent with the hypothesis that BI-D prevented the polymerase from releasing RNA within the promoter region . However , the data described above from cell-based assays do not exclude the possibility that the truncated transcripts detected on Northern blots arose because BI-D affected the stability of the full-length mRNA and replicative RNAs . Therefore , to confirm if BI-D could cause abnormal elongation of the ~25 nt transcript , RNA synthesis was examined in an in vitro RNA synthesis assay using purified recombinant RSV L-P complexes and an RNA oligonucleotide template , as described previously [6] ( Fig 10 ) . Because it is challenging to purify the RSV polymerase to a high concentration , we were not able to perform assays in which the initiation and elongation steps were uncoupled , to measure the elongation step of RNA synthesis specifically , neither was it possible to examine the kinetics of elongation . However , it was possible to determine if BI-D affected how far the polymerase could elongate the RNA by analyzing the radiolabeled products using denaturing gel electrophoresis , and then comparing the amounts of shorter and longer transcripts produced in the presence of BI-D , relative to those produced in its absence . The same assay design has been used previously to study how the elongation properties of the VSV polymerase are altered under various conditions [42–44] . RSV polymerase activity was reconstituted with purified L-P complexes , an RNA oligonucleotide consisting of nucleotides 1–40 of the tr promoter and NTPs , including [α32P]GTP . In previous studies , we have shown that in the in vitro assay , the RSV polymerase initiates at positions 1 and 3 , as it does in cells , with initiation from +3 being dominant to the extent that the readily detected bands are due to initiation at this position [6 , 45] ( see also Fig 11 ) . Analysis of RNA generated from the 40 nt template in the absence of BI-D showed two sets of products . At the top of the gel were bands > 40 nt that represent the products of 3' extension ( labeled 3'ext . ) . These arise when the template folds into a panhandle structure , with nucleotides 1 and 2 base pairing with nucleotides 13 and 14 ( or 15 and 16 ) , and the polymerase adds up to three nucleotides onto the 3' end of the RNA in a templated fashion , an activity that occurs in infected cells [6] . The level of 3' extension product was not significantly affected by increasing BI-D concentration . The bands ≤ 40 nt represent newly synthesized RNAs that had been initiated at the promoter . In the absence of BI-D the polymerase did not extend these transcripts to the end of the template , but instead paused or released RNAs along the length of template , with the major products being ≤ 25 nt in length ( Fig 10B ) . This correlates with what is found in cells , and shows that the polymerase is not able to efficiently elongate the promoter transcript beyond approximately 25 nt . Titration of BI-D into the reactions did not change the migration patterns of the RNA transcripts ≤ 25 nt ( indicating that it did not affect initiation site selection; see legend to Fig 11 ) , but resulted in production of lower amounts of shorter transcripts and the appearance of longer RNA transcripts that were extended to the end of the template ( Fig 10B , the asterisk indicates full-length products ) . This finding shows that BI-D increased the elongation properties of the polymerase and allowed it to overcome the trigger that normally causes release of the RNA after approximately ~25 nt . We used a similar approach to examine if CRV variant polymerases had elongation defects within the promoter . For this analysis , we focused on two of the L variants which showed different deficiencies in the minigenome system: G1264A and R1339A . G1264A was capable of initiating mRNA synthesis at the gs signal , albeit at a lower level than wt polymerase , indicating that it was capable of RNA synthesis ( Fig 3 ) . However , it was so deficient in RNA synthesis within the promoter region that RNA initiated from positions 1 and 3 was barely detectable by primer extension analysis , even with the primer that hybridized at positions 13–35 relative to the 5' end of the replicative RNA ( Fig 5 ) . In contrast , R1339A was capable of producing RNAs that could be detected with the 13–35 primer , but similarly to G1264A , was defective in producing full-length replication product ( Figs 4 and 5 ) . The polymerase complexes containing these variant L proteins were expressed and purified and compared by colloidal blue staining to ensure that similar the preparations contained similar amounts of L and P proteins ( Fig 11B ) . To examine their abilities to elongate RNA within the promoter region , we used a template consisting of the first 25 nt of the tr promoter ( Fig 11A ) , as this is the longest distance that the wt polymerase can synthesize RNA ( in the absence of BI-D ) and is more efficient than the 1–40 template ( probably because it is less prone to adopting secondary structures ) . Previously , we have shown that the polymerase tends to abort RNA synthesis after formation of the first dinucleotide [46] . Therefore , the RNA products from the same RNA synthesis reactions were analyzed using two different gel electrophoresis conditions that were optimized for resolving either the first dinucleotide initiated at position 3 ( Fig 11D ) , or 11–25 nt products , which would be a mixture of RNAs initiated at positions 1 and 3 , with the +3 products being dominant , as described above ( and being up to 23 nt in length ) ( Fig 11C ) . The background in the region of the gel which resolved RNA 3–10 nt in length was typically too high to clearly detect these products . Nonetheless , comparison of the levels of each of the 2 and 11–23 nt products generated by the variants to those generated by the wt polymerase allowed us to gain information regarding how efficiently the variants elongated the RNA relative to the wt polymerase ( Fig 11E ) . This analysis showed that the R1339A substitution caused subtle , but consistent , differences in the elongation properties of the polymerase , resulting in increased accumulation of products of 17–20 nucleotides , relative to the wt polymerase , and reduced accumulation of products of 21–23 nt ( Fig 11C , compare lanes 3 and 5 , and 7 and 9; 11E , note that the red dotted line represents the level of each RNA produced by the wt polymerase ) . This finding shows that the R1339A variant inhibited elongation , resulting in increased accumulation of shorter RNAs . Although the G1264A variant produced dinucleotide at a similar level as the R1339A variant , it produced lower levels of longer RNAs , indicating that this variant was even more deficient in elongation ( Fig 11C and 11E ) . The 3' extension products could be detected in reactions containing [α32P]GTP and were similar for each of the variants ( Fig 11C , lanes 3–5 , 11E ) . This result shows that all the variants were competent for RNA synthesis per se , and that CRV has relatively little impact on the 3' extension activity of the polymerase . However , the G1264A and R1339A substitutions inhibited elongation within the promoter region . The variants might also have been altered in their abilities to initiate RNA synthesis , but we were unable to determine if this was the case because the inability to detect the 3–10 nt products above a high background signal meant that we could not quantify total RNA synthesis relative to the wt polymerase .
The goal of this study was to determine if CRV of the L protein plays a role in RNA synthesis in addition to its anticipated role in mRNA capping . We show that this is the case , and that CRV influences polymerase elongation within the promoter region , with substitutions in CRV reducing elongation distance , and BI-D increasing it . Based on the previous characterization of the VSV PRNTase domain , and the fact that it is highly conserved between different nsNSVs , it was anticipated that at least some of the substitutions introduced into RSV L in this study would inhibit mRNA capping . Consistent with this , we found that polymerase variants with substitutions at G1264 , N1335 and R1339 , and to a lesser extent , T1267 and E1269 , were deficient in mRNA transcription ( Fig 2 ) . Primer extension analysis showed that these variants were capable of initiating RNA synthesis at the gs signal , but produced transcripts that either lacked a cap , or that were capped inefficiently ( Fig 3 ) . Thus , it appears that the enzymatic domain for PRNTase activity is shared between the rhabdo- and pneumoviruses , although there are some subtle differences , for example with respect to the significance of residue T1267 , which is essential for capping in VSV [24 , 30] . Single amino substitutions within CRV revealed that this region is also involved in regulating polymerase elongation . Analysis of the G1264A and R1339A variants in an in vitro RNA synthesis assay showed that they were less efficient than the wt polymerase at elongating RNA through the promoter region ( Fig 11 ) . This is consistent with primer extension analysis of RNAs produced in the minigenome system , which showed that for most of the variants , replication products could be detected with a primer proximal to the 5' end of the RNA ( primer 13–35 ) , but not a primer that hybridized further from the 5' end ( primer 24–48 ) ( Fig 5 ) . In a Sendai virus polymerase study , double or triple amino acid substitutions were introduced into the CRV [47] . Analysis of the phenotypes of these variants showed that in some cases , they were able to produce abundant levels of le RNA , but could not produce full length replication product [47] , similarly to what is described here . This suggests that a role of CRV in regulating elongation during the initial stages of replication may be a common feature of nsNSVs . Studies with BI-D showed that this compound also affected polymerase elongation through the promoter region . There are three possible mechanisms by which BI-D affected elongation: ( i ) it could have bound CRV directly , affecting CRV function , ( ii ) it could have bound elsewhere on the polymerase and allosterically altered the function of CRV , or ( iii ) it could have bound to CRV and elicited a long-range allosteric effect on the polymerase , which affected elongation . It might be necessary to resolve high-resolution structures of the RSV polymerase , both in isolation and in complex with BI-D , to distinguish between these possibilities . However , based on the fact that substitutions in CRV also affect polymerase elongation we propose that BI-D is most likely exerting an influence by either directly or indirectly altering the properties of CRV . However , whereas amino acid substitutions in CRV caused the polymerase to elongate less efficiently through the promoter , BI-D had the opposite effect , causing the polymerase to be abnormally efficient in elongation through the promoter . Analysis of RNA synthesis in vitro showed that whereas the polymerase typically could not elongate beyond ~ 25 nt , in the presence of BI-D this release signal was overcome , resulting in longer RNAs ( Fig 10 ) . This finding was consistent with what was detected in the minigenome assay . In this assay , BI-D prevented the polymerase from releasing the small le and tr transcripts after ~25 nt . In the case of the transcription-competent minigenome , this caused the polymerase to read into the first gene instead ( Fig 7B and 7H , and Fig 9D ) and inhibited mRNA initiation at the first gs signal ( Fig 7B ) . This is expected as the polymerase should only be able to reinitiate at the gs signal if it had previously released the le transcript . It has previously been shown ( and confirmed here ) that failure to cap the 5' end of an mRNA results in abortive RNA synthesis after ~40 nt [24 , 32 , 33] . It is thought that this is due to a checkpoint at this point in mRNA synthesis such that the polymerase can only elongate the RNA beyond ~40 nt if it can detect a cap . Thus , a role for CRV in elongation had already been suggested . However , the defect in elongation through the promoter region , described here is distinct from that which comes into play at the cap checkpoint . Data from the minigenome system showed that the T1267A variant was partially active in transcription and could produce full-length mRNAs , but was completely defective in replication elongation , whereas the converse was true for the N1335A variant ( Figs 2 and 4 ) . This indicates that the capping checkpoint and elongation within the promoter can be separated by single amino acid substitutions . Thus , the data presented here describe a novel function for CRV , namely to modulate elongation within the promoters so that the polymerase is neither hypo- nor hyper-processive . This might be of particular importance in the promoter regions , in which the polymerase must be particularly sensitive to a number of signals to be able to transcribe and replicate the genome . The mechanism by which CRV affects elongation is not known . Structural analysis of VSV polymerase indicates that the lobe containing CRV must undergo a conformational change and move away from the polymerization active site following initiation of RNA synthesis , to allow RNA egress [34] . It is possible that this lobe remains in contact with the RNA to modulate polymerase elongation . CRV could affect polymerase elongation distance by altering a number of factors , including elongation rate ( nt polymerized per unit time ) , and template and transcript affinity . Further experiments will be required to determine which of these factors is influenced by CRV . While the main finding of this paper is that CRV plays a role in polymerase elongation , the data obtained also indicate that this region has a role in initiation of RNA synthesis . For example , in the case of the minigenome with a le promoter , BI-D caused an apparent increase in initiation from +3 ( Fig 7; although as noted above , this could be due to more efficient elongation facilitating primer binding ) , and in the case of the replication-specific minigenome that contained the tr promoter at the 3' end , BI-D inhibited initiation at +1 ( Fig 9 ) . A role for CRV in initiation is not surprising given that in a pre-initiation state of L , a loop within this region protrudes into the polymerization active site ( Fig 1C ) [34] . Further studies are underway to more completely characterize the effect of CRV perturbations on initiation at the +1 and +3 sites , and to understand why there appears to be a promoter-specific effect . The data described above indicate how aberrant elongation induced by BI-D inhibited transcription by preventing the polymerase from being able to reinitiate at the first gs signal ( Fig 7 ) . They also indicate how the CRV substitutions inhibited replication , as inhibition of elongation within the promoter would not allow replication to occur ( Fig 5 ) . However , it is not clear exactly how BI-D inhibited RNA replication . In the case of the tr promoter , inhibition of initiation at +1 could be at least partially responsible ( Fig 9 ) , but in the case of the le promoter , BI-D had little effect on +1 initiation ( Fig 7 ) , and increased elongation caused by BI-D might have been expected to facilitate replication , rather than inhibit it . We speculate that BI-D inhibited replication by interfering , either directly or indirectly , with encapsidation . There are two models for encapsidation initiation in the nsNSVs . One is that the 5' end of the replicative RNA contains a sequence that binds to the N protein , thus nucleating the encapsidation process . The other model is that a replicase form of the polymerase binds to N and delivers it to the nascent replicative RNA . These two models suggest two possible mechanisms by which BI-D could inhibit replication . First , it is possible that polymerase elongation through the promoter region must be appropriately calibrated for encapsidation to occur and that BI-D interferes with that . Studies with human RNA-polymerase II have shown that the polymerase pauses after generating a 20–65 nt RNA to allow the recruitment of factors that can modulate gene expression [48] . It is possible that the RSV polymerase must pause in a similar fashion shortly after beginning RNA synthesis , to allow recruitment of N protein to begin the encapsidation process , and that by altering the elongation properties of the polymerase , BI-D prevents this , resulting in production of unencapsidated RNA . The second possibility is that CRV harbors an N binding site that is essential for encapsidation . If this were the case , BI-D might inhibit encapsidation by occluding this site , again resulting in production of unencapsidated RNA . If CRV were to contain an N binding site , this could explain why the R1339A substitution was so deleterious to replication , despite having relatively subtle effects on elongation in the in vitro assay . The results shown here demonstrate that BI-D has pleiotropic effects on the RSV polymerase . However , it is worth noting that the effects of BI-D on abortive transcript production described here differ from what was found in a previous study to examine the mechanism of action of the BI inhibitors [25] . In that study , the mechanism of action of BI-D was not examined , but BI-A and BI-E ( compounds related to BI-D ) were studied . BI-A caused accumulation of transcripts less than 50 nt in length , and BI-E caused accumulation of transcripts with a 5' triphosphate instead of a 5' cap [25] . Therefore , it was thought that this family of compounds inhibits capping . The results presented here are consistent with this: BI-D inhibited production of capped RNA ( Fig 7 ) , however , it resulted in production of abortive transcripts longer than those described previously ( Fig 7J ) . At this time , it is not clear if this is due to differences in the effects of BI-A versus BI-D on elongation , or if it is a reflection of the assays or concentrations of compounds that were used . For example , the assay used in the study by Liuzzi and coworkers used RSV infected cell fractions enriched for viral nucleocapsids [25 , 49] . While this assay had the advantage of examining polymerase activity on an encapsidated template , a significant drawback is that it did not allow the researchers to determine the provenance of the RNAs produced . Therefore , it is not known if the abortive transcripts detected in that study were derived predominantly from polymerase that had initiated at one of the viral promoter regions , or from the ten gs signals ( by polymerases already engaged on the template ) . The differences between the RNA sequences within these regions , and between the RSV genome and minigenome , could potentially affect polymerase elongation in the presence of compound . It was already appreciated that CRV of L would be a good target for antiviral drugs . Cap addition in the nsNSVs involves an enzymatically distinct mechanism than cellular capping , and cap addition is essential for viral gene expression [24–26 , 29 , 32 , 50] . The data presented here show that compounds targeting CRV might have the additional benefit of inhibiting replication , which could reduce the possibility of resistance mutations arising . Both defects would result in increased production of RNAs containing a 5' triphosphate , resulting in activation of RIG-I and augmentation of the interferon response , as shown previously [35] . Thus , the findings presented here add further support to the advantages of targeting CRV of L as a strategy for developing antiviral drugs .
HEp-2 cells ( ATCC ) were used to confirm expression of the mutant L proteins and were grown in Opti-MEM reduced serum media supplemented with 2% fetal bovine serum ( Invitrogen ) . BSR-T7 cells , that are engineered to constitutively express T7 RNA polymerase [51] , were used to reconstitute RSV RNA synthesis using a minigenome template . BSR-T7 cells were grown in G-MEM supplemented with 10% fetal bovine serum , 1X MEM non-essential amino acids solution , 6 mM L-glutamine and 1 mg/ml geneticin ( Invitrogen ) . Sf21 insect cells were used for expression of the RSV L-P complex for assaying RSV RNA synthesis in vitro . These cells were cultured in suspension in SF-900 II serum free medium ( Invitrogen ) . Three types of plasmid expression constructs were generated . To confirm mutant L proteins were expressed in mammalian cells , a codon optimized version of the L ORF of RSV strain A2 containing an N-terminal hexahistidine tag sequence was cloned under the control of the T7 polymerase promoter in pTM1 . Mutations in the GDNQ motif of L ( D811A ) and in CRV ( G1264A , T1267A , E1269A , S1333A , N1335A and R1339A ) were generated by QuickChange site-directed mutagenesis ( Agilent Technologies ) in a fragment of the L ORF subcloned into a pGEM T easy vector ( Promega ) . The mutated ORF fragment was sequenced and then substituted in the L ORF of pTM1 by restriction digest and ligation . For analysis of L protein function in the minigenome system , a similar panel of plasmids was generated , lacking the hexahistidine tag . For high-level baculovirus expression of an L-P complex containing substitutions in the GDNQ motif of L ( D811A ) and in CRV ( G1264A and R1339A ) , CRV mutations were introduced into an L ORF contained in a pFastBac Dual vector . The vector also contained the RSV A2 P ORF tagged with the tobacco etch virus ( TEV ) protease cleavage site and a hexahistidine sequence , as described previously [6] . Bacmids and recombinant baculoviruses were generated from the pFastBac Dual vectors using the Bac-to-Bac system ( Invitrogen ) , according to the manufacturer’s instructions . All DNA clones were sequenced to confirm the presence of mutations , and to ensure that no other changes were introduced during PCR-mediated mutagenesis . To confirm that each of the variant L proteins could be expressed in mammalian cells , HEp-2 cells were transfected with pTM1 plasmids containing histidine-tagged versions of wt or variant L plasmids . Cells were transfected with 0 . 5 μg of pTM1 P and 0 . 25 μg wt or mutant L using lipofectin ( Invitrogen ) . Cells were simultaneously infected with modified vaccinia virus Ankara-T7 ( MVA-T7 ) to express the T7 polymerase [52] . The cells were incubated at 37°C and at 48 h post-transfection , they were collected and disrupted with lysis buffer ( 750 mM NaCl , 50 mM NaH2PO4 , 20 mM imidazole , 0 . 5% NP40 ) . The cell lysates were clarified by brief centrifugation in a bench top centrifuge and incubated with Ni-NTA agarose ( Invitrogen ) with gentle rotation for 2h at 4°C The beads were pelleted by centrifugation and after two washes with lysis buffer , the Ni-NTA agarose was pelleted and resuspended in 2X SDS sample buffer . Proteins were analyzed by 10% SDS-PAGE and colloidal blue staining . Polymerase activities were assessed in the minigenome system using pTM1 vectors containing untagged versions of the L proteins . The minigenome templates that were used have been described previously [7 , 21] . Minigenome RNA synthesis and protein expression were reconstituted in BSR-T7 cells at 32°C , as described previously [20] , note that the cells were not co-infected with vaccinia virus or treated with actinomycin D . For samples treated with BI-D , BI-D was diluted in DMSO and added to the media at the time of transfection at the concentrations indicated . DMSO was maintained at a consistent concentration in each reaction . At 48 h post-transfection , cells were harvested to isolate total intracellular RNA . RNA samples were prepared using Trizol according to the manufacturer’s instructions ( Invitrogen ) . To detect full length mRNA and replication products generated from minigenomes , RNA samples were subjected to electrophoresis in 1 . 5% agarose-formaldehyde gels in MOPS buffer and subjected to Northern blot analysis as previously described [7] . The identities of each of the bands were determined by comparing the RNAs produced by minigenomes that have different arrangements of gs and ge sequences , and by examining RNAs with gene specific probes ( e . g . as described in [13] ) . In each experiment , the levels of input minigenome RNA were determined by probing Northern blots with a positive sense CAT riboprobe . To detect low molecular weight RNAs , RNA was analyzed by gel electrophoresis in 6% polyacrylamide gels containing 7 M urea in Tris-borate-EDTA buffer . The ladders used were either a low-range ssRNA ladder ( NEB ) , which was excised from the gel prior to transfer , stained with ethidium bromide for visualization with UV light and then realigned with the Northern blot , or the Decade Marker System ( Ambion ) which was Northern blotted directly . Direct comparison showed correlation between the two markers . Northern Blots were analyzed by autoradiography and phosphorimager analysis . 5'-ends of the RNA transcription and replication products were analyzed by primer extension as described previously [7 , 20 , 21] . In addition , a primer that hybridized at position 91–113 of the positive sense RNA products was used ( 5' TATCCAGTGATTTTTTTCTCCAT ) . RNA samples were reverse transcribed at 37°C using the Sensiscript RT kit ( Qiagen ) and radiolabeled primers . To detect capped RNA , reverse transcription reactions were performed using Thermoscript reverse transcriptase ( Invitrogen ) at 55°C . In cases in which the guanosine cap of transcription products were digested with pyrophosphatase , RNA was treated with Cap-Clip Acid Pyrophosphatase ( CellScript ) for 2 h at 37°C prior to primer extension . Primer extension reactions were analyzed by electrophoresis in 6% or 8% polyacrylamide gels containing 7 M urea in Tris-borate-EDTA buffer . End-labeled oligonucleotides corresponding in sequence to cDNA representing initiation from +1 and +3 of the le and tr promoters and the first position of the gs signal were used as markers . Primer extension reactions were analyzed by autoradiography and phosphorimager analysis . The RSV L and P proteins were expressed in insect cells using recombinant baculovirus and the L-P complex was purified as described previously [6] , except that the proteins were eluted from Ni-NTA beads with sodium phosphate buffer containing 250 mM imidazole and then dialyzed into 150 mM NaCl , 20 mM Tris HCl pH 7 . 4 , 10% glycerol , 1 mM DTT . To compare wt and variant polymerases , polymerase preparations were tested for RNase contamination by incubating them for 30 minutes , 1 h , and 2 h in a reaction buffer of the same composition used for RNA synthesis assays , with an end labeled RNA corresponding in sequence to the product of the 25 nt template , and the products were analyzed by denaturing gel electrophoresis , as described below . Polymerase preparations showing evidence of nuclease contamination were not included in the results presented here . RNA synthesis was reconstituted in vitro using a similar approach as described previously [6 , 7] , but with some modifications . A PAGE-purified RNA oligonucleotide corresponding to nucleotides 1–25 or 1–40 of the tr promoter was used as a template . Positions 23 , 32 , and 33 relative to the 3' end of the 1–40 template were changed to C , U , and U respectively to limit secondary structure . RNA ( 2 μM ) was combined with the purified L-P complex ( containing ~200 ng of L protein ) in RNA synthesis buffer ( 50 mM Tris HCl pH 7 . 4 , 8 mM MgCl2 , 5 mM DTT , 10% glycerol ) , with 500 μM rNTPs , and 10 μCi of [α-32P] GTP or [α-32P] ATP in a final volume of 50 μl . For reactions containing BI-D , BI-D was diluted in DMSO and added to reactions at the concentrations indicated . DMSO was maintained at a consistent concentration in each reaction . Reactions were incubated at 30°C for 2 h , followed by incubation at 90°C for 3 minutes to inactivate the polymerase , and then diluted in an equal volume of stop buffer ( deionized formamide containing 20 mM EDTA , bromophenol blue , xylene cyanol ) . To resolve 2 nt products , the reactions were analyzed on a 25% polyacrylamide gel containing 7 M urea in Tris-taurine-EDTA buffer . Longer RNA products were resolved by electrophoresis on a 20% polyacrylamide gel containing 7 M urea in Tris-borate-EDTA buffer . The nucleotide lengths of the RNA products were determined by comparison with a molecular weight ladder generated by alkaline hydrolysis of [γ32P] ATP end-labeled RNA oligonucleotides representing the anticipated 25 and 23 nt RNA products produced from the +1 and +3 sites , respectively or by alkaline hydrolysis of [γ32P] ATP end-labeled 1–40 RNA . Bands were visualized by autoradiography and quantified using Licor Image Studio Lite .
|
Respiratory syncytial virus ( RSV ) is a leading cause of respiratory illness in infants , elderly , and immunocompromised individuals , yet treatment is limited to supportive medical care . The large polymerase protein ( L ) of RSV is essential for transcribing viral mRNAs and replicating the genome , and is an attractive target for antiviral intervention . Previous studies from related viruses have shown that a region of the L protein co-transcriptionally adds a cap to nascent mRNAs . In this study , we showed that perturbation of this region of the RSV L protein resulted in capping failure , as expected . In addition , we found that this region altered how efficiently the polymerase could elongate the RNA , impairing recognition of essential signals . Thus , this region of the RSV L protein has more than one function during viral multiplication , making it a particularly appealing target for antiviral development .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"nucleic",
"acid",
"synthesis",
"molecular",
"probe",
"techniques",
"gene",
"regulation",
"rna",
"analysis",
"messenger",
"rna",
"dna-binding",
"proteins",
"northern",
"blot",
"polymerases",
"molecular",
"biology",
"techniques",
"rna",
"synthesis",
"dna",
"gel",
"electrophoresis",
"promoter",
"regions",
"chemical",
"synthesis",
"research",
"and",
"analysis",
"methods",
"rna",
"polymerase",
"primer",
"extension",
"electrophoretic",
"techniques",
"electrophoretic",
"blotting",
"artificial",
"gene",
"amplification",
"and",
"extension",
"proteins",
"gene",
"expression",
"primer",
"extension",
"molecular",
"biology",
"biosynthetic",
"techniques",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"biochemistry",
"rna",
"nucleic",
"acids",
"nucleic",
"acid",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"primer",
"extension",
"analysis"
] |
2017
|
RNA elongation by respiratory syncytial virus polymerase is calibrated by conserved region V
|
PRODH , encoding proline oxidase ( POX ) , has been associated with schizophrenia through linkage , association , and the 22q11 deletion syndrome ( Velo-Cardio-Facial syndrome ) . Here , we show in a family-based sample that functional polymorphisms in PRODH are associated with schizophrenia , with protective and risk alleles having opposite effects on POX activity . Using a multimodal imaging genetics approach , we demonstrate that haplotypes constructed from these risk and protective functional polymorphisms have dissociable correlations with structure , function , and connectivity of striatum and prefrontal cortex , impacting critical circuitry implicated in the pathophysiology of schizophrenia . Specifically , the schizophrenia risk haplotype was associated with decreased striatal volume and increased striatal-frontal functional connectivity , while the protective haplotype was associated with decreased striatal-frontal functional connectivity . Our findings suggest a role for functional genetic variation in POX on neostriatal-frontal circuits mediating risk and protection for schizophrenia .
Schizophrenia is a debilitating illness affecting about 1% of the population with a $62 . 7 billion estimated economic burden in the United States alone [1] . Twin studies and adoption studies have shown that disease risk is largely genetic , though characterizing the neural mechanisms underlying the disorder and mediating this genetic risk is a challenging process . Certainly , the genetic architecture of schizophrenia is complex and involves multiple genes . A candidate gene for schizophrenia that has received extensive investigation , with conflicting results , is PRODH , encoding proline oxidase ( POX ) . This enzyme , among other functions , is rate limiting in the conversion of proline to glutamate in mitochondria . PRODH along with its pseudogene are located at 22q11 is implicated in schizophrenia through a hemideletion syndrome at this locale , 22q11 deletion syndrome ( 22q11DS , Velo-Cardio-Facial syndrome , VCFS: OMIM #192430 ) , which may be the strongest known risk factor for schizophrenia besides having a monozygotic twin with the disorder [2]–[6] . Furthermore , a linkage peak at 22q11 has also been identified in a meta-analysis of families with multiple affect patients with schizophrenia [7] . Several recent studies have found genetic associations of schizophrenia with tag and functional single nucleotide polymorphisms ( SNPs ) in PRODH [8]–[11] , although this association has not been observed in all samples [12]–[18] . In addition to these genetic data , hyperprolinemia has been noted in populations of psychotic patients [19]–[26] and has been associated with neurological problems in 22q11 syndrome [24] . Recently , Bender et al [27] in a sample of European descent characterized several common functional genetic polymorphisms in POX that influence a wide range of enzymatic activity from reductions to below 30% of the reference sequence to increasing activity to 120% of the reference sequence . These findings offer the opportunity to study the impact of functionally characterized variation in POX on risk for schizophrenia . Statistical genetic association studies can provide a link between genes and complex polygenetic constructs like mental illness , but this approach does not illuminate the possible underlying pathophysiology impacted or the mechanisms of association . Here we used a staged investigation ranging from clinical genetic association to studies of the associated pathophysiology through a multimodal imaging approach to examine the impact of variation in PRODH on risk for schizophrenia and function and structure in human brain of neural circuitries implicated in the pathophysiology of schizophrenia . First , we identified variants in PRODH related to risk for schizophrenia in a family-based sample . Then , we used imaging genetic techniques to investigate genetically correlated physiology by identifying neural systems on which these variations impact in normal subjects who carry risk associated alleles or haplotypes [28] . Investigating susceptibility genes in the normal population allows for the investigation of complex trait susceptibility alleles without confounding factors of disease , drugs of abuse or treatment effects . In this study , we used Voxel Based Morphometry ( VBM ) for whole brain quantitative structural analysis , and a robust working memory task ( NBACK ) previously shown to be sensitive to genetic variation associated with schizophrenia susceptibility [29] , [30] .
We found that the functional variants in PRODH were associated with risk for schizophrenia . ( Table 1 ) Single marker analysis revealed that the minor allele for rs450046 was most strongly positively associated with schizophrenia in families . This allele has previously been associated with increased POX enzyme activity [27] . Conversely , the minor alleles of rs4819756 and rs2870983 , linked to decreased POX activity [27] , were significantly negatively associated with schizophrenia . Three-SNP sliding window haplotype analysis confirmed these separate individual SNP associations by showing change from positive associations to negative associations for haplotypes as they differentially included the 2 functional SNPs , rs2870983 and rs450046 separated by a tag SNP rs385440 , in the 3′ prime end of the gene . The CAC haplotype at these three SNPs was positively associated and TGT was negatively associated with schizophrenia , as seen clearly in Table 1 . Haplotype analysis comprised of the 3 functional SNPs ( hereafter referred to as the “functional risk or protective haplotypes” ) showed a significant positive disease association with the reference alleles for rs4819756 and rs2870983 and rs450046 minor alleles ( risk haplotype: GCC ) and negative association with haplotype of the minor allele for rs4819756 and the major alleles of rs2870983 and rs450046 ( protective haplotype: ACT ) . In a post hoc analysis to test if the SNPs were truly protective , we tested whether excess transmission to unaffected siblings of schizophrenic patients was observed with protective alleles . We found a trend for the major T allele of rs450046 towards over transmission ( z score = 2 . 142 , p = 0 . 077 ) . The results of the case control analysis , comparing unaffected siblings with unrelated controls , revealed a genotypic effect for the protective SNP rs4819756 , with a heterozygote odds ratio of 1 . 8 ( 95% C . I . 1 . 17–2 . 90 , OR p value = 0 . 009 ) and a minor allele homozygote odds ratio of 2 . 3 ( 95% C . I . 1 . 26–4 . 27 , OR p value = 0 . 007 ) . There was a significant association with the protective haplotype ( ACT ) and a negative association with the reference haplotype ( GCT ) ( Table 2 ) . The functional risk-associated haplotype ( GCC ) was associated with reduced regional gray matter volume in the neostriatum in the normal control sample ( 21 , 22 , −6 , z = 4 . 92 , P<0 . 05 whole brain corrected false discovery rate ( FDR ) ( Figure 2 A and B ) . In a post hoc exploratory analysis , the protective haplotype ( ACT ) revealed a non significant trend towards gray matter volume increase in the frontal lobes ( −21 , 53 , 8 , z = 3 . 97 , P<0 . 001 uncorrected ) ( Figure 2 A and B ) and occipital lobes ( 21 , −85 , 17 , z = 4 . 26 , P<0 . 001 uncorrected ) ( Table S1 ) . To further characterize these findings functionally , we analyzed the effect of the risk haplotype during working memory in the normal sample . Network activation patterns differed , with reduced blood oxygenation level-dependent ( BOLD ) signal in ventral lateral prefrontal cortex ( VLPFC , z = 3 . 18 , p<0 . 05 corrected ) and parietal lobes ( Brodmann area ( BA ) 40 , z = 3 . 52 , p<0 . 05 corrected ) ( Figure 2C ) relative to the reference haplotype . ( Table S2 ) Functional connectivity for the bilateral striatal seed regions showed trends for increased dorsal lateral prefrontal-striatal functional connectivity ( Figure 2D ) for the risk haplotype carriers . ( Table S3 ) In contrast , the protective haplotype carriers had decreased BOLD in bilateral striatum compared to reference ( −22 , 0 , 12 , z = 3 . 23 , p = 0 . 05 corrected ) ( Figure 3D ) , decreased recruitment of areas of early visual processing ( BA 18 , fusiform gyrus , BA6 , BA19 , z = 3 . 92 , p<0 . 05 whole brain corrected ) ( Figure 3C ) ( Table S4 ) and a decreased striatal-frontal functional connectivity , z = 3 . 75 , p<0 . 037 corrected for region of interest ( ROI ) ( Figure 3E and F ) ( Table S3 ) . Performance was not significantly different between haplotype groups ( Table S5 ) . Post hoc analysis including performance as a nuisance covariate had similar activation patterns . In order to assess whether the impact on POX function is a likely proximal cause of the systems-level findings reported above , we performed the same set of imaging analyses with haplotypes constructed from the 3 SNPs with no observed clinical association or known function . These did not yield any significant genetic effects for all phenotypes studied above .
Using a translational approach , we show that functional polymorphisms in PRODH , encoding POX , are linked to risk for schizophrenia and associated with alterations in prefrontal-striatal brain circuits involved in working memory and cognitive gating . We found a positive disease association with the SNP previously shown [27] to increase enzyme activity and a negative association between 2 polymorphisms that decreased POX activity . One of the strengths of the present study is that it assesses direct association of multiple functional variants with disease status , instead of relying on linkage disequilibrium ( LD ) between non-functional variants and unobserved unknown variants; this is especially important given that previous association studies did not test multiple functional SNPs . Another advantage of this approach particularly in this gene is that tag SNP mapping in this gene is complicated by the presence of a pseudogene in the same chromosomal region leading to low population SNP to SNP r2 [31] . Additionally , in the functional assays of POX , it has been observed that the enzymatic activity resembles the lowest activity polymorphism when there is more than one in the haplotype [27]; therefore , without testing functional haplotypes the association may not be observed . Despite the compelling finding that directionality of disease association corresponded to the effects on enzyme activity , our data do not exclude effects from variants in long range linkage disequilibrium . However , the diminished LD structure of this area of the genome and the results of our three SNP sliding window analysis make this unlikely , as do the results from our control experiment showing that nonfunctional variants had no significant effect on brain phenotypes . We also tested post hoc the hypothesis that putative protected alleles should be enriched in an unaffected sibling population . Indeed , we found that unaffected sibling status is positively associated with the POX variants that decrease the enzyme's activity . There are several reasons why this observation was difficult to appreciate in the previous literature . In the original finding , Lui et al [10] found a three SNP haplotype including one of the functional SNPs rs450046 positively associated with schizophrenia . Williams et al in two studies [15] , [16] looked first at only the 3′ end of the gene with rs383964 , rs372055 , and rs450046 in 677/679 case control sample from the UK and the Irish Republic and 55 trios from Bulgaria . While they tested directly the increased enzymatic activity variant , they may not have observed positive results due to not testing the other decreasing activity SNPs and/or the difference in the haplotype structure in case/control sample . Their second study [16] used a more extensive map of SNPs but used a 368/368 case/control pooling method that would make inferring functional haplotypes impossible . Fan et al [17] typed rs372055 in a small Chinese family trio sample and observed no association in their sample . Li et al [8] found rs385440 and two-marker haplotypes consisting of rs385440 and rs450046 ( G-G ) and rs372055 and rs385440 ( T-A ) associated with schizophrenia in a 528 trio and sibling pair sample . Abu Jamra and colleagues [18] found no association with the three putatively non functional SNPs , rs16983466 , rs372055 , and rs383964 , individually or in haplotypes in a German sample . Glaser et al [12] typed 4 SNPs ( rs450046 , rs383964 , rs372055 and rs385440 ) in a sample of 488 Bulgarian trios . They determined that only 2 SNPs could be used by calculating r2 in a subsample and therefore typed only rs383964 and rs372055 and were not able to find an association in their full sample . This fact may reflect a difference in linkage disequilibrium in the greater sample or confounding by the other functional variants . Recently in a genome wide case control association study , Sullivan and colleagues using a large schizophrenia treatment study sample found a positive association for one of the three SNPs in PRODH at p = 0 . 023 , but in the setting of a genome wide association study were unable to reach genome wide significance [11] . Our imaging findings provide further genetic evidence for fronto-striatal loop dysfunction as part of the pathophysiology of schizophrenia . This neural circuit has been previously defined in the animal [32] and human [33] literature as comprised of parallel , but interacting cortical-subcortical loops highly relevant for sensorimotor , emotional and cognitive performance . While the majority of our previous imaging genetic studies have focused on cortical working memory function “inefficiency” signal for increased activation of the dorsal lateral prefrontal cortex ( DLPFC ) with the recruitment of the VLPFC when performance is maintained , we are beginning to observe striatal to frontal “bottom up” processes also implicated in the pathophysiological processes . We previously have shown that polymorphisms in the PPP1R1B and in AKT1 linked to increased striatal frontal connectivity and decreased striatal volume were associated with genetic risk for schizophrenia [34]–[36] . Here , we again observe decreased striatal gray matter volumes in carriers of the risk haplotype in concert with increased functional connectivity with DLPFC . Interestingly , risk haplotype carriers with evidence for altered striatal structure and connectivity ( decreased striatal activations coupled with an increase of striatal to dorsal lateral prefrontal functional connectivity ) showed alterations in a working memory network that suggest a decreased engagement of the VLPFC and parietal cortex . Paralleling these results , Tan and coworkers , using the same functional paradigm as here in schizophrenia , found evidence for shifts to cortically hierarchical lower level working memory processes ( the VLPFC and parietal cortex ) in the setting of DLPFC dysfunction and related to genetic risk [37] . Tan et al ( 2007 ) further showed that an epistatic genetic interaction of risk variants for schizophrenia involving the cortical glutamate and dopamine systems revealed similar regions of activation as seen here in healthy controls carrying PRODH risk variants but in the opposite direction with the two risk variants in COMT and GRM3 showed increased engagement of the parietal cortex and VLPFC while PRODH risk haplotype showed an increase in striatal activity and decrease in activation of the parietal lobe and VLPFC [38] . This observed result may therefore be related to the interaction of the dopamine and glutamatergic systems through the POX enzyme but through the frontal-striatal loops . Of note , is that PRODH haplotypes groups had no significant performance differences and post hoc analysis including performance as a nuisance covariate did not change the activation patterns . Taken together , these convergent findings strongly support the concept that altered prefrontal-striatal processing is a genetic risk mechanism for schizophrenia [28] , [35] . In contrast to the risk haplotype , the protective haplotype had opposing correlations on striatal functional connectivity ( bottom-up processing ) in the performing the Nback task . In addition , the protective haplotype was associated with trend-level increases in frontal gray matter , a finding that is of interest since prefrontal cortex is one of the areas in which reduced gray matter volume is most consistently found in schizophrenia [39]–[43] . This finding , however , must be viewed with caution as it did not meet stringent statistical thresholds . If independently confirmed in future studies , this genetic-structural association could suggest that allele-associated increases in prefrontal grey matter volume or integrity could contribute to a protective effect through this intermediate phenotype . Though not hypothesized , we speculate that our observation of increased efficiency of early visual processing areas seen for the protective haplotype could correspond to long range prefrontal cortex control on the early visual processing stream previously shown to be a “top-down” mechanism relevant for attentional selection [44] and therefore be related to more efficient prefrontal control of attentional resources . At this time , while our experiments and control experiments suggest that it is the impact on POX function that is the proximal event linking genetic variation in PRODH to risk for schizophrenia , brain structure and function , the biochemical and cellular events mediating these observations are unknown . However , several mechanisms can be speculated about as a point of departure for further studies . First , L-Proline can directly modulate glutamatergic transmission in the brain and selective expression of a brain specific high-affinity proline transporter ( SLC6A7 ) occurs in a subset of glutamatergic synapses [45] . Dysfunction in glutamatergic neurotransmission has been widely implicated in the pathophysiology of schizophrenia [46] . In a mouse model of type 1 hyperprolinemia , disruption of glutamatergic transmission was observed together with altered cortical dopamine transmission and increased catechol-o-methyltransferase gene expression [47] . The protective SNPs decrease POX enzyme activity and may increase levels of the precursor and possibly decrease glutamate synthesis in sensitive neuronal populations . Secondly , POX plays a role in a redox shuttle in the mitochondria , while also interacting with the intrinsic and extrinsic pathway of apoptosis [48] , [49] . Mitochondrial dysfunction has been implicated in schizophrenia through expression data [50] , [51] and abnormalities with redox reactions have been associated with schizophrenia [52] . Finally , Proline is also a recognized osmolite and it has been shown recently that increased intracellular proline content modulates stability and aggregation of polyglutamate proteins [53] . While the patho-etiological evidence for a polyglutamate process in schizophrenia is limited [54] , several intracellular structural proteins have previously been associated with this disorder [55] . Molecular mechanisms unknown at this time may also contribute to this association , for example , serine racemase is not simply an enzyme to produce D-serine . D-amino acid oxidase does not simply degrade D-serine , but also degrade other D-amino acids . Our results appear inconsistent with earlier associations of overt hyperprolinemia and psychosis [20] , [21] , [25] . We believe this discrepancy may be more apparent than real . Notably , blood proline levels have a wide normal distribution [25] . We found that common variants that would presumably translate into relative increases of proline levels in the normal range were protective , while risk was associated with the common variants that decreased levels in the normal range; therefore , our results raise the possibility that increased risk of psychosis in the context of clinical hyperprolinemia [20] , [21] , [25] operate through different molecular mechanisms; for example , proline's ability to directly activate glutamatergic NMDA receptors at extremely high levels [25] or hypoglutamatergic neuronal state due to the lack of the metabolic precursor , Δ′-pyrroline-5-carboxylic acid or other cellular role of POX . Many other roles of POX in the cell are still being discovered [56]–[58] and will been illuminating for discovering the necessary molecular mechanism that POX increases susceptibility for schizophrenia . In conclusion , our data provide evidence that PRODH genetic variations that modulate the enzymatic activity of the POX enzyme contribute to the risk of schizophrenia through an impact on fronto-striatal processing . Our results may offer a novel target for the development of future therapeutic interventions .
For the genetics studies , 303 probands with schizophrenia spectrum disorders , their unaffected siblings , their parents and 370 controls were studied as part of the Clinical Brain Disorders Branch Sibling Study ( Protocol 95-M0150 , DRW PI ) at the National Institute of Mental Health . All subjects gave written informed consent before participation in accordance to the Internal Review Board of the NIMH . Only subjects of European ancestry were used in the present study to minimize effects of population stratification . An independent sample of psychiatrically and neurologically screened healthy individuals was used for the neuroimaging control data sets ( Table S5 ) . 4 common functional nonsynonymous SNPs and one synonymous SNP , previously associated with schizophrenia , were studied ( Table 1 ) . For additional coverage of the gene , fourteen additional tag SNPs were chosen from the Hapmap project CEU data for NCBI build 36 ( dbSNP build 126 ) [59] . For selection of the tagging SNP set , we used a 2–3 SNP aggressive tagging algorithm [60] as implemented in Haploview with minor allele frequency >0 . 05 and r2>0 . 80 [61] . One tag SNP failed on design of the genotype assay , 3 tag SNPs including one functional SNP failed due to poor amplification and one functional SNP and 3 tag SNPs failed due to non specificity because of the pseudo-PRODH gene detected by deviance from Hardy-Weinberg equilibrium >0 . 05 . We used standard methods to extract DNA from lymphoblastoid cell lines using the Puregene DNA purification kit ( Gentra Systems , Minneapolis , MN ) . PRODH genotyping was performed using the Taqman 5′-exonuclease allelic discrimination assay [62] obtained from Applied Biosystems ( Foster City , CA ) with primers and probes sets from Assays by Design . LD mapping and association test result plotting was performed using the R package snp . plotter [63] . We used a Bayesian method for haplotype construction ( Phase v2 . 1 ) [64] , [65] using the full Caucasian normal control sample ( N = 368 ) for the remaining 13 SNPs using 5000 iterations , one thinning interval and a burn-in of 1000 . We then recoded the inferred haplotypes in relationship to the 3 functional SNPs; rs450046 , rs2870983 and rs4819756 . 259 individuals were considered for imaging analysis if the functional haplotype were inferred at a greater than 90 percent probability . Only the reference , risk and one protective haplotype existed at appreciable population frequency rates >5% for haplotype imaging analysis . 17 of the 132 individuals with Nback data and 20 of the 124 individuals with VBM data were excluded due to missing data of a functional SNP which resulted in ambiguous haplotype assignment . As a control experiment , to ascertain whether the impact on POX function is a likely contributor to our findings , we performed this procedure for three SNPs that had no clinical association , no known function ( in particular , no impact on POX activity ) and r2<0 . 2 with any known functional SNP , rs9604911 , rs17743056 , and rs5746640 and analyzed these haplotypes exactly as stated above . 64 individuals were used in the VBM analysis and 55 were used in the Nback analysis . Family-based association testing was done via FBAT for single SNPs , 3-SNP sliding window haplotypes and haplotypes comprised of the three functional SNPs . All p-values were obtained via permutation testing using 1 , 000 replicates and are not adjusted for multiple testing . If the protective haplotype decreases risk for schizophrenia , it should be overtransmitted from the parents to unaffected siblings of schizophrenia . Therefore , we additionally performed a post hoc “sibcentric” analysis , by considering the transmissions to the unaffected siblings to be the “case” and non-transmission of parental alleles to be “pseudocontrols” of interest using FBAT ( n = 235 families ) and also examined them in a logistic regression using only one unaffected sibling per family single functional SNP genotypic case/control ( n = 167/350 ) method in STATA and with a 3 SNP functional haplotype analysis using the R package haplo . stats [66] . For structural analysis , scans for 92 individuals passed quality control and also had usable genotypes . Of those , 39 subjects were included in the risk compared to reference haplotype analysis and 85 subjects were included in the protective compared to reference haplotype analysis , Table S5 . Structural scans were performed on a 1 . 5 T GE scanner ( General Electric , Milwaukee , WI ) using a T1-weighted SPGR sequence ( repetition time , 24 msec; echo time , 5 msec , 256×256; field of view , 24×24 cm; flip angle 45° ) , with 124 sagittal slices at a thickness of 1 . 5 mm and an in-plane resolution of 0 . 94×0 . 94 mm . Optimized VBM [67] , [68] was performed using custom templates in SPM2 http://www . fil . ion . ucl . ac . uk/spm/; ( Wellcome Department of Imaging Neuroscience , London , UK ) as previously described [69] . Modulated and segmented grey matter images were smoothed with a 12-mm Gaussian kernel prior to statistical analysis . The Nback task has been shown to reliably activate the working memory network in normal subjects , shows an abnormal pattern in schizophrenia patients and their healthy relatives and is sensitive to genetic variation [30] , [70] . Briefly , we used a block design version of the Nback task where 100% of stimuli were both target and probes . A 0-back control task block where the subject simply responded with the current digit presented ( 1–4 in a diamond shaped box ) was alternating with the 2-back block in which the subject serially responded with numbers presented 2 previous ( “n” = 2 ) . Data were available for 108 subjects . 48 subjects were used in the risk compared to reference haplotype analysis and 103 subjects were used in the protective compared to reference haplotype analysis ( Table S5 ) . BOLD fMRI was performed on a GE Signa 3-T scanner using gradient echo EPI ( 24 axial slices , 6 mm thickness , 1 mm gap , TR/TE = 2000/28 ms , FOV = 24 cm , matrix = 64×64 ) . Images were processed on the first level using SPM99 with a 2>0 back contrast . Based on structural findings of the risk haplotype , which identified striatum , prefrontal connectivity with striatum was characterized using BOLD response maps with the Nback task by computing functional connectivity , as described previously [35] . Briefly , seed ROI for functional connectivity were placed within the combined bilateral caudate , head and body , and putamen as defined in the Wake Forest University brain atlas ( WFU ) ( www . fmri . wfubmc . edu ) [71] , and median BOLD time course in these regions were then correlated across all voxels , yielding a map of correlation coefficients in every voxel with the seed region activity . These functional correlation maps were then analyzed in a random-effects model in SPM for haplotype effects as below . Second level linear regression analyses as implemented using the general linear model in SPM2 of structural data and SPM99 for functional data [72] . Imputed PRODH haplotypes were covariates of interest entered into the model with the following nuisance covariates: for VBM: gender , total gray matter volume , age , and second-order polynomial age expansions and for Nback: age and gender . For all analyses , a stringent threshold of p<0 . 05 , corrected for multiple comparisons using false discovery rate , was used , either for whole brain or small volume corrected for a priori hypothesized ROI . ROI analysis for the Nback was based on known regions involved in functional compensatory mechanisms during Nback working memory as previously described in Tan et al . [37] and regions engaged in gene×gene interaction of glutamate signaling and dopamine systems [73] . In addition , the bilateral striatum ROI constructed as above was used to examine haplotype dependent BOLD signal . Based on our previous genetic findings [35] for striatal frontal connectivity , analysis was restricted to ROI of dorsal lateral prefrontal cortex ( DLPFC BA 45 , 46 , 9 ) . We have recently shown that this methodology affords excellent protection against false positives in imaging genetics [74] .
|
Schizophrenia is a major mental illness affecting 1% of the population . It is known that genetics plays a role in the disease susceptibility , and it is thought that the illness is a complex disorder involving multiple genes . We show that the schizophrenia susceptibility gene , PRODH , conveys its risk through a variation that increases its enzyme activity . We further show that protection is associated with variations that decrease enzyme activity and these protective variations are enriched in their unaffected siblings . We then used brain imaging of structure and memory function to dissect the risk and protective haplotypes differential effects , and found that the schizophrenia risk haplotype was associated with decreased striatal gray matter volume and increased subcortical to frontal lobe functional connectivity , while the schizophrenia protective haplotype was associated with trend-level increase of frontal lobe volume and decreased subcortical to frontal lobe connectivity . These findings indicate a new target for treating schizophrenia and characterize associated structural and functional deficits .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"neurological",
"disorders/neuroimaging",
"neurological",
"disorders/neurogenetics",
"radiology",
"and",
"medical",
"imaging/magnetic",
"resonance",
"imaging",
"genetics",
"and",
"genomics/medical",
"genetics"
] |
2008
|
Functional Polymorphisms in PRODH Are Associated with Risk and Protection for Schizophrenia and Fronto-Striatal Structure and Function
|
Cellular electrophysiology experiments , important for understanding cardiac arrhythmia mechanisms , are usually performed with channels expressed in non myocytes , or with non-human myocytes . Differences between cell types and species affect results . Thus , an accurate model for the undiseased human ventricular action potential ( AP ) which reproduces a broad range of physiological behaviors is needed . Such a model requires extensive experimental data , but essential elements have been unavailable . Here , we develop a human ventricular AP model using new undiseased human ventricular data: Ca2+ versus voltage dependent inactivation of L-type Ca2+ current ( ICaL ) ; kinetics for the transient outward , rapid delayed rectifier ( IKr ) , Na+/Ca2+ exchange ( INaCa ) , and inward rectifier currents; AP recordings at all physiological cycle lengths; and rate dependence and restitution of AP duration ( APD ) with and without a variety of specific channel blockers . Simulated APs reproduced the experimental AP morphology , APD rate dependence , and restitution . Using undiseased human mRNA and protein data , models for different transmural cell types were developed . Experiments for rate dependence of Ca2+ ( including peak and decay ) and intracellular sodium ( [Na+]i ) in undiseased human myocytes were quantitatively reproduced by the model . Early afterdepolarizations were induced by IKr block during slow pacing , and AP and Ca2+ alternans appeared at rates >200 bpm , as observed in the nonfailing human ventricle . Ca2+/calmodulin-dependent protein kinase II ( CaMK ) modulated rate dependence of Ca2+ cycling . INaCa linked Ca2+ alternation to AP alternans . CaMK suppression or SERCA upregulation eliminated alternans . Steady state APD rate dependence was caused primarily by changes in [Na+]i , via its modulation of the electrogenic Na+/K+ ATPase current . At fast pacing rates , late Na+ current and ICaL were also contributors . APD shortening during restitution was primarily dependent on reduced late Na+ and ICaL currents due to inactivation at short diastolic intervals , with additional contribution from elevated IKr due to incomplete deactivation .
The first step toward preventing sudden cardiac death is understanding the basic mechanisms of ventricular arrhythmias at the level of ion channel currents and the single myocyte action potential ( AP ) , using both experiments[1] and theoretical models[2] . Obtaining ventricular myocytes from human hearts for the study of arrhythmia mechanisms is both rare and technically challenging . Consequently , these mechanisms are usually studied with human channels expressed in non myocytes , or with non human ( rodent or other mammalian ) myocytes . However , these approaches have limitations , because functionally important accessory subunits and anchoring proteins native to ventricular myocytes[3] are absent in expression systems , and even among mammalian ventricular myocytes , ion channel kinetics[4] , [5] , [6] and consequently arrhythmia mechanisms are strongly species dependent . These issues limit the applicability of results from animal studies to human cardiac electrophysiology and clinical arrhythmia[7] . Measurements from undiseased human ventricular myocytes are a requisite for understanding human cell electrophysiology . Here , we present data from over 100 undiseased human hearts for steady state rate dependence , and restitution of the ventricular AP . Importantly , we also obtained essential new measurements for the L-type Ca2+ current , K+ currents , and Na+/Ca2+ exchange current from undiseased human ventricle . These previously unavailable data are critically important for correct formulation of mathematical models for simulation of electrophysiology and cellular arrhythmia mechanisms[8] . Using the new data together with previously published experiments , a detailed mathematical model of undiseased human ventricular myocyte electrophysiology and Ca2+ cycling was developed and thoroughly validated over the entire range of physiological frequencies . This model is referred to as the ORd ( O'Hara-Rudy dynamic ) model throughout the text . Model comparisons are conducted with the ten Tusscher-Panfilov ( TP ) model[9] , and the Grandi-Bers ( GB ) model[10] . The ORd model was used to describe cellular electrophysiology mechanisms specific to human ventricular myocytes . Underlying mechanisms of AP duration ( APD ) rate dependence and APD restitution were investigated . The effects of Ca2+/calmodulin-dependent protein kinase II ( CaMK ) on known ionic current and Ca2+ cycling targets were incorporated and studied . Early afterdepolarizations ( EADs ) and alternans were reproduced by the model . These are important arrhythmogenic phenomena that must be reproduced in order to study the mechanisms of cardiac arrhythmias in human and simulate clinical interventions such as drugs .
Data for ICaL in the undiseased human ventricle are from Magyar et al . [11] and Fulop et al . [12] ( both at 37°C , model validation in Figure 1C ) . Magyar et al . measured steady state activation , steady state inactivation , and the current voltage ( I–V ) curve . Fulop et al . measured recovery from inactivation . However , neither study separated Ca2+ dependent inactivation ( CDI ) from voltage dependent inactivation ( VDI ) . In fact , no published data are available which separate CDI and VDI in the undiseased or nonfailing human ventricle . For this measurement , we made new recordings in undiseased human ventricular myocytes at 37°C ( Figure 1 , current traces and white circles ) . Measurements were carried out with Ca2+ as charge carrier , allowing both CDI and VDI , or with Ba2+ as charge carrier , allowing only VDI . Results for Sr2+ were similar to those for Ba2+ . From holding potential of −60 mV , 75 ms steps were to potentials ranging from −40 to +50 mV ( 10 mV increments , 3 second interpulse interval , Figure 1A ) . 75 ms was sufficient for comparison of CDI and VDI , since it is in the early phase of decay in which CDI effects are most prominent[13] . Simulated current traces for CDI+VDI and for VDI–alone were similar to the experiments . Fractional remaining current ( FRC , at time t and voltage Vm , FRC ( t , Vm ) = I ( t , Vm ) /Ipeak ( Vm ) ) quantified the voltage and time dependence of inactivation for comparison between charge carriers . Figure 1B compares FRC for Ba2+ ( experiments top left , simulations right ) , and Ca2+ ( experiments bottom left , simulations right ) . With Ba2+ as the charge carrier , FRC monotonically decreased with increasing voltage at all times after peak current . This finding is consistent with dependence of inactivation on voltage alone . In contrast , for Ca2+ currents , FRC did not decrease monotonically with increasing voltage . Rather , Ca2+ current FRC curves appear to be effectively voltage independent . FRC for CDI+VDI was statistically different from FRC for VDI-alone at the more hyperpolarized potentials ( −20 to 0 mV , unpaired two-tailed t-test , p<0 . 01 ) . Ca2+ ions caused additional inactivation at these voltages , where VDI-alone was relatively weak . Since the only difference between Ca2+ and Ba2+ cases was the charge carrier , it follows that Ca2+ ions themselves were the source of the additional inactivation . This is evidence that currents carried by Ba2+ inactivate due to VDI only , while Ca2+ currents inactivate due to both VDI and CDI[14] . There is evidence that Ba2+ can cause ion dependent inactivation[15] . However , Ba2+-dependent inactivation was estimated to be 100-fold weaker than CDI[16] , and its effects were not appreciable in FRC experiments . To modulate VDI versus CDI in the model , the n gate was introduced , the value of which represents the fraction of channels operating in CDI mode . Under physiological conditions , ICaL inactivation is caused by a combination of both CDI and VDI . That is , n is between 0 ( all VDI ) and 1 ( all CDI ) . This model was based on experiments by Kim et al . [17] , where CDI was observed to function as a faster VDI , activated by elevated Ca2+ . Thus , both CDI and VDI are voltage dependent . The rate of decay in CDI mode is faster than that in VDI mode . The Mahajan et al . [18] and Decker et al . [19] ICaL models work similarly . The n gate is diagrammed in Figure 1E . Rates k1 and k-1 represent binding/unbinding of Ca2+ to channel bound calmodulin ( CaM ) [20] . There are four identical binding sites . Rates k2 and k-2 represent activation/deactivation of CDI mode ( black circle , asterisk ) , which occurs when all Ca2+ binding sites are occupied . We considered that the four Ca2+ binding transitions are in rapid equilibrium and solved the reversible two state reaction of Ca2+/CaM binding and CDI mode activation to obtain the differential equation describing the n gate ( Supplement Text S1 , page 10 ) . In both CDI and VDI modes , there are two weighted time constants for inactivation ( time constant weighting described in Methods ) . We determined time constants for CDI and n gate kinetics in an attempt to represent the shape and magnitude of the FRC measurements ( i . e . CDI reduced FRC , particularly at negative potentials ) . Time constants for VDI gates were determined by inactivation of Ba2+ currents ( Figure 1C ) . AP clamp simulations using the formulated ICaL model were similar to AP clamp experiments , where ICaL was defined as the 1 µM nisoldipine sensitive current ( Figure 1D ) . Specifically , currents showed spike and dome morphology . In experiments , peak current was −3 . 0 µA/μF . It was −2 . 7 µA/µF in simulations . Fast inactivation was 2 . 5 fold faster when phosphorylated by CaMK , similar to the Decker et al . dog ICaL model[19] and as measured experimentally[21] . The model for Ito was formulated based on newly measured experimental data . The measurements were from isolated undiseased human ventricular myocytes at 37°C ( Figure 2A , white circles ) , and were carried out with the addition of 1 µM nisoldipine to the standard bath solution ( see Methods ) to block ICaL . The holding potential was −90 mV . Currents were activated by a 300 ms step to various potentials . Inactivation time constants were determined from exponential fits to decay of these traces . To measure steady state inactivation , 500 ms steps from −90 mV to various potentials were followed by test pulses to 50 mV . Recovery from inactivation was determined at −90 mV , using P1/P2 pulses of 200 ms to 50 mV at varying interpulse intervals in a double pulse protocol . The time constant for activation was determined by fitting time to peak from a digitized current trace ( Amos et al . [22] , their Figure 12C , in undiseased human ventricle at 37°C; τa = 2 . 645 ms at Vm = +40 mV ) . Greenstein et al . [23] showed time to peak for hKv4 . 3 expressed in mouse fibroblast cells . The model provides a qualitative match to these data ( considering temperature and expression system differences ) . That is , the model activation time constant decreases from a peak value of 6 . 5 to 1 . 5 ms in near linear fashion with increasing voltage from −20 to 60 mV . The inactivation gate has two time constants , each with voltage dependent weighting . Inactivation kinetics and the I–V curve are accurate to the experimental data . A small divergence between simulations and experiments was observed at hyperpolarized potentials along the I–V curve ( simulated current was less than in experiments ) . This may be due to the fact that experimentally measured currents were small and difficult to measure at these potentials . In fact , current was not measureable in 21 , 11 , 5 , and 1 out of 23 cells at Vm = −40 , −30 , −20 , and −10 mV , respectively . Currents with zero values were not included in the experimental I–V averages . However , these currents were included in averages for obtaining steady state activation and steady state inactivation curves in the model . This prevented over representation of the window current ( small , appearing late during phase-3 of the AP , shown later ) . The conductance of the Ito model was set so that phase-1 behavior of the simulated AP would be similar to undiseased human endocardium experiments ( small in endocardium; maximum value ∼1 µA/µF ) . Measured endocardial APs showed rapid phase-1 repolarization , but did not show positive time derivatives during phase-1 ( true notching was generally not observed ) . Thus , model Ito conductance was set to the maximum level which did not violate these observations ( ∼1 µA/µF peak current at 1 Hz pacing ) . CaMK effects on Ito were incorporated based on measurements by Tessier et al . [24] and Wagner et al . [25] . As in Tessier et al . , CaMK shifted the voltage dependence of steady state activation 10 mV in the depolarization direction , and the time constant for development of inactivation was increased ( multiplicative factor fit to match the voltage dependent increase ) . Wagner et al . showed that the time constant for recovery from inactivation was affected by CaMK ( ∼2 fold faster ) . The INaCa model was formulated using measurements from undiseased human ventricular myocytes at 37°C ( Figure 2B , white circles ) . The model was based on the framework established by Kang and Hilgemann[26] , which allows for unlikely occurrence of inward Na+ leak , without Ca2+ exchange . The Hilgemann model shows Na+:Ca2+ exchange stoichiometry slightly greater than 3 . 0 , as has been observed by others[27] , [28] . Though the Hilgemann model is mechanistically novel in this way , it can still reproduce all Na+ , Ca2+ and voltage dependent properties observed by Weber et al . [29] in the nonfailing human ventricle . Compare Hilgemann and Weber data to our simulated reproductions in Supplement Figures S1 , S2 and S3 in Text S1 . As in the Faber-Rudy[30] and Hund-Decker-Rudy models[19] , [31] , we included 20% of the exchanger in the Ca2+ diffusion subspace[32] , [33] . The choice to include 20% in the subspace in human is validated based on its effect on the rate dependence of peak [Ca2+]i ( results in Supplement Figure S17 in Text S1 ) . Values above or below 20% disrupt the demonstrated correspondence of peak [Ca2+]i rate dependence with experiments ( see section on Na+ and Ca2+ rate dependence ) . The model for IK1 was constructed based primarily on new experimental data , measured at 37°C in undiseased isolated human ventricular myocytes as the 0 . 5 mM BaCl2 sensitive current ( Figure 2C , white circles ) . Current was elicited with steps from −90 mV to various potentials for 300 ms . The current that remained at the end of the steps was recorded as IK1 . Two gates were used in the model: RK1 , the instantaneous rectification gate , and xK1 , the time dependent inactivation gate . Importantly , previous models[9] , [10] , [34] , [35] have ignored both inactivation gating , and detailed [K+]o-dependence of IK1 ( exception , IK1 equations by Fink et al . [36] ) . There are nonfailing human ventricular measurements which we utilized to include these effects[37] , [38] . Steady state rectification was determined by dividing current by driving force , then normalizing . Rectification was shown to be [K+]o-dependent in the nonfailing human ventricle by Bailly et al . [37] . A linear shift in V1/2 for rectification toward more depolarized potentials with elevated [K+]o was incorporated , as was shown experimentally ( compare to Bailly et al . , their Figure 4B ) . Bailly also showed the voltage and [K+]o-dependence of inactivation . We introduced the time dependent xK1 gate , based on these data . As was shown experimentally , both V1/2 and the slope factor for inactivation depend linearly on [K+]o . The time constant for inactivation was based on measurements in nonfailing human ventricular myocytes by Konarzewska et al . [38] ( their Figure 1C ) . Conductance was observed to be in proportion to the square root of [K+]o in the human ventricle[37] . When assembled , the IK1 model demonstrated correspondence with the measured amplitude and rectification profile , and with Bailly data for [K+]o-dependence . As in Jost et al . [39] , IK1 was voltage dependent , but not pacing rate dependent ( Supplement Figure S4 in Text S1 ) . The model for IKr was constructed using experimental data measured in isolated undiseased human ventricular myocytes at 37°C ( Figure 3A , white circles ) . Measurements were carried out with/without addition of 1 µM E-4031 to the standard bath solution in order to obtain the difference current . Tail currents were elicited by stepping from −40 mV to various potentials for 1000 ms , and then stepping back down to −40 mV . The deactivation time constant was determined by fitting the tail current decay . The time constant for activation was found by stepping from −40 mV to various potentials for various durations preceding a step back to −40 mV . The rate with which the envelope of tail currents developed at different voltages was measured with an exponential fit to obtain the time constant for activation . Since this process was well fit by a single exponential , we made the fast and slow time constants in the model converge on the activation limb , at depolarized potentials . The steady state activation curve was determined from the I–V curve , after dividing by the driving force , assuming maximal activation at the time of peak tail current . Slow deactivation of IKr ( experiments and simulations , Figure 3B ) , suggests its participation in AP shortening during steady state pacing at fast rate and at short diastolic intervals during restitution; this hypothesis will be explored in a later section . The fast inactivation ( rectification , instantaneous in the model ) RKr gate was determined so that current profile matched experiments using a human AP voltage clamp ( Figure 3C ) . Important features of the experimental AP clamp trace that the model reproduced include 1 ) the early recovery phase , where approximately half maximal current appeared by the beginning of the AP plateau , followed by 2 ) quasi-linear current increase until peak current was reached during late phase-3 of the AP . Since enzymes used to disaggregate myocytes can significantly degrade IKr[40] , conductance was scaled to provide correct APD90 in control and with IKr block , measured in small tissue preparations . Indeed , APD90 was a function of IKr conductance ( parameter sensitivity , Supplement Figure S15 in Text S1 ) . As in undiseased human ventricle experiments[39] , IKr was voltage dependent , but not pacing rate dependent ( Supplement Figure S5 in Text S1 ) . Data from Virág et al . [41] , measured in isolated undiseased human ventricular myocytes at 37°C , were used to construct the model for IKs ( Figure 3D ) . The model has two gates: xs1 and xs2 . The xs1 gate is responsible for activation . Deactivation was controlled by xs2 . Activation/deactivation separation was based on the fact that activation was much slower than deactivation . Setting τx1>>τx2 at hyperpolarized potentials , where deactivation dominated , and τx2<<τx1 at depolarized potentials , where activation dominated , allowed for separation of these processes as two gates . As in the case of IKr , it is understood that IKs is damaged ( reduced ) by enzymatic disaggregation of myocytes[42] . Therefore , we used IKs specific drug block ( 1 µM HMR-1556 ) effects on APD90 , measured in small tissue preparations , to determine the correct conductance . Ca2+ dependence of IKs was incorporated based on measurements by Tohse et al . [43] . The effect of this dependence was negligible under physiological Ca2+ concentration conditions . Fast INa was formulated using nonfailing human ventricular data from Sakakibara et al . [44] ( Figure 4A ) . Since Sakakibara experiments were performed at 17°C , a temperature adjustment was used to obtain the final model equations , representing behavior at 37°C . The effect of temperature on steady state gating was shown by Nagatomo et al . [45] . For activation , V1/2 shift with temperature change from 23 to 33°C was +4 . 3 mV . For inactivation , the shift was +4 . 7 mV . We shifted V1/2 by twice these amounts , assuming linearity ( adjust to 37°C from data taken at 17°C , a change of 20°C; Nagatomo showed a change of 10°C ) . Time constants were adjusted to 37°C using Q10 . We set Q10 = 2 since Q10 was given as “about two” by Nagatomo . Hanck and Sheets[46] documented a shift in V1/2 with the passage of time after patch clamp commencement . For activation , the shift was −0 . 47 mV/min . It was −0 . 41 mV/min for inactivation . Sakakibara reported the time elapsed between patching and measurement for steady state activation and inactivation as between 10 and 20 min , ∼15 min for both . Thus , we reversed the time dependent shifts in V1/2 . CaMK effects on INa were based on available data[47] . We took into account the measured −6 . 2 mV V1/2 shift in steady state inactivation , the roughly 3-fold slowing of current decay , and the 1 . 46-fold slowing of recovery from inactivation . The non-temperature adjusted model I–V curve matches Sakakibara data at 17°C . We determined appropriate channel conductance at 37°C based on conduction velocity , and maximum dVm/dt . Conduction velocity in a one dimensional fiber simulation was 45 cm/s during 1 Hz pacing , consistent with available ( dog ) experiments[48] . It was 70 cm/s when stimulated from quiescence , consistent with in vivo measurements in nonfailing human hearts[49] . Maximum dVm/dt was 254 mV/ms in single cells at 1 Hz pacing , consistent with measurements from nonfailing human ventricular myocytes at 37°C ( 234±28 mV/ms ) [50] . Data used in the formulation of late INa were from Maltsev et al . [51] , measured in the nonfailing human ventricle ( Figure 4B ) , functionally defined in experiments and simulations as the Na+ current remaining after 200 ms from the onset of depolarization . Steady state activation was derived from the I–V curve ( current divided by driving force , then normalized ) . The time constant for activation of late INa was identical to that for fast INa . It is not possible to measure the time to peak for late INa because of the interfering effects of the much larger INa . However , in the model scheme , the measurement is irrelevant for the same reason . The hL gate is responsible for both development of and recovery from inactivation . The time constant for development was adjusted using Q10 = 2 . 2 , as measured by Maltsev et al . [52] ( hNav1 . 5 channels expressed heterologously ) . The time constant was voltage independent[51] . Maltsev et al . [51] reported a maximum late INa of −0 . 356 pA/pF in nonfailing human ventricular myocytes ( average current between 200 and 220 ms during step to −30 mV from −120 mV , their Table 2 , donor heart average ) . We scaled the Maltsev I–V curve to the donor value and used it to determine the model conductance . We do not consider fast and late Na+ currents to be separate channels . Rather , they have long been understood to represent different gating modes ( experiments[52] , and simulations by our group[53] ) , separated functionally in time . In experiments , and in simulated reproductions of experiments , late INa was functionally defined as the INa current persisting 200 ms after onset of depolarization . CaMK dependence was implemented ( −6 . 2 mV V1/2 shift in steady state inactivation , and 3-fold slowing of inactivation time constant , as measured[47] ) . The model for INaK was reformulated based on the work of Smith and Crampin[54] . The Smith and Crampin model includes more detail than standard formulations employed by other ventricular AP models[9] , [10] , [34] , [35] . Importantly , the Smith and Crampin framework includes [K+]i dependence and inputs for ATP and pH sensitivity . Here , we set ATP and pH values to normal physiological levels ( pH was dynamic when stated ) . Dynamically changing [K+]i is a known and meaningful pump regulator that is a functioning part of this model . High [K+]i ( combined with low ATP ) can make the pump reverse , bringing Na+ in , as has been observed in isolated hearts[55] . The Smith and Crampin model ( schematized in Supplement Figure S6 in Text S1 ) was adjusted to reproduce the basic findings of Nakao and Gadsby[56] , demonstrating [Na+]o dependence , [Na+]i dependence with high and low [Na+]o , and [K+]o dependence with high and low [Na+]o ( Supplement Figure S7 in Text S1 ) . To determine human ventricle appropriate conductance for INaK , we used [Na+]i-frequency data presented by Pieske et al . [57] as a target ( nonfailing human left ventricular myocytes at 37°C ) . The INaK formulation is based on known biophysical properties[54]; its behavior reproduces available experimental observations[56] ( Supplement Figure S7 in Text S1 ) . However , no direct measurement of INaK has been made in the nonfailing or undiseased human ventricle . To endow human ventricle specificity to INaK , our strategy was indirect; reproducing the rate dependence of intracellular Na+ concentration , [Na+]i , measured in the nonfailing human ventricle was the target . This choice assumes that the major role for INaK is maintenance of physiological [Na+]i . In the model , [Na+]i and its relative changes with pacing rate are controlled by INaK conductance ( ∼0 . 5 mM change per 20% change in INaK conductance , Supplement Figure S18 in Text S1 ) . In the absence of direct human ventricle INaK measurements , validation of the INaK formulation employs this relationship . Figure 5 shows a schematic diagram of the human ventricular AP model . The scheme was largely unchanged from the recent dog ventricular model by Decker et al . [19] . However , additional targets for CaMK were included , as described above , based on new findings . Currents were reformulated based on new undiseased or published nonfailing human experiments . These are colored gray in Figure 5 . Currents and fluxes colored white in the figure were based on human specific measurements of rate dependence of intracellular Na+ and Ca2+ concentrations ( [Na+]i and [Ca2+]i , respectively ) , which these currents/fluxes affect . Equations for currents and fluxes were not adopted from other human or animal models without substantive modification; all equations were reformulated with the exceptions of Ca2+ buffers , CaMK kinetics , and background currents , for which we used Decker et al . [19] formulations and adjusted conductances . Model equations for all major currents were completely reformulated ( i . e . fast INa , late INa , Ito , ICaL , IKr , IKs , IK1 , INaCa , and INaK ) . Relevant details precede equations in Supplement Text S1 . Microelectrode AP recordings from undiseased human ventricular endocardium at 37°C were used to validate basic human model AP characteristics . Figure 6A shows simulated APs and experimentally measured example APs for comparison during steady state pacing at the cycle lengths ( CLs ) indicated . We also compared simulated values for resting voltage , maximum voltage , and the maximum upstroke velocity , dVm/dt , with experiments ( Figure 6B ) . These comparisons were made for a single beat , stimulated from the quiescent state . For steady state rate dependence , we compared APD30–90 after pacing at different CLs ( Figure 7A ) . For restitution , we compared APD30–90 after steady state S1 pacing at CL = 1000 ms , followed by a single S2 extrasystolic stimulus delivered at various diastolic intervals ( DIs , measured relative to APD90 , Figure 7B ) . Model AP repolarization from 30 to 90% quantitatively reproduced this extensive dataset ( simulation results were within experimental error bars ) . Generally , electrotonic effects of tissue coupling were minor ( see Discussion and Supplement Figure S8 in Text S1 ) . The rate of repolarization in the model was gradual , as in experiments ( APD30–90 were well separated in time , Figure 7C ) . Other models repolarized more rapidly and late compared to these experiments ( simulations were all endocardial cell types ) . Koller et al . [58] measured dynamic restitution in the nonfailing human ventricle with monophasic AP electrodes . Following the Koller protocol ( explained in Methods ) , the human model matched Koller results ( Figure 7D ) . Simulations predict a bifurcation ( alternans ) at shortest DIs ( <90 ms ) , which is also observed in the experiments . Steady state rate dependence and restitution of the undiseased human ventricular APD were also measured in the presence of channel-specific blockers ( Figure 8 , white squares , see Methods for further details ) . In Figure 8 , drugs and applied doses are provided for each experiment . Simulated block was based on experimental dose-response measurements ( E-4031[59] , HMR-1556[60] , nisoldipine[61] , BaCl2[62] , ryanodine[63] , and mexiletine[64] , for block of IKr , IKs , ICaL , IK1 , Jrel , and late INa , respectively ) . Simulations matched these experiments; that is , simulation results were within experimental error bars . As pacing CL was decreased from 2000 to 300 ms , currents in the human ventricular AP model changed accordingly ( Figure 9 ) . Due to increased refractoriness at faster rates , maximum fast INa , late INa , and Ito were reduced . By contrast , peak ICaL increased , due in part to CaMK-phosphorylation induced facilitation[65] . IKr and IK1 were largely rate independent . Mild IKs accumulation[66] caused rate dependent increase in current . INaK became larger due to intracellular Na+ accumulation at fast pacing rates ( details below ) . INaCa , i , and INaCa , ss became more inward , in order to remove increasing Ca2+ . Changes in mRNA and protein expression across the transmural wall using undiseased human ventricles were measured[67] , [68] , [69] . Functional data for transmural changes in Ito were measured in nonfailing human ventricular myocytes[70] . These results were compiled to create a complete dataset for transmural differences between endocardial ( endo ) , mid-myocardial ( M ) , and epicardial ( epi ) cell types . We considered transmural differences in Nav1 . 5 , Cav1 . 2 , HERG1 , KvLQT1 , Kir2 . 1 , NCX1 , Na/K ATPase , Kv1 . 5 , RyR2 , SERCA2 , and CALM3 to be represented in the model by late INa , ICaL , IKr , IKs , IK1 , INaCa , INaK , IKb , Jrel , Jup , and CMDN , respectively . Whenever an expression ratio was not available , we chose unity . Using this analysis , models for M and epi cells were derived from the thoroughly validated endo model ( Figure 10A–10D; equations on page 19 in Supplement Text S1 ) . In Figure 10E1 , our experimental measurements for endo APD90 were scaled by M/endo and epi/endo APD90 ratios measured by Drouin et al . [50] and compared to simulations . Drouin experiments did not show results for CL<1000 ms . Epi simulations seem to deviate from Drouin experiments at faster pacing rates . However , epi simulations were consistent with nonfailing human epi experimental measurements at fast pacing rates ( CL <1000 ms ) recorded using optical mapping by Glukhov et al . [71] ( panel E2 ) . The rate dependence of simulated AP morphology in the different cell types ( Figure 10F ) was similar to Drouin recordings[50] . Relative shape and duration of simulated transmural APs were also consistent with those recorded by Glukhov et al . [71] from the heart of a 20 year old healthy human male ( Supplement Figure S9 in Text S1 ) . The transmural repolarization gradient direction was such that the pseudo-ECG T-wave was upright and rate dependent[72] as expected ( Figure 10G ) . Experiments from Guo et al . [73] in isolated nonfailing human ventricular endo myocytes showed EADs when paced very slowly ( CL = 4000 ms ) in the presence of the IKr blocker dofetilide ( 0 . 1 µM dose , ∼85% IKr block[74] ) . In Figure 11A , we display Guo experimental results and simulation results of the same protocol using the ORd model , and the GB and TP models ( all for endo cells at steady state ) . As in the experiment , the ORd model produced an EAD when paced at slow rate ( CL = 4000 ms ) with block of IKr ( 85% ) . Experiments and simulations both show a single , large EAD deflection . The GB and TP models failed to produce an EAD following the same protocol ( CL = 4000 ms ) , even with complete block of IKr ( 100% ) . EADs in the ORd model were caused by IKr block induced prolongation of the time at plateau voltages , allowing ICaL reactivation . When ICaL recovery was prevented , the EAD was eliminated ( inactivation gate clamping protocol , Figure 11B ) . This mechanism is the same as shown previously in other species[75] . Using data from nonfailing human ventricle , we validated rate dependent changes in concentrations of intracellular Na+ and Ca2+ . For [Na+]i changes with pacing rate , we used data from Pieske et al . [57] , measured in the nonfailing human ventricle , normalized to 0 . 25 Hz pacing rate ( Figure 12A ) . Reproduction of this curve implied that INaK magnitude was accurate ( INaK conductance controls intracellular Na+ , thus rate dependence of relative accumulation , Supplement Figure S18 in Text S1 ) . For Ca2+ , we used data from Schmidt et al . [76] , normalized to the value at 0 . 5 Hz pacing rate . A personal correspondence with senior author J . Gwathmey revealed that pacing in the experiments was for about 100 beats ( long enough to reach apparent steady state ) . Following this protocol , we showed the reduction in peak Ca2+ observed at the fastest pacing rates ( Figure 12B ) . However , at true steady state , peak Ca2+ increased monotonically with pacing rate ( shown in Figure 13 ) . Using Fura-2-AM fluorescence data measured in an undiseased isolated human ventricular myocyte at 37°C , we determined that the ORd model showed accurate intracellular Ca2+ decay ( Figure 12C and 12D ) . Time constant fits were a single exponential decay from time of peak Ca2+ . The decrease in decay time constant observed with increase in pacing rate is a measure of frequency dependent acceleration of relaxation , an importa10 . 1371/journal . pcbi . 1002061 . g012Figure 12Rate dependence of intracellular ion concentrations . Simulation results are solid lines . A ) [Na+]i versus pacing frequency ( normalized to 0 . 25 Hz ) . Experiments are from nonfailing myocytes ( Pieske et al . [57] , black squares ) . B ) Peak Ca2+ transient ( normalized to 0 . 5 Hz ) . Experiments are from nonfailing myocytes ( Schmidt et al . [76] , black squares ) . C ) Ca2+ transients from experiments ( Fura-2-AM ) and simulations . Results are normalized to illustrate the time course of decay . The arrow indicates pacing CL changes . D ) Frequency dependent acceleration of relaxation . Undiseased human experimental data are white circles . Simulations are the black line . nt validation of Ca2+ cycling . As pacing rate increased , so did the CaMK active fraction ( CaMKactive , Figure 13A , validated previously[31] , [77] ) . CaMK was important for controlling rate dependence of Ca2+ cycling in the model . In the absence of CaMK: Ca2+ transient amplitude was reduced , diastolic Ca2+ was elevated , JSR Ca2+ content and evacuation were rate independent , and Ca2+ reuptake ( Jup ) and release ( Jrel ) were severely blunted ( Figure 13B ) . Koller et al . [58] showed that in the nonfailing human ventricle ( in vivo monophasic AP recordings ) , APD alternans appeared at CLs <300 ms ( rates >200 bpm ) . The amplitude of APD alternans was ∼10 ms . These findings were reproduced by the human model ( APD alternans of 11 ms at CL = 250 ms , Figure 14 ) . Pacing at rates faster than 230 ms in the model caused 2 to 1 block ( i . e . failed APs every other beat ) , because APD began to encroach upon the pacing cycle length , leading to enhanced refractoriness of Na+ current due to incomplete repolarization . Since Koller measurements were performed in intact hearts , electrotonic coupling effects would have played a role . Therefore , simulations in a strand of 100 coupled endo cells were conducted to test whether alternans occurred in coupled tissue as well . Indeed , during CL = 280 ms steady state pacing , alternans developed in the multicellular fiber ( results shown in Supplement Figure S10 in Text S1 ) . As in Livshitz et al . [77] , beat to beat alternans in the Ca2+ subsystem were the cause of the APD alternans in the model . Longer APs coincided with larger Ca2+ transients . For steady state pacing at 250 ms pacing cycle length ( shown in Figure 14A ) , we found that clamping the subspace Ca2+ concentration to either the odd or even beat waveforms eliminated alternans , but clamping of the voltage , myoplasmic Ca2+ , ICaL , or INaCa did not eliminate alternans ( odd or even beat clamp , not shown ) . Cutler et al . [78] demonstrated that 30% SERCA upregulation eliminated alternans . Similarly , in our human model , a 20% increase in Jup magnitude eliminated alternans ( shown in Supplement Figure S11 in Text S1 ) . CaMK suppression also eliminated alternans in the model ( Figure 14A and 14B , gray traces ) . At slower pacing rates , APD was minimally affected by CaMK suppression . However , the peak Ca2+ concentration was markedly reduced , especially at faster rates ( Figure 14C ) . In order to describe the mechanisms underlying steady state rate dependence and restitution of the APD in the model , it is instructive to first systematically determine which currents participate in these phenomena . In Figure 15 , currents were plotted versus Vm during steady state and S1S2 restitution pacing for a variety of CLs and DIs , respectively . If I–V curves are CL or DI independent ( i . e . curves overlap ) , then that current did not participate in steady state rate dependence or restitution , respectively . Conversely , if I–V curves depended greatly on CL or DI , then that current played at least some role in these phenomena . As CL or DI decreased , fast INa , responsible for the maximum AP upstroke velocity and maximum Vm , was reduced ( see Figure 9 , and principles detailed in Luo and Rudy[79] ) . This is because shortened time at resting potential between beats prevents complete recovery from inactivation . Thus , at fast pacing rates , and short DIs , maximum Vm and upstroke velocity were reduced , explaining some of what follows . During steady state pacing , IKs was strongly rate dependent ( Figure 15A ) . The I–V curves were dramatically different at different pacing CLs . However , IKs was a relatively small contributor to the rate dependence of APD because IKs density in human ventricle is small under basal conditions ( no β-adrenergic stimulation ) , and changes relative to slow rate values produced minimal additional outward current at fast rates . Late INa , ICaL , INaCa and INaK also showed CL dependent changes during steady state pacing ( Figure 15A ) . INaK became more outward at fast rates . The changes in INaK were dramatic , and the current density was relatively large . Thus , INaK was an important contributor to APD shortening at fast pacing rates . Late INa became dramatically less inward at fast rates , making it a secondary contributor to APD shortening at fast rates . Changes in ICaL and INaCa opposed APD shortening at fast rates; these currents became more inward at short CLs . INaCa increased to match the increased Ca2+ extrusion burden . Importantly , ICaL increased despite reduced channel availability . ICaL inactivation gates recovered less between beats as pacing rate increased ( ∼20% less at CL = 300 ms compared to CL = 2000 ms ) . The same mechanism caused reduced late INa at fast rates ( availability at CL = 300 ms was ∼1/3 that at CL = 2000 ms ) . However , influences of increased CaMK facilitation combined with increased driving force ( reduced maximum Vm ) actually caused ICaL to become larger at fast rates . If Na+ is clamped to small values associated with slow pacing ( [Na+]i and [Na+]ss = 6 . 2 mM at CL = 2000 ms ) , preventing its accumulation at fast rates , INaK remains small and CL independent ( this mechanism is described later in detail ) , causing plateau voltages to become relatively CL independent . Thus , with Na+ clamp , ICaL changes with pacing rate are different than under control conditions . CL independent plateau voltages confer CL independence to the driving force for plateau ICaL . Na+ clamping reduced Ca2+ ( via INaCa ) which reduced activated CaMK and thus ICaL facilitation . An interesting consequence is that with Na+ clamp , ICaL changes with CL help to cause APD shortening at fast rates , whereas in control ( i . e . no Na+ clamp ) , ICaL changes with CL oppose APD shortening . During restitution , late INa , Ito , ICaL , IKs and INaCa showed DI dependent changes ( Figure 15B ) . Dramatically less inward late INa at short DIs helped shorten the APD . The mechanism was reduced availability due to residual inactivation at the start of the S2 beat . ICaL was reduced for the same reason . This was evident during the plateau . CaMK facilitation did not depend on DI because Ca2+ accumulation ( necessary for CaMK activation ) is slow , occurring only after long term pacing to steady state . Similarly , Na+ did not accumulate at short DIs , which kept INaK constant . Therefore , plateau potentials and ICaL driving force during the plateau were relatively DI independent . Just as in the case of Na+ clamp , these properties combined to allow reduced availability of ICaL at short DI to dominate the behavior . However , reduced maximum Vm increased the driving force during the time of peak ICaL , which caused peak current to generally increase at short DIs . At extreme DI of 5 ms , the slow AP upstroke ( i . e . reduced dVm/dt ) caused mild ICaL inactivation coincident with activation , so the peak current was reduced compared to DI = 10 ms . Changes in other currents ( Ito , IKs and INaCa ) , though nonzero , were relatively minor due to timing . DI dependent changes that increased or reduced current during phase-1 of the AP had little effect on final repolarization time . The exception is IKr . IKr is large enough that early spiking helped shorten APD at very short DIs ( detailed simulations follow ) . Steady state rate dependence of the APD was largely caused by accumulation of intracellular Na+ at fast rates . This is illustrated in Figure 16A . When [Na+]i and [Na+]ss were clamped to values from steady state pacing at CL = 2000 ms , APD lost much of its sensitivity to pacing rate and remained relatively long . Conversely , when the clamp was to [Na+]i and [Na+]ss from steady state pacing at CL = 300 ms , the APD remained relatively short at all rates . Pacing rate dependent [Na+]i and [Na+]ss changes are linked to the AP via INaK , which responds to [Na+]i levels . INaK increased with [Na+]i at fast rate . However it did not increase , regardless of the pacing rate , when [Na+]i and [Na+]ss were kept low ( Na+ at CL = 2000 ms; Figure 16C , right ) . Moreover , APD remained long at all CLs when INaK was clamped to its slow rate waveform ( not shown ) . Steady state APD rate dependence was not completely eliminated by Na+ clamp alone . That is , clamping [Na+]i and [Na+]ss to slow rate values did not cause APD curves to become absolutely flat with respect to CL , especially at fast pacing rates ( Figure 16A , shaded box CL = 300 to 700 ms , solid gray line ) . This signifies that other mechanisms are involved . When in addition to clamping [Na+]i and [Na+]ss to their slow rate values , we also reset the inactivation gates for late INa , and especially for both late INa and ICaL to their CL = 2000 ms values at the start of each beat , the APD curve flattened further at fast rates ( Figure 16A , dashed gray and dashed-dot-dot gray lines , respectively ) . Importantly , resetting these inactivation gates alone , without also clamping Na+ , had little effect on APD rate dependence ( Figure 16B ) . As described previously , without Na+ clamp , fast pacing caused late INa reduction and ICaL increase; the former helped while the latter opposed APD shortening . However , with Na+ clamp , both currents became less inward with fast pacing . Thus , resetting ICaL inactivation gates to slow rate values had different effects with , versus without Na+ clamping . Na+ clamp prolonged the APD . The prolongation and changed ICaL behavior after Na+ clamp rendered late INa and ICaL gate resetting more potent effectors of further AP prolongation; especially at fast rates where residual inactivation between beats was substantial . Rate dependent Na+ changes only occurred with the steady state pacing protocol due to slow ion accumulation after lengthy pacing regimes . For APD restitution , clamping [Na+]i and [Na+]ss to values from S1 pacing during the S2 beat did not affect APD ( Figure 16E ) . However , restitution was dramatically affected by resetting inactivation gates for late INa and/or ICaL to their S1 starting values at the start of the S2 beat ( Figure 16D ) . APD remained long for all DIs . Conversely , when late INa and/or ICaL inactivation gates were reset to S2 starting values for DI = 5 ms , APD remained short for all DIs . Again , resetting these inactivation gates to their slow rate values had only minor effects on steady state APD rate dependence ( Figure 16B ) . At very short DIs , IKr played an important role in APD restitution . In Figure 17A , the fast and slow time dependent deactivation gates ( xrfast and xrslow , respectively ) were reset to their value at DI = S1 = 1000 ms ( dashed gray line , compare to control solid black line ) . Deactivation of IKr is slow ( Figure 3B ) . For DI = S1 , deactivation was complete between beats . At short DIs , it was incomplete at the start of the S2 beat , enhancing IKr availability ( early IKr spiking , Figure 17B , bottom ) and outward current that contributes to APD shortening . The enhanced availability only mattered at very short DIs , because at these DIs APD was short enough that increased outward current during phase-1 of the AP affected final repolarization time . Changes to currents during later AP phases 2 and 3 ( during the plateau and early repolarization , e . g . late INa and ICaL ) , generally have greater impact on the APD . Early IKr spiking reduced maximum Vm , which affected all other currents , including late INa and ICaL . Several important differences exist between the ORd model presented here and other human models ( e . g . TP[9] and GB[10] models ) . Notably , model differences in the rate of repolarization and EAD formation were examined in direct comparison with experiments ( Figures 7C , and Figure 11A , respectively ) . Readers wishing to simulate the human ventricular AP have a choice of models . To help further differentiate the models , additional comparisons are shown in Figure 18 . Undiseased human ventricular measurements of steady state rate dependence of APD90 , 70 , 50 and 30 were accurately reproduced by the ORd model ( Figure 18A , same data as in Figure 7A ) . Rate dependence of APD90 is fairly accurate in the TP model . However , rate dependence of APD70 , 50 and especially APD30 are not accurate . The GB model repolarization rate is more accurate , but divergence from the measurements is large for APD30 . At fast pacing rates , GB model APD90 is accurate . Slow pacing APD90 is long compared with experiments ( at CL = 2000 ms , APD90 is ∼40 ms longer than in experiments ) . In addition , APD rate dependence does not plateau at CL = 2000 ms . In Figure 18B , the AP , major currents , and [Na+]i and [Ca2+]i were compared between models . Simulations were in single endo cells paced to steady state at CL = 1000 ms . Of note , the TP and GB models do not include late INa . The width of the ICaL peak and the morphology were model dependent . It was “cigar shaped” in the TP model . In the GB model , the ICaL peak was broad and poorly defined . The ORd model ICaL peak was sharp , as seen in undiseased human ventricle experiments ( AP clamp , Figure 1D ) . IKr was relatively small in the GB model , but shared a similar morphology with the ORd model . The TP IKr morphology is characterized by an early spike and a wider late spike . The IKs density in the TP model was much larger than in the other models ( ∼10 fold larger ) . Density and morphology of INaCa was model dependent . INaCa was smallest in the ORd model ( based on undiseased human measurements , Figure 2B ) . INaK was roughly 1 . 5 and 2 fold greater in GB and TP models , respectively , compared with ORd . The Ca2+ transient peak was much larger in the TP model than in the other models , which were similar to each other . The decay rate of [Ca2+]i was somewhat slower in the ORd model ( accurate to undiseased human measurements; Figure 12 panels C and D ) . Model [Na+]i was 7 . 2 , 8 . 2 , and 9 . 7 mM in ORd , GB , and TP models , respectively .
One of the most important aspects of the model is its close correspondence to experimental measurements of not only APD90 , but also to APD30 , 50 and 70 at all physiologically relevant pacing rates and for S1S2 restitution . This large pool of data has previously been unavailable . Accurate repolarization rate ( i . e . time between APD30 and 90 ) for the restitution protocol is crucial for simulating any phenomenon related to reentrant arrhythmia , where head-tail interactions determine refractoriness and vulnerability[83] . Use of new undiseased data for currents that are active during the plateau and phase-3 of the AP ( ICaL , INaCa , IKr and IKs ) contributed to the correct repolarization rate . The rate of repolarization and its effects on ICaL control EAD formation in this model , as in canonical EAD explanations [75] , [84] . Failure of the TP and GB models to reproduce EADs may be due in part to their accelerated repolarization rates ( Figure 7C ) . It may also be caused by inaccurate formulation of ICaL inactivation , developed in absence of the essential undiseased human data presented here . Due to the small amplitude and rapid deactivation kinetics of IKs in the human ventricle in absence of β-adrenergic stimulation , it does not play a major role in determining APD , APD rate dependence , or APD restitution under basal conditions[85] ( Figure 8 ) . This is in contrast to guinea pig ventricle , where slower deactivation and larger amplitude IKs make it the most important current for steady state APD rate dependence ( simulations[86] and experiments[87] ) . Phosphorylation by PKA in the case of β-adrenergic stimulation greatly enhances both the activation rate and amplitude of IKs[88] . With β-adrenergic stimulation , IKs plays an important role in steady state APD rate dependence[89] . Clearly , IKs is important under various circumstances – the AP repolarizes in human ventricle experiments even when IKr is blocked[85] , and clinical long QT syndrome type 1 is caused by IKs loss of function[90] . Typically , isolated myocyte patch clamp experiments underestimate IKs due to enzymatic degradation[42] . In ORd , the role of IKs was validated using small tissue preparations , where selective IKs block prolonged APD , but only very modestly under basal conditions ( no β-adrenergic stimulation , <15 milliseconds in experiments and simulations at CL = 1000 ms , Figure 8 ) . Block of IKr caused the most severe changes to the human AP ( rate dependence and restitution , Figure 8 ) . However , Supplement Figure S5 in Text S1 , and Figure 15A show that IKr is rate independent , as in experiments[39] and therefore was not responsible for causing APD changes with pacing rate . Rather , our simulations identified rate dependent changes in INaK secondary to [Na+]i accumulation as a primary cause of APD rate dependence ( Figure 16A , 16C ) . This finding is not new . Simulations in dog ventricle[19] , human atrium[91] , and in the GB human ventricle[10] models all led to this conclusion . However , findings from the Iyer human model[34] differ , at least in part , regarding this mechanism . In the Iyer model , [Na+]i affected APD rate dependence via INaCa , which is primarily outward at fast rates . Rate dependence in the TP model[9] is less [Na+]i dependent because , as Grandi discussed[10] , IKs is exaggerated . Experiments by Pieske et al . [57] investigated [Na+]i in heart failure , versus nonfailing human ventricular myocytes . Pieske experiments demonstrate that rate dependent [Na+]i accumulation is an important phenomenon in health and disease . However , additional experiments are needed to determine whether and how [Na+]i affects INaK and APD in human ventricle . In addition to INaK and INaCa ( both included in the ORd model ) , intracellular Na+ is also mediated by fluxes related to H+ , CO2 , and HCO3- homeostasis . Exchangers and cotransporters move Na+ ions down the electrochemical gradient in order to offset the cost of H+ , CO2 , and HCO3- pumping . Na+ rate dependent handling and consequently INaK should be affected by these processes , which were not explicitly included in the ORd model . In the absence of H+ , CO2 , and HCO3- fluxes , it is possible that the role of INaK might have been over estimated . It is important to address this because INaK and its response to Na+ accumulation was a major cause of APD rate dependence in the model . Thus , we performed simulations where H+ , CO2 , and HCO3- effects on Na+ were explicitly included , using Crampin and Smith equations[92] ( Supplement Figure S12 in Text S1 ) . Quantitative details of Na+ handling , INaK and APD rate dependence were affected when we included H+ , CO2 , and HCO3- handling processes . However , the qualitative outcomes were not affected . INaK increase with fast pacing , secondary to Na+ accumulation , was still the primary determinant of APD rate dependence during steady state pacing . Removal of the effects of Na+ accumulation on steady state APD rate dependence by clamping [Na+]i and [Na+]ss did not completely eliminate APD rate dependence . Especially at fast rates ( Figure 16A , shaded box CL = 300 to 700 ms , solid gray line ) , APD was not absolutely flat with respect to CL . APD rate dependence was largely unaffected by resetting inactivation gates for late INa , and/or ICaL to their slow rate values at the start of each beat ( Figure 16B ) . Interestingly , if these gates were reset while also clamping Na+ to slow rate values , the APD-CL curve became almost completely flat , even at fast rates ( Figure 16A , dashed gray and dashed-dot-dot gray lines , respectively ) . Thus , accumulation of Na+ and consequent effects on INaK is a major cause of APD rate dependence , however , not the only cause . Other currents also participate at fast pacing rates . Though the GB model[10] demonstrated the Na+/INaK/APD rate dependence mechanism , it did not show the additional effects of late INa and ICaL . The GB model cannot show these multi-factorial causes of APD rate dependence because it does not include late Na+ current ( Figure 18 ) , and because ICaL kinetics are inaccurate due to lack of experimental data . Due to charge conservation , accumulation of [Na+]i is associated with an equal reduction in [K+]i and a volume converted [K+]o increase in tissue clefts and interstitial spaces[93] . This can affect behavior by increasing IK1 ( its [K+]o sensitivity is included in this model ) , which depolarizes resting voltage and reduces excitability . However , our simulations represent experiments in an isolated myocyte in a large bath , where [K+]o is constant . Even in vivo , [K+]o is tightly controlled via regulation by the lymphatic system and kidneys . We showed that in contrast to steady state rate dependence , [Na+]i had no effect on APD restitution . Rather , restitution was primarily caused by the time course of recovery from inactivation of late INa and ICaL; processes which had little effect on steady state-rate dependence of APD ( absent Na+ clamp ) . At very short DIs , slow deactivation of IKr caused increased availability and spiking , which helped shorten the APD . APD rate dependence was caused primarily by concentration changes , while APD restitution was caused by gating kinetics . Previous studies have not made this important distinction between steady state rate dependence and restitution mechanisms in human . The role of ICaL and its inactivation kinetics in APD restitution reiterates the primacy of ICaL in determining basic physiological behaviors , highlighting the importance of the new ICaL experimental data , presented here , to model development and validation . A role for late INa in restitution could not have been hypothesized using TP or GB models , which have no late INa . The density of late INa was constrained in the ORd model by experiments from nonfailing human ventricular myocyte measurements by Maltsev et al . [51] , where the late current was measured 200 ms after the start of the voltage clamp step ( ∼0 . 35 µA/µF I–V curve maximum ) . The maximum late INa during the free running AP model was much smaller ( ∼0 . 15 µA/µF , about half the I–V curve maximum ) even at slow pacing rates , where late INa was largest . Late current is difficult to measure directly , and it is possible that the current density was overestimated due to selection bias . That is , late INa is small , and not all cells produced measureable late current ( 2 of 3 myocytes[51] ) . However , we consider the model density of late INa to be accurate based on model reproduction of experiments which consistently showed substantial APD90 shortening following application of 10 µM mexiletine in undiseased human myocardium ( 90% late INa block in simulations , Figure 8A ) . Previously published human ventricle AP models have not incorporated the CaMK signaling pathway . Our human simulations show , as in dog simulations[31] , [77] , that CaMK plays an important role in determining frequency dependence of Ca2+ cycling ( Figure 13 ) . The model also shows that the integrated electrophysiological consequence of CaMK effects on target channels is minimal . That is , CaMK suppression had only minor effects on APD rate dependence and AP morphology . At very fast pacing ( CLs <300 ms ) , the Ca2+ cycling consequences of CaMK phosphorylation were central to alternans formation . Suppression of CaMK eliminated alternans . CaMK related findings are in agreement with simulations using other models developed by our group[77] , models from other groups[94] , and experiments[95] . However , experiments showing the effects of pharmacological suppression of CaMK on rate dependent behaviors ( e . g . by Wehrens et al . [96] with KN-93 in rabbit ) should be performed in human ventricular myocytes to validate model predictions . The method used for implementation of the transmural cell types ( M and epi cell ) , based on the thoroughly validated endo cell framework , was simplistic . That is , we considered that channel conductance was proportional to transmural gradients in mRNA or protein expression for alpha subunits of ion channels . Only in the case of Ito were functional current measurement data available[70] . Staying within error bars for mRNA or protein data[67] , [68] , [69] , channel conductances were modulated so that the simulated transmural AP differences were consistent with experiments[50] , [71] . The effect of transmural heterogeneity of accessory β-subunits was not considered in the transmural cell type definitions . However , in the case of IKs , the KCNE1 β-subunit is transmurally heterogeneous . KCNE1 protein was about two times greater in M-cells compared to epi cells[69] . The presence of KCNE1 carries two important functional consequences 1 ) ∼5 fold slower activation and 2 ) ∼5 fold larger conductance[97] . Therefore , theoretically , twice as much KCNE1 in M-cells may increase the variable stoichiometry ratio of KCNE1 to alpha subunit KCNQ1[98] , slowing activation and increasing conductance . We conducted simulations to evaluate the influence of KCNE1 heterogeneity on IKs and the AP ( Supplement Figure S13 and related description in Text S1 ) . Due to the small amplitude of human IKs in the absence of β-adrenergic stimulation , implementation of KCNE1 heterogeneity did not appreciably affect the AP ( Supplement Figures S13 and S19 in Text S1 , where transmural APDs are shown to be relatively insensitive to changes in IKs conductance ) . Interestingly , the simulated effects of KCNE1 on activation rate and conductance offset one another . That is , slowed activation and larger conductance in M-cells yielded IKs current that was remarkably close to the control case . Similar results were found for epi cell simulations: the effects of faster activation and reduced conductance were offsetting such that their combined effect was minimal . Steady state rate dependence of APD and APD restitution were the focus of the simulations and experiments in this study . However , the time course of APD response to abrupt changes in pacing rate has been shown in human by Franz et al . [99] , and simulated in the TP model by Pueyo et al . [100] as a marker for arrhythmia risk . Simulations of APD accommodation in our model compare favorably to Franz experiments ( same pacing protocols used in experiments were used in the simulations , Supplement Figure S14 in Text S1 ) . Single exponential curves were fit to the time dependence of APD changes . For abrupt CL reduction from 750 to 480 ms , the time constant was 250 and 284 seconds in experiments and simulations , respectively . Time constants were 300 and 299 seconds in experiments and simulations , respectively , when CL was abruptly returned to 750 ms . When the CL reduction was more severe , from CL = 750 to 410 ms , the time constants were 252 and 165 seconds , in experiments and simulations , respectively . For return to CL = 750 ms , the time constants were 350 and 289 seconds , respectively . Pueyo used time to 90% accommodation to compare model with experiments demonstrating similar accuracy . Both simulation studies also show initial overshoot , or “notching” , as observed and described by Franz . As in Romero et al . [101] , we performed a sensitivity analysis to determine factors participating in important model outputs , including 1 ) steady state APD90 rate dependence ( Supplement Figure S15 in Text S1 ) , 2 ) S1S2 APD90 restitution ( Supplement Figure S16 in Text S1 ) , 3 ) rate dependence of maximum ( systolic ) [Ca2+]i ( Supplement Figure S17 in Text S1 ) , 4 ) rate dependence of [Na+]i ( Supplement Figure S18 in Text S1 ) , and 5 ) transmural cell type APD90 at steady state ( Supplement Figure S19 in Text S1 ) . The findings from our analysis were similar to those shown by Romero et al . using the TP human AP model[101] . That is , in ORd and TP models , IKr and ICaL affect APD90 while ICaL , INaCa , and INaK affect peak [Ca2+]i . One important difference is the role for IKs . A much larger role was played by IKs in the TP model ( ∼10 fold larger density than in other human models , Figure 18B ) . In the TP model , IKs is responsible for steady state rate dependence of the APD ( shown by Grandi et al . [10] ) . IKr conductance changes affect APD90 substantially in our model . This was expected , since IKr is the largest outward current ( also in experiments , Figure 8 , and in Romero's analysis using the TP model ) . Though IKr affects APD , it is not responsible for its rate dependence . Conductance changes in INaK did not substantially affect APD90 because INaK is a relatively small current . Yet , rate dependent changes in INaK ( secondary to Na+ accumulation at fast rate ) were the primary determinant of APD rate dependence . [Na+]i at different pacing rates , and thus its relative changes with rate , was by far most sensitive to INaK conductance ( Supplement Figure S18 in Text S1 ) . This supports our strategy for setting INaK conductance to reproduce rate dependence of [Na+]i in nonfailing human myocytes[57] . To keep the ORd model computationally efficient and parameters well constrained , the Hodgkin-Huxley formalism was used in formulating current equations . This choice was made as a design principal with the thought that interested users can modularly replace any current or flux with more detailed Markov formulations of mutation or drug effects as desired ( e . g . [53] , [66] ) . Similarly , intracellular Ca2+ handling can be modified ( e . g . more spatial detail , Markov ryanodine receptor implementation ) , or various signaling pathways and related effects on ion channels can be added ( e . g . [31] , [89] , [102] ) . The basic ORd model has 41 state variables . In the absence of CaMK and its effects on target currents and fluxes , the number of state variables is 31 . Exclusion of Markov models increases parameter constraint . It also prevents the system of differential-algebraic equations from being overly stiff . This promotes model stability and computational tractability . Using the rapid integration technique described in Supplement Text S1 ( Computational Methodology section ) , the model arrives at true and accurate steady state in under one minute of runtime ( ∼1000 beats are needed , depending on the CL , Visual C++ running on a desktop PC; more details in Supplement Text S1 ) . ORd equations are all smoothly varying functions , free of singularities and “if” conditionals . Thus , the model can readily be implemented in any of a variety of automated numerical integrators , such as Matlab ( The MathWorks Inc . ) , CellML ( http://www . cellml . org/ ) , CHASTE[103] , or CARP ( CardioSolv LLC . ) . Direct measurement of INaK in the undiseased or nonfailing human ventricular myocyte is lacking . Therefore , INaK was validated by reproduction of important biophysical properties ( Supplement Figure S7 in Text S1 ) , and by reproduction of [Na+]i rate dependence measured in nonfailing human ventricular myocytes ( Pieske et al . [57] , Figure 12A ) . However , independent and direct experimental measurement of INaK in undiseased or nonfailing human ventricular myocytes would provide additional support for the mechanistic conclusion that INaK changes secondary to Na+ accumulation at fast pacing rates is a major determinant of steady state APD rate dependence . This conclusion is consistent with several other modeling studies which proposed the same mechanism ( dog ventricle[19] , human atrium[91] , and human ventricle[10] ) . The relationship between [Na+]i , INaK and steady state APD rate dependence was robust . It was not disrupted by including the effects of Na+/H+ and Na+/HCO3- exchange fluxes on Na+ handling ( Crampin and Smith equations[92] , Supplement Figure S12 in Text S1 ) . Na+ accumulation and INaK response were not the only cause of APD rate dependence in the ORd model . At fast pacing rates ( CL = 300 to 700 ms ) , late INa and ICaL were also involved ( Figure 16A , and related discussion ) . Measurements of undiseased human endo APs were performed in small tissue preparations ( 1–3 gram pieces ) . This was to avoid possible enzymatic degradation of K+ channel proteins[40] , [42] , affecting currents and the AP . However , electrical loading in tissue subtly affects behavior[19] . We performed simulations using a multicellular fiber model to include loading effects , which had only minor consequences ( Figure S8 ) . APD was ∼275 ms in our human endo preparations at CL = 1000 ms , well matched by the model ( 273 ms ) . In vivo noninvasive electrocardiographic imaging of the activation-recovery interval , an indicator of the cellular epi APD , was ∼260 ms in healthy adults[104] . Human monophasic AP measurements are also in this range[58] . Measurements from Drouin et al . showed longer APDs ( ∼350 ms in endo cells on the cut transmural face at CL = 1000 ms ) . Having validated the endo model based on more than 100 of our own endo AP measurements , we thought it reasonable to use Drouin transmural APD ratios , rather than the uniformly longer APDs themselves , for validation of the transmural cell type models . The presence of M cell APs in the nonfailing human heart was observed by Drouin et al . [50] , and more recently by Glukhov et al . [71] . However , there is controversy regarding the M cell definition and its role in human . Our M cell model was based on data where the M cell was defined by its transmural location . The resulting simulated M cell AP corresponded with the “max cell” observed by Glukhov . Recently , Sarkar and Sobie developed a method for quantitative analysis of parameter constraint and relationships between parameters and target outputs in AP models[105] . We did not apply this analysis during model development . However , the extensive validation of channel kinetics and the emergent response of the AP to a variety of dynamic pacing protocols , used in development and validation of the model , ensures sufficient parameter constraint . The parameter sensitivity tests we performed were instructive , though relatively limited ( conductance changes only ) . Application of Sarkar and Sobie's analysis to our model is beyond the scope of this paper , but should provide worthwhile insights regarding inter-relatedness of processes in the human ventricle , in addition to formally testing parameter constraint .
During the last 15 years , undiseased hearts were donated for research in compliance with the Declaration of Helsinki and were approved by the Scientific and Research Ethical Committee of the Medical Scientific Board of the Hungarian Ministry of Health ( ETT-TUKEB ) , under ethical approval No 4991-0/2010-1018EKU ( 339/PI/010 ) . Data from 140 hearts were used in this study . Of these , 78 were from male donors ( 56% ) . The average donor age was 41 years old with standard deviation of 12 years . Tissue transport and ventricular endo preparations were performed as previously described[85] . Tissue was carefully pinned and placed in a modified Tyrodes superfusate ( in mM: NaCl 115 , KCl 4 , CaCl2 1 . 8 , MgCl2 1 , NaHCO3 20 , and glucose 11 , pH 7 . 35 , 37°C ) , and point stimulation was via bipolar platinum electrodes . Drug solutions were made fresh on the day of use . Drugs included in this study were , in µM: E-4031 1 , HMR-1556 1 , nisoldipine 1 , BaCl2 100 , ryanodine 5 , mexiletine 10 . Simulated application of these drugs was 70% IKr[59] , and 90% IKs[60] , ICaL[61] , IK1[62] , RyR[63] , and late INa[64] block , respectively . Tissue transport and myocyte isolation for the undiseased donor hearts were as previously described[85] . Myocyte isolation commenced immediately upon arrival in the laboratory , using the perfusion disaggregation procedure , previously described[85] . Data were obtained using conventional whole cell patch-clamp techniques . Micropipette fabrication and data acquisition were as described previously for undiseased donor heart[85] . Axopatch 200 amplifiers , Digidata 1200 converters , and pClamp software were used ( Axon Instruments/Molecular Devices ) . Experiments were performed at 37°C . The standard bath solution contained , in mM: NaCl 144 , NaH2PO4 0 . 33 , KCl 4 . 0 , CaCl2 1 . 8 , MgCl2 0 . 53 , Glucose 5 . 5 , and HEPES 5 . 0 at pH of 7 . 4 , and pipette solutions contained K-aspartate 100 , KCl 25 , K2ATP 5 , MgCl2 1 , EGTA 10 and HEPES 5 . The pH was adjusted to 7 . 2 by KOH ( +15−20 mM K+ ) . For L-type Ca2+ current measurement , the bath solution contained in mM: tetraethylammonium chloride ( TEA-Cl ) 157 , MgCl2 0 . 5 , HEPES 10 , and 1 mM CaCl2 , or BaCl2 , or SrCl2 ( pH 7 . 4 with CsOH ) . The pipette solution contained ( in mM ) CsCl 125 , TEA-Cl 20 , MgATP 5 , creatine phosphate 3 . 6 , EGTA 10 , and HEPES 10 ( pH 7 . 2 with CsOH ) . For Na+/Ca2+ exchange current measurement , the bath solution contained , ( in mM ) : NaCl 135 , CsCl 10 , CaCl2 1 , MgCl2 1 , BaCl2 0 . 2 , NaH2PO4 0 . 33 , TEACl 10 , HEPES 10 , glucose 10 and ( in µM ) ouabain 20 , nisoldipine 1 , lidocaine 50 , pH 7 . 4 . The pipette solution contained ( in mM ) : CsOH 140 , aspartic acid 75 , TEACl 20 , MgATP 5 , HEPES 10 , NaCl 20 , EGTA 20 , CaCl2 10 ( pH 7 . 2 with CsOH ) . Isolated myocytes from the undiseased donor hearts were used to measure the Ca2+ transient during point stimulation via bipolar platinum electrodes , indicated by Fura-2-AM , as was described previously[106] . Bath temperature was 37°C . For both experiments and simulations , we determined APD at 30 , 50 , 70 and 90% of complete repolarization ( APD30–90 , in ms ) . The start of the AP was the time of maximum dVm/dt . The time of APDX occurred once membrane voltage was X% of the resting value . Resting voltage was measured immediately prior to each paced beat . For APD rate dependence , pacing was to steady state . For APD restitution ( S1S2 , or static restitution ) , S1 pacing was at cycle length ( CL ) = 1000 ms . The S2 beat was delivered at variable diastolic intervals ( DIs ) , measured relative to APD90 . The dynamic restitution protocol was simulated as in experiments by Koller et al . [58] . Pacing was at a variety of rates ( 30 seconds at CLs from 230 to 1000 ms , no S2 beats ) . APD95 was plotted against DI ( where DI = CL – APD95 ) . Unlike static S1S2 restitution , the dynamic restitution protocol allows for more than one APD to be associated with a given DI . This is significant because bifurcation in the dynamic restitution curve is believed to be arrhythmogenic[107] . For all channels affected by CaMK , we created separate models for the fully CaMK phosphorylated channels , and the totally non phosphorylated channels . Then , based on the degree of CaMK activation ( CaMKactive ) , we determined the proportion of channels affected by CaMK . To calculate the CaMK affected current ( or flux ) , we added the weighted sum of fully affected and totally unaffected channels , using the proportionality . The model employed for CaMK activity was validated previously[31] , [77] . When measurements called for a gating process to be represented by both a fast and a slow process , we included separate fast and slow gates . However , we did not simply multiply fast and slow gates to modulate conductance as others have done previously . To do so allows the fast process alone to completely control deactivation/inactivation , and the slow process alone to completely control activation/recovery . Rather , since measurements of bi-exponential behaviors provide the relative weight of fast/slow processes , we modeled the measurements accordingly , and used the weighted sum of fast and slow processes . We computed the pseudo-ECG using a 1-dimensional model of the transmural wedge preparation[108] , [109] . In brief , the spatially weighted sum of the voltage gradient was determined at a point 2 cm from the epi end of a heterogeneous multicellular fiber , along the fiber axis . Cells 1–60 were endo , 61–105 were M , and 106–165 were epi . The stimulus was delivered to cell 1 . Cells 15 from both ends of the fiber were excluded from the gradient measurement due to confounding edge effects . Pacing was for 100 beats using steady state initial conditions from paced single cells . All model equations , hardware and software used , and rapid integration methods are provided in Supplement Text S1 . Model code can be downloaded from the research section of our website: http://rudylab . wustl . edu .
|
Understanding and preventing irregular heart rhythms that can lead to sudden death begins with basic research regarding single cell electrical behavior . Most of these studies are performed using non-human cells . However , differences between human and non-human cell properties affect experimental results and invoke different mechanisms of responses to heart rate changes and to drugs . Using essential and previously unavailable experimental data from human hearts , we developed and validated an accurate mathematical model of the human cardiac cell . We compared cellular behaviors and mechanisms to an extensive dataset including measurements from more than 100 undiseased human hearts . The model responds to pacing rate and premature beats as in experiments . At very fast pacing rates , beat to beat alternations in intracellular calcium concentration and electrophysiological behavior were observed as in human heart experiments . In presence of drug block , arrhythmic behavior was observed . The basis for these and other important rhythmic and irregular rhythm behaviors was investigated using the model .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"bioengineering",
"biomedical",
"engineering",
"biochemistry",
"medicine",
"biotechnology",
"ion",
"channels",
"arrhythmias",
"proteins",
"electrophysiology",
"biology",
"engineering",
"cardiovascular"
] |
2011
|
Simulation of the Undiseased Human Cardiac Ventricular Action
Potential: Model Formulation and Experimental Validation
|
In response to replication stress cells activate the intra-S checkpoint , induce DNA repair pathways , increase nucleotide levels , and inhibit origin firing . Here , we report that Rrm3 associates with a subset of replication origins and controls DNA synthesis during replication stress . The N-terminal domain required for control of DNA synthesis maps to residues 186–212 that are also critical for binding Orc5 of the origin recognition complex . Deletion of this domain is lethal to cells lacking the replication checkpoint mediator Mrc1 and leads to mutations upon exposure to the replication stressor hydroxyurea . This novel Rrm3 function is independent of its established role as an ATPase/helicase in facilitating replication fork progression through polymerase blocking obstacles . Using quantitative mass spectrometry and genetic analyses , we find that the homologous recombination factor Rdh54 and Rad5-dependent error-free DNA damage bypass act as independent mechanisms on DNA lesions that arise when Rrm3 catalytic activity is disrupted whereas these mechanisms are dispensable for DNA damage tolerance when the replication function is disrupted , indicating that the DNA lesions generated by the loss of each Rrm3 function are distinct . Although both lesion types activate the DNA-damage checkpoint , we find that the resultant increase in nucleotide levels is not sufficient for continued DNA synthesis under replication stress . Together , our findings suggest a role of Rrm3 , via its Orc5-binding domain , in restricting DNA synthesis that is genetically and physically separable from its established catalytic role in facilitating fork progression through replication blocks .
The replication machinery is constantly at risk of encountering obstacles such as protein-DNA complexes , DNA secondary structures , transcribing RNA polymerases , RNA-DNA hybrids , and DNA damage , all of which can block fork progression . If these structures cannot immediately be resolved the paused fork may eventually collapse as replisome components become irretrievably inactivated . The 5’ to 3’ DNA helicase Rrm3 is a member of the Pif1 family , which is conserved from yeast to humans [1 , 2] . Saccharomyces cerevisiae RRM3 was first discovered as a suppressor of recombination between tandem arrays and ribosomal DNA ( rDNA ) repeats [3] . Without Rrm3 , extrachromosomal rDNA circles accumulate , suggesting a role in maintaining rDNA repeat stability , and cells accumulate recombination intermediates at stalled replication forks , which has lead to the suggestion that Rrm3 facilitates DNA unwinding and the removal of protein blocks to help fork convergence during replication termination [4–7] . Additionally , replication fork pausing has been observed in the absence of Rrm3 at centromeres , telomeres , tRNA genes , the mating type loci , inactive origins of replication , and RNA polymerase II-transcribed genes [3 , 5 , 6] . The mechanism by which Rrm3 aids fork progression is poorly understood , but it is thought that the ATPase/helicase activity of Rrm3 facilitates replication through protein blocks and may also be able to remove RNA transcripts [5 , 8] . Within each rRNA coding region are two intergenic spacers that contain termination sites that are bound by the replication terminator protein Fob1 to promote fork arrest and to prevent unscheduled transcription [9–11] . Termination site function also requires the intra-S phase checkpoint proteins Tof1 and Csm3 , which form a complex with the replisome and antagonize Rrm3 function [12 , 13] . It is thought that Rrm3 removes Fob1 and other non-histone proteins from DNA before the replication fork encounters them . This ability of Rrm3 to promote replication fork progression is dependent on its catalytic activity [7] . Further supporting a role of Rrm3 in fork progression are synthetic fitness defects or lethality between rrm3Δ and mutations that disrupt genes involved in maintaining the integrity of stalled forks , including rad53Δ , mec1Δ , srs2Δ , sgs1Δ , mrc1Δ , and rtt101Δ [5 , 14–16] . Rrm3 possesses an N-terminal PCNA-interacting peptide ( PIP ) box , associates with the replication fork in vivo and is hyperphosphorylated by Rad53 under replication stress [1 , 17 , 18] . The replication damage that arises in the absence of Rrm3 causes constitutive , Mec3/Mec1/Rad9-dependent activation of the checkpoint kinase Rad53 [5 , 17 , 19 , 20] . As a result , Dun1 kinase is activated , leading to degradation of the ribonucleotide reductase ( RNR ) inhibitor Sml1 and an increase in the dNTP pool [21 , 22] . This increased dNTP pool has been associated with enhanced DNA synthesis in hydroxyurea ( HU ) in chromosome instability mutants [22] . Here we show that cells lacking Rrm3 fail to inhibit DNA replication in the presence of HU-induced replication stress and that this failure is not caused by the increased dNTP pool resulting from constitutive DNA-damage checkpoint activation . This novel replication function of Rrm3 is independent of its ATPase/helicase activity and , thus , distinct from Rrm3’s established catalytic role in facilitating fork progression through replication blocks . Instead , we have identified dependency on a novel functional domain in the Rrm3 N-terminus that we find is involved in binding the Orc5 subunit of the origin recognition complex ( ORC ) . Together with our finding that Rrm3 associates with a subset of replication origins , this suggests that Rrm3 may control DNA synthesis by controlling origin activity . Quantitative mass spectrometry and genetic analyses further implicate Rad5-dependent error-free DNA damage bypass and Rdh54 translocase as novel repair mechanisms for DNA lesions that result from inactivating the catalytic activity of Rrm3 , whereas these DNA repair factors are dispensable when the Orc5-binding domain is disrupted , leading us to conclude that the types of DNA lesions that result from the inactivation of the two independent Rrm3 functions are distinct .
In the absence of Rrm3 cells accumulate replication pause sites at the rDNA locus , in tRNA genes and at centromeric regions , as well as many other sites throughout the genome [5 , 6 , 15] . To identify DNA metabolic pathways that deal with stalled forks , we sought to identify proteins whose association with chromatin changed in the absence of Rrm3 using stable isotope labeling by amino acids in cell culture ( SILAC ) -based quantitative mass spectrometry [23 , 24] . We extracted the chromatin fraction from nuclei purified from a mixture of wildtype and rrm3Δ cells grown in the presence of heavy or light isotopes of arginine and lysine , respectively ( Fig 1A and 1B ) . Across chromatin fractions from three biological replicates we identified 490 peptides from 137 different proteins , with the abundance of 11 proteins changing significantly in at least two of the three replicates ( Fig 1C ) . The largest change in chromatin association was a 5 . 1-fold increase ( p<0 . 001 ) of Rad5 , which belongs to the SWI/SNF family of ATPases and defines an error-free pathway for bypassing replication-blocking DNA lesions [25–28] . The increase in Rad5 was followed by smaller , but significant , increases for Top2 ( 1 . 9-fold , p<0 . 01 ) , a type II topoisomerase that is important for the decatenation of replication intermediates , and Rdh54 ( 1 . 8-fold , p<0 . 01 ) , a chromatin remodeler with a role in homologous recombination that is still largely unclear . Like Rad5 , Rdh54 is a member of the SWI/SNF family of ATPases; it possesses translocase activity on double-stranded ( ds ) DNA and has been shown to be capable of modifying DNA topology , especially in chromatinized DNA [29–31] . We observed significant decreases in chromatin association for the Rsc1 subunit of the chromatin-structure-remodeling ( RSC ) complex ( 2-fold , p<0 . 01 ) , the Mcm4 subunit of the minichromosome maintenance ( MCM ) replicative DNA helicase ( 1 . 9-fold , p<0 . 01 ) , and the catalytic subunit Hda1 of the histone deacetylase ( HDAC ) complex ( 1 . 7-fold , p<0 . 01 ) . Upon treatment with HU , which induces replication stress by reducing the nucleotide pool [21] , Rdh54 abundance in the chromatin fraction of the rrm3Δ mutant increased the most ( 2 . 6-fold , p<0 . 01 ) whereas the histone deacetylase Set3 and the Rsc9 subunit of the RSC chromatin remodeling complex saw the largest decreases ( 2 . 7-fold , p<0 . 05 ) ( Fig 1D and 1E ) . The complete list of proteins that underwent significant changes in the HU-treated or untreated rrm3Δ mutant , including the FANCM-related Mph1 helicase , the recombination factor Mgm101 , and the cohesin components Smc1 , Smc3 and Scc3 , is provided in S1 Table . Rrm3 helicase is required to prevent excessive replication fork pausing at nonhistone-protein-bound sites , possibly by acting as a protein displacement helicase [6] . The role of Rdh54 as a dsDNA translocase that can act on chromatinized DNA [32 , 33] and the fork reversal activity of Rad5 suggest that they are recruited to chromatin to recover forks that are blocked due to the lack of Rrm3 or to substitute for Rrm3 in preventing fork pausing . We therefore examined the effect of deleting RAD5 and RDH54 in the rrm3Δ mutant on genome stability and sensitivity to DNA damage caused by methyl methanesulfonate ( MMS ) and to replication stress caused by HU . We found synergistic increases in sensitivity to HU and MMS in the rrm3Δ rad5Δ and rrm3Δ rdh54Δ mutants ( Fig 2A ) . The negative genetic interaction between rrm3Δ and rad5Δ was particularly strong; both single mutants were no more sensitive to HU than wildtype , but the double mutant failed to form colonies on 100 mM HU and grew very poorly even on 20 mM HU . In contrast to HU , the rad5Δ mutant was extremely sensitive to MMS , and deleting RRM3 led to a further synergistic increase in MMS sensitivity . Inactivation of the ATPase activity of Rrm3 ( rrm3-K260A/D ) caused the same hypersensitivity in the rad5Δ mutant as an RRM3 deletion ( Fig 2B ) . We also identified a negative genetic interaction between rrm3Δ and rdh54Δ , which was especially strong on MMS . The increased sensitivity of rdh54Δ cells to HU and MMS upon deletion of RRM3 extended to diploid cells ( Fig 2G ) , suggesting that the lesions generated in the absence of Rrm3 are also substrates for recombination between homologous chromosomes that is controlled by Rdh54 . Even though the rrm3Δ rad5Δ mutant was hypersensitive to MMS and HU , deletion of RDH54 caused further synergistic increases in sensitivity to both chemicals , indicating that Rad5 and Rdh54 define important pathways for dealing with DNA lesions that arise in the absence of Rrm3 , and that they perform ( at least some ) independent roles . In addition to structure-specific helicase activity , Rad5 also possesses a RING motif associated with ubiquitin ligase activity that plays a role in polyubiquitination of PCNA [34–37] . Two mutations in Rad5 , Q1106D and C914A/C917A , were recently described to disrupt its helicase and ubiquitin-ligase activity , respectively [27 , 38] . Disrupting either of these Rad5 activities in the rrm3Δ mutant significantly increased sensitivity to MMS and to HU , indicating that both activities make important contributions to the repair of DNA lesions that arise in the absence of Rrm3 ( Fig 2B ) . Although Rad5 and Rdh54 chromatin association increased most in the absence of Rrrm3 ( Fig 1E ) , gross-chromosomal rearrangements ( GCRs ) did not accumulate at higher rates in the rrm3Δ rad5Δ or rrm3Δ rdh54Δ mutants compared to the single mutants , even after exposure to HU and MMS ( Table 1 ) . However , disruption of both , RAD5 and RDH54 , in the rrm3Δ mutant caused a significant , albeit small , increase in chromosome instability compared to disruption of the single genes , especially upon exposure to HU or MMS , supporting independent contributions of Rdh54 and Rad5-mediated repair mechanisms to genome stability and DNA damage tolerance in the absence of Rrm3 . That the effect on the GCR accumulation was small despite a synergistic effect on hypersensitivity to HU and MMS ( Fig 2A ) could indicate that the DNA lesions are not substrates for GCRs or that chromosomal rearrangements that form in these mutants are not viable . Whereas rdh54Δ and rad5Δ cells moved through an undisturbed cell cycle with similar kinetics as wildtype cells , rrm3Δ cells were delayed in progressing to G2/M , consistent with previous observations [6 , 14] . We find that this delay was enhanced when RDH54 or RAD5 were deleted ( Fig 2C ) . To examine progression of rrm3Δ rad5Δ and rrm3Δ rdh54Δ cells through S phase under replication stress , we released α-factor arrested cells from G1 phase in the presence of HU and trapped them in G2/M with nocodazole ( Fig 2D and 2E ) . After 140 minutes , virtually all wildtype cells had reached 2C DNA content , whereas rrm3Δ and rdh54Δ showed a marked delay ( Fig 2D , 120 minute time point ) . When we combined rrm3Δ and rdh54Δ mutations , this slowdown was so severe that most cells still had near 1C DNA content 100 minutes after release from G1 arrest , consistent with the synergistic increase in HU hypersensitivity of the rrm3Δ rdh54Δ mutant . Similarly , rad5Δ rrm3Δ cells were delayed in reaching 2C DNA content in HU ( Fig 2E ) . However , all mutants were able to recover from a 2-hour arrest in 100 mM HU and resume the cell cycle normally ( S1 Fig ) . When we examined the ability of nocodazole-arrested cells in G2/M to complete mitosis and reach G1 phase , we found that most wildtype cells were in G1 after 10 minutes , whereas rrm3Δ , rad5Δ , rdh54 cells and the double mutants showed 2C DNA content 20 minutes after release ( Fig 2F ) . Together , these findings indicate that Rad5 and Rdh54 facilitate the progression of rrm3Δ cells through S phase , both in the presence and in the absence of HU , and that in the absence of Rrm3 , cells accumulate DNA damage that impairs exit from mitosis . In addition to Rad5 and Rdh54 , which exhibited the most significant increases in chromatin association in the absence of Rrm3 ( Fig 1E ) , we tested DNA damage sensitivity of cells that lacked Rrm3 in combination with other nonessential factors revealed in the proteome screen ( Fig 1C and 1D ) , including Mgm101 , Hda1 , Set3 , and Mph1 . Whereas deletions of MGM101 , HDA1 or SET3 had no effect on sensitivity of wildtype or rrm3Δ cells to HU or MMS ( S2 Fig ) , deletion of MPH1 caused a synergistic increase in HU and MMS sensitivity of the rrm3Δ mutant ( S3 Fig ) , consistent with our previous finding [39] . In the absence of Mph1 , rrm3Δ cells progressed very slowly through an undisturbed cell cycle and accumulated in G2/M when they were released from a 2-hour incubation in 100 mM HU ( S3B and S3C Fig ) . When cells were released from HU arrest into media with 40 mM HU and α-factor , virtually all wildtype cells and the single mutants were trapped in G1 phase after 60 minutes ( with a slight S phase delay in the mph1Δ mutant ) , whereas the majority of rrm3Δ mph1Δ cells accumulated in S phase , never forming a majority peak at 1C DNA content in the 120-minute time course ( S3D Fig ) . Deleting MPH1 in the rrm3Δ rad5Δ mutant led to a slight increase in sensitivity to HU compared to the double mutants , but not to MMS ( S3E Fig ) . These findings implicate Mph1 as another crucial factor for overcoming spontaneous and DNA-damage-induced replication-blocking lesions when Rrm3 is absent and suggest pathways for error-free bypass of DNA polymerase blocking lesions , implicated by Rad5 and Mph1 , and homologous recombination , implicated by Rdh54 , as independent mechanisms that are recruited to chromatin to act on blocked replication forks . All functions of the Rrm3 helicase known to date are dependent on its ATPase/helicase activity . During our analysis of cell cycle progression , however , we observed that cells with a deletion of RRM3 continue to replicate DNA in the presence of HU , similar to a rad53Δ checkpoint mutant , whereas the helicase-defective rrm3-K260A and rrm3-K260D mutants maintained near 1C DNA content after 2 hours in HU , similar to wildtype ( Fig 3B and 3C ) . This observation suggested the presence of a previously unknown , ATPase/helicase-independent function of Rrm3 in DNA replication . Since this replication defect was independent of the ATPase/helicase activity located in the ordered C-terminal domain of Rrm3 ( residues 250–723 ) , we explored a possible involvement of the 230-residue , disordered N-terminal tail ( Fig 3A , S4A Fig ) . The only motifs previously identified in this tail are a putative PCNA-interacting peptide ( PIP ) box between residues 35–42 [18] and a cluster of phosphorylated residues between S85 and S92 [17] . Deletion or mutation of the PIP-box ( rrm3-ΔN54 , rrm3-FFAA ) had no effect on DNA replication in HU , whereas deletion of the entire N-terminal tail ( rrm3-ΔN230 ) caused the same replication defect as deleting RRM3 ( rrm3Δ ) ( Fig 3C ) . By constructing a series of N-terminal truncations ( Fig 3A and 3C ) we determined that a deletion of up to 186 residues , which also eliminates the PIP-box and the phosphorylation site , was able to maintain the wildtype replication phenotype in HU , whereas deletions of 212 or 230 residues caused the same inability to restrict DNA synthesis in HU as rrm3Δ ( Fig 3D–3G ) , thus narrowing down the critical functional site for control of DNA replication to the 26 residues between residues 186–212 . This defect was not due to changes in protein stability or levels of expression of the rrm3 mutant alleles ( Fig 3H ) , which had been observed for other Rrm3 truncations before [20] . The importance of residues 186–212 for controlling DNA replication was limited to HU , and not observed when cells were exposed to the alkylating agent MMS ( S4B Fig ) . Deletion of RRM3 or inactivation of its ATPase/helicase activity was recently reported to partially suppress the HU hypersensitivity of the rad53Δ mutant [17] . We obtained the same findings , and observed that the new rrm3-ΔN212 allele does not act as a suppressor ( Fig 3I ) , indicating that the rrm3-ΔN212 allele codes for a functional ATPase/helicase . To verify this , we disrupted the Walker A motif in the rrm3-ΔN212 mutant ( rrm3-Δ212-K260A ) and , as expected , it now suppressed the HU hypersensitivity of the rad53Δ mutant to the same extent as the ATPase-defective rrm3-K260A allele ( Fig 3I ) . The Rad53 checkpoint kinase was constitutively activated in the rrm3-Δ212 mutant just like in the ATPase/helicase-defective rrm3-K260A/D mutants , and Rad53 activation in both mutants was dependent on the mediator of the DNA damage checkpoint Rad9 ( Fig 3J and 3K ) . Through degradation of the ribonucleotide reductase ( RNR ) inhibitor Sml1 , the nucleotide pool increases upon Rad53 activation , and this correlates with enhanced fork progression [22] . We found that the rrm3 mutants that continued DNA replication in HU ( rrm3Δ , rrm3ΔN212 ) as well as the rrm3 mutant that maintained a peak at 1C DNA content ( rrm3-K260D ) had constitutively increased nucleotide pools upon entrance into S phase ( Fig 3L ) , indicating that the continued DNA replication in HU seen in the rrm3-ΔN212 mutant could not be explained by a larger nucleotide reservoir prior to its depletion by HU addition . In fact , based on quantification of the cell cycle profiles obtained by flow cytometry and by visual analysis of morphology the vast majority of rrm3Δ , rrm3-ΔN212 and rrm3-ΔN230 cells entered S phase in HU and continued to progress , whereas the majority of wildtype cells and the other rrm3 mutants did not enter S phase during the 2-hour incubation in HU , with the peaks of DNA content remaining at 1C ( Fig 3B–3G , S4C Fig ) . Together , these findings suggest a new function of Rrm3 in restricting DNA replication in the presence of HU and prevention of S phase damage , which maps to residues 186–212 of the N-terminal tail and does not require Rrm3’s established activity as an ATPase/DNA helicase . Long disordered tails , such as the N-terminal 230 residues of Rrm3 that extend from a structured catalytic core , typically serve as sites for protein binding and posttranslational modification [40] . The phenotype of the rrm3-ΔN212 allele in the rad53Δ mutant indicates that it encodes a proficient ATPase/helicase , suggesting that the replication defect of this allele is caused by loss of a protein-binding site in the disordered tail . Because deletion of the putative PIP-box and the recently identified phosphorylation site did not impair the ability of Rrm3 to control DNA replication , we explored the possibility that Orc5 , an ATP-binding subunit of the origin recognition complex ( ORC ) , binds to the N-terminal tail of Rrm3 . An interaction between the two full-length proteins had previously been identified in a yeast-two-hybrid screen [41] . When we combined ORC5 with the various rrm3 truncation alleles in a yeast two-hybrid assay , we found that deletion of 186 residues did not diminish Orc5 binding to Rrm3 , in the presence or absence of MMS or HU , whereas deletion of 212 or 230 residues eliminated binding ( Fig 4A ) . To verify the importance of the N-terminal region of Rrm3 for Orc5 binding in vivo we tested the ability of Orc5 to co-immunoprecipitate myc-epitope-tagged Rrm3 , rrm3-ΔN186 and rrm3-ΔN212 . Consistent with the yeast-two hybrid assay , wildtype Rrm3 and rrm3-ΔN186 bound efficiently to Orc5 whereas binding of rrm3-ΔN212 was impaired ( Fig 4B ) . These findings show that the same site of Rrm3 that restricts DNA replication in HU is required for a physical interaction with Orc5 in vivo and raise the possibility that Rrm3 may control DNA replication by affecting replication origins . To test the hypothesis that Rrm3 acts on replication origins we tested if Rrm3 and rrm3-ΔN212 associate with origins and if this association is affected by the presence of HU . Since progression of rrm3Δ and rrm3-ΔN212 mutants into S phase in the presence of HU is not as pronounced as in the absence of the Rad53 checkpoint kinase , which modulates the timing of origin firing and S phase progression upon exposure to HU , we considered that Rrm3 might act only on a subset of origins . We therefore selected a variety of replication origins for analysis by chromatin immunoprecipitation ( ChIP ) , ranging from early to late-initiating origins and including two origins near telomeres ( ARS319 , ARS501 ) . We performed ChIP on asynchronous cultures , cultures synchronized in G1 with α-factor , as well as cultures that were released from G1 into S phase for 45 minutes in the presence or absence of HU . We observed that Rrm3 and rrm3-ΔN212 associated with ARS305 , ARS601 , ARS603 , ARS607 , and ARS1414 in asynchronous cultures , in G1 and in S phase , but not with ARS1411 ( Fig 4C ) . In fact , the lack of a PCR product for ARS1411 in the asynchronous culture indicates that Rrm3 does not associate with this replication origin for any extended period during the cell cycle; Rrm3 was also not at ARS306 , ARS319 , ARS416 , ARS522 , ARS606 , and ARS609 ( S4D Fig ) . This suggests that Rrm3 associates with a subset of origins of replication in unperturbed G1 and S phase independently of its Orc5-binding site . However , when we released cells into S phase in the presence of 200 mM HU , rrm3-ΔN212 lost its association with ARS602 , ARS603 , ARS607 , and ARS1414 whereas wildtype Rrm3 remained bound ( Fig 4C ) , suggesting that the failure of the rrm3-ΔN212 mutant to halt DNA synthesis and progression into S phase in the presence of HU might be due to a failure of rrm3-ΔN212 to act on a subset of replication origins ( 40% of origins tested in this study ) . To investigate the link between Rrm3 functions and DNA replication , we examined the replication checkpoint . Replication mutants exhibit strong genetic interactions with Mrc1/Claspin , which acts as a mediator of the replication stress checkpoint–a Rad9-independent pathway of the intra-S-phase checkpoint [42–46] . Mrc1 is also a component of normal replication forks , which is loaded at origins of replication and stays associated with the replisome [42 , 44 , 47 , 48] . Mrc1 , like Rrm3 , is required for efficient replication [49] . The function of Mrc1 in DNA replication is essential for the viability of cells lacking Rrm3 [47] whereas Mrc1 phosphorylation on SQ and TQ sites linked to its checkpoint function is dispensable [19] . However , the role of this functional interaction between Rrm3 and Mrc1 in DNA replication has remained unclear . We therefore tested if the ability of Rrm3 to control DNA replication was required for the viability of the mrc1Δ mutant . For this purpose , we transformed diploids heterozygous for the mrc1Δ and rrm3Δ mutations with plasmids expressing N-terminal truncations of Rrm3 and analyzed the viability of meiotic products . Fig 5A shows that the rrm3-ΔN186 allele restored viability to the rrm3Δ mrc1Δ mutant as effectively as the wildtype RRM3 allele , whereas the helicase-dead alleles and the rrm3-ΔN212 allele were as ineffective as the null allele ( empty plasmid ) . Thus the helicase activity of Rrm3 is not sufficient for viability of the mrc1Δ mutant; Rrm3’s new Orc5-binding domain for controlling DNA replication is also required . In addition to Mrc1 , Tof1 promotes normal progression of the replication fork; however , in contrast to Mrc1 , its requirement for fork progression appears more limited , assisting primarily replication through non-histone protein complexes with DNA [50] . TOF1 deletion was not lethal in the rrm3Δ mutant and neither single mutant was hypersensitive to HU or MMS . The combined loss of Rrm3 and Tof1 , however , caused a synergistic increase in DNA-damage sensitivity ( Fig 5B ) . Identical to the functional requirements in the absence of Mrc1 both , the ATPase/helicase activity of Rrm3 and the Orc5 binding domain , were required for growth in the presence of DNA damage and replication stress in the absence of Tof1 . In contrast to mrc1Δ and tof1Δ mutants , we found that only the ATPase/helicase activity of Rrm3 was required for the suppression of HU and MMS hypersensitivity of the rdh54Δ mutant ( Fig 5C ) . The N-terminal tail , including its function in controlling DNA replication , was dispensable , with the rrm3-ΔN212 allele exhibiting a wildtype phenotype in the rdh54Δ mutant . When we extended this analysis to the Rdh54 homolog Rad54 , and the HR factor Rad51 , which both Rdh54 and Rad54 interact with [29 , 51] , we observed that the rrm3 alleles caused the same phenotypes in the rad51Δ mutant as in the rdh54Δ mutant , but had no effect on the rad54Δ mutant ( Fig 5D and 5E ) . Like rdh54Δ and rad51Δ mutants , rad5Δ and mph1Δ mutants exhibited increased sensitivity to HU and MMS only if the ATPase activity of Rrm3 was disrupted , whereas the rrm3-ΔN212 allele caused the same phenotype as the RRM3 wildtype allele ( Fig 5F and 5G ) . Together , these findings suggest two separable functions of Rrm3 in DNA replication . First , an ATPase/helicase-dependent function that facilitates fork progression through protein-DNA complexes , which if disrupted ( rrm3-K260A/D ) causes aberrant replication intermediates that can be rescued by Rad5 , Rdh54/Rad51 or Mph1 mechanisms . Second , an N-terminal function that restricts DNA replication in the presence of HU , mediated by Rrm3 association with replication origins , which if disrupted ( rrm3-ΔN212 ) requires the replication checkpoint factors Mrc1 and Tof1 for viability and DNA damage survival . This differential requirement of factors involved in DNA repair and DNA damage tolerance pathways in the rrm3-ΔN212 and rrm3-K260A/D mutants also suggests that the types of DNA lesions that accumulate upon inactivation of the two Rrm3 functions are different , but both lead to dependence on Mrc1 for survival and both are sufficient for constitutive activation of the DNA-damage checkpoint . If Rrm3 is important for the response to replication stress induced by HU , cells lacking the catalytic activity of Rrm3 or its Orc5-binding domain might be prone to accumulating mutations at higher rates than wildtype cells . To test this , we measured forward mutation rates at the CAN1 locus and the accumulation of GCRs on chromosome V in the presence and absence of HU or MMS ( Table 2 ) . Two-fold ( ung1Δ ) to 50-fold ( rad27Δ ) increases in CAN1 forward mutation rates compared to wildtype have previously been reported for numerous DNA metabolism mutants [52] . Deletion of RRM3 or disruption of its ATPase/helicase activity caused a significant increase in spontaneous CAN1 mutations ( Table 2 ) . Of the truncation alleles , which encode catalytically active rrm3 mutants ( Fig 3I ) , rrm3Δ-N186 was indistinguishable from wildtype whereas rrm3Δ-N212 caused a small , but significant , increase in the CAN1 mutation rate in untreated cells and upon exposure to HU . In contrast , expression of the rrm3Δ-N212 allele had no effect on the CAN1 mutation rate if cells were treated with MMS , consistent with our observation that the rrm3-ΔN212 mutant exhibits a defect in controlling replication in HU , but not MMS . GCRs accumulated at increased rates in the rrm3Δ and rrm3-K260A/D mutants in the absence and presence of HU or MMS , but accumulated at wildtype levels in cells expressing N-terminal truncations under all conditions . These mutator phenotypes , albeit mild , reveal that Rrm3’s ATPase/helicase activity helps to suppress all tested mutation types induced by either HU or MMS , or in their absence , whereas the N-terminal plays a role specifically in the suppression of spontaneous and HU-induced mutations , but not for the suppression of MMS-induced mutations , or GCRs under any conditions .
By quantifying changes in chromatin composition we have identified Rad5 and Rdh54 as novel factors that respond to increased replication fork stalling induced by the absence of Rrm3 , and affirmed the importance of Mph1 . These factors suggest that error-free post-replicative repair ( PRR ) , implicated by Rad5 and Mph1 , and HR , implicated by Rdh54 , act on DNA polymerase blocking sites that arise throughout the genome in the absence of Rrm3 . The N-terminal unstructured tail , including the Orc5-binding site identified in this study , is dispensable for this ATPase/helicase-dependent role of Rrm3 in facilitating fork progression . Instead , we have discovered that the N-terminal tail encodes a new function of Rrm3 –to control DNA replication in the presence of HU . This function of Rrm3 is distinct from its established role as an ATPase/helicase , is not regulated by the previously identified phosphorylation cluster [17] or the PIP-box [18] and , in contrast to the ATPase/helicase activity of Rrm3 , does not contribute to the HU hypersensitivity of the rad53Δ mutant . Based on changes in DNA content as measured by flow cytometry , we observed that wildtype cells maintained near 1C DNA content for 180 minutes after release from G1 phase into HU , whereas rad53Δ , rrm3Δ and rrm3-ΔN212 did so for only 60 minutes ( Fig 3 , S4C Fig ) . The extent of continuing DNA replication in the presence of HU , however , was not as pronounced in the rrm3 mutants as in the rad53Δ mutant . Although the DNA-damage checkpoint is chronically activated in the rrm3-ΔN212 mutant and , as a consequence , nucleotide levels are increased as cells are about to enter S phase , these increased nucleotide levels do not appear to be not sufficient for ability of the rrm3-ΔN212 mutant to continue DNA replication upon HU exposure because the rrm3-K260A/D mutants showed the same nucleotide level increase and DNA-damage checkpoint activation , but maintained a peak at 1C DNA content in HU with only a small percentage of cells entering S phase ( Fig 3G ) . Therefore , considering Rrm3’s known function as an accessory ATPase/helicase that facilitates progression of the replication fork , and its new function in controlling DNA synthesis reported here , we propose a model ( Fig 6 ) where Rrm3 performs two genetically and physically separable functions to deal with challenges during genome duplication: ( 1 ) the N-terminal tail of Rrm3 plays a structural role in preventing untimely replication in the presence of replication stress ( HU ) and in normal S phase , and ( 2 ) the C-terminal ATPase/helicase domain plays a catalytic role in preventing fork pausing . The site between residues 186 to 212 , which is in a segment of the N-terminal tail not previously assigned a function , is not only involved in restricting DNA synthesis in HU , but also for the association of Rrm3 with Orc5 and a subset of origins of replication . That rrm3-ΔN212 fails to restrict DNA synthesis and S phase progression in the presence of HU suggests that the association of Rrm3 with origins of replication is inhibitory and that this inhibition is realized through binding Orc5 , the ATP-binding subunit of ORC . By binding Orc5 , Rrm3 could act as an inhibitor of ORC ATPase activity , which is required for loading of minichromosome maintenance ( MCM ) proteins and for initiation of DNA replication [53 , 54] , or Rrm3 could block the recruitment of another replication protein . Although ORC is associated with origins throughout the cell cycle , Orc5 does not appear to play a role in the completion of S phase , or the remainder of the cell cycle [55 , 56] . Besides the association of Rrm3 with origins in HU , which depends on the N-terminal tail , our ChIP data from the rrm3-ΔN212 mutant also show association with origins in normal G1 and in unperturbed S phase in a manner that does not require the N-terminal tail , invoking the presence of another protein binding site in the rrm3-ΔN212 polypeptide , which is made up almost entirely of the catalytic domain . Indeed , the phenotype of rrm3-ΔN212 mutants deficient in HR ( rdh54Δ , rad51Δ ) or PRR ( rad5Δ , mph1Δ ) suggests that the association of Rrm3 with the replisome [1] , appears to occur outside of the N-terminal tail . While the N-terminal tail is required for binding Orc5 , it is not required for association with origins of replication , unless cells are exposed to replication stress . Together with continued DNA synthesis and progression into S phase in the presence of HU our findings , thus , raise the possibility that Rrm3 performs a replication checkpoint-like function in response to HU . Instead of a global role in controlling origin activity , the wildtype level of HU sensitivity of rrm3Δ cells , the less pronounced S phase progression in HU than that of the rad53Δ mutant , and the importance of Rrm3 for replicating through certain nonhistone-protein-bound regions suggest that Rrm3 may play a role at origins in specific loci , such as those in highly transcribed regions and regions with converging transcription , which are often late-firing [57] , rRNA and tRNA coding loci , or highly transcribed metabolic genes , where ORC has been found to be bound to the open reading frames , possibly to coordinate the timing of replication with transcription [58] . Indeed , our analysis so far has revealed that Rrm3 appears to associate only with a subset of replication origins . The group of origins not associated with Rrm3 includes early and late replicating origins as well as two origins near telomeres . It is currently unclear what distinguishes the origins that associate with Rrm3 from those that do not . For example , there are no consistent differences in how their activity in HU is affected by mutations in the DNA-damage checkpoint or the DNA replication checkpoint [59] , the time they are activated [60 , 61] , or any obvious chromosomal features . Association with some but not other origins in unperturbed G1 and S phase does indicate , however , that Rrm3 interacts with a factor specific to some origins rather than a replication protein common to all pre-RCs or replisomes . The role of Rrm3 at certain origins in HU is likely to be the cause of enhanced DNA synthesis and S phase progression in the rrm3Δ and rrm3-ΔN212 mutants . However , DNA-damage checkpoint activation in rrm3-ΔN212 , but not rrm3-ΔN186 , and synthetic lethality between mrc1Δ and rrm3-ΔN212 , but not rrm3-ΔN186 , suggest that there are conditions of replication stress other than exposure to HU that require the integrity of the Orc5-binding domain of Rrm3 . The role of Rrm3 in controlling DNA replication is not affected by inactivation of the ATPase/helicase activity ( Fig 6C ) . Instead , it impairs Rrm3’s established function in facilitating fork progression through replication blocks , leading to the accumulation of DNA lesions that activate the DNA-damage checkpoint and can give rise to mutations . By identifying changes in chromatin composition combined with genetic assays we have identified Rad5 and Rdh54 as novel factors that contribute to the maintenance of genome stability in the absence of Rrm3’s ATPase/helicase activity . Rad5 defines an error-free pathway for the bypass of DNA polymerase blocking lesions [26–28 , 62–64] . As a structure-specific DNA helicase , Rad5 is capable of regressing replication forks in vitro [25] . Such a regressed fork is thought to provide an alternative template for DNA synthesis , generating enough nascent DNA to eventually bypass the replication block . The ATPase activity of Rad5 and the RING motif involved in polyubiquitination of PCNA [34–36 , 65] contribute to DNA damage tolerance in the absence of Rrm3 . Evidence for a role of the ATPase activity of Rad5 in remodeling blocked replication forks has been obtained in vitro [25] whereas a role of Rad5-dependent polyubiquitination of PCNA in activating HR-dependent template switching has more recently been suggested [66] . Evidence that these two Rad5 activities can function independently , as we determined here in the rrm3Δ mutant , was also observed for bypass of MMS-induced lesions by sister-chromatid recombination [66] . Besides fork regression , Rad5 has also been implicated in DNA damage bypass by HR-dependent template switching between sister-chromatids [66] and the major HR factors Rad51 , Rad52 and Rad54 as well as Sgs1 have been implicated in error-free DNA lesion bypass [67] . It was therefore surprising that Rdh54 , a dsDNA translocase that is known to play a major role in meiotic , but not mitotic , HR [68–70] , is recruited to chromatin when Rrm3 is absent—both in the presence and absence of HU . Rdh54 was only required in the absence of the ATPase/helicase activity of Rrm3 , but not in the absence of the Orc5-binding domain , implicating Rdh54 in repair of DNA lesions that arise when Rrm3 cannot facilitate fork progression through replication blocks . Even though Rdh54 does not affect gene conversion repair of a DSB , a role specifically in repair that involves template switches was recently reported [71] , and could be related to its increased chromatin association and the DNA-damage hypersensitivity of the rrm3Δ rdh54Δ and rrm3-K260A rdh54Δ mutants . That Rad51 was required for this repair , but Rad54 was not , suggests a Rad51-dependent HR pathway in mitotic cells where Rdh54 takes the place of Rad54 . Although it is unknown how Rdh54 acts in template switching , its activities in vitro seem compatible with those that may be required to rescue a paused fork . Like Rad5 and the human Rad5 ortholog , HTLF , Rdh54 is a dsDNA translocase of the SWI/SNF family [29 , 30 , 72] . In vitro , it can dislodge Rad51 from dsDNA and introduces negative supercoiling into dsDNA that can cause strand separation [29–31] . These Rdh54 activities could help to regulate restart at fork pause sites in Rad5-mediated pathways , such as fork regression/reversal or template switching , and in HR-mediated events . Whereas Rdh54 can remove proteins from dsDNA and remodel chromatinized DNA , an ability to remove bound proteins from DNA has not yet been shown for Rad5 , and RecQ-like helicases are only capable of acting on forked DNA structures that are protein-free [32 , 33] . The synergistic interactions between rad5Δ and rdh54Δ in the absence of RRM3 clearly identify a requirement of Rdh54 outside of a Rad5 mechanism . In addition to facilitating template switching HR when error-free PRR is inactivated , Rdh54 could act in the avoidance of replication fork pausing in a manner similar to Rrm3 by removing certain proteins from dsDNA , such as shown for Rad51 , which appears to have a tendency to associate with nonrecombinogenic dsDNA [29 , 31 , 70] . That the ATPase activity of Rrm3 is required in the absence of Rad5 , Rdh54 , Rad51 or Mph1 , whereas the role of Rrm3 in controlling DNA replication is dispensable strongly suggests that the types of DNA damage checkpoint activating DNA lesions in the rrm3-K260A and rrm3-ΔN212 mutants are different , and that Rad5 , Rdh54 , Rad51 and Mph1 act on DNA lesions that form when replications forks are unable to move through obstacles , but not on DNA lesions that form during untimely DNA replication ( rrm3-ΔN212 mutant ) . In contrast , Mrc1 and Tof1 were required for viability and DNA damage tolerance when either of the two Rrm3 activities was disrupted . Mrc1 , the mediator of the replication stress checkpoint , mediates Rad53 phosphorylation specifically in response to replication fork pausing , leading to intra-S checkpoint activation and inhibition of late-origin firing [44 , 59] . That synthetic lethality between rrm3Δ and mrc1 is limited to those mrc1 alleles that cause DNA damage accumulation during S phase [49] , whereas the checkpoint function of Mrc1 is dispensable [19 , 49] suggests that the additive accumulation of S phase damage due to lack of both , Mrc1 and Rrm3 , is lethal and suggests that dysregulated replication in the rrm3-ΔN212 mutant ( Fig 6B ) also leads to S phase damage , consistent with our observation of Rad9-dependent activation of Rad53 . Finally , the new N-terminal Rrm3 function in controlling DNA replication is separated from Rrm3’s established C-terminal function as an ATPase/helicase in facilitating fork progression not only by the differential requirement for Rad5 , Rdh54 , Rad51 and Mph1 , but also by different spontaneous and DNA-damage induced mutation spectra . This supports that the N-terminal tail is neither involved in the recruitment of Rrm3 to active replication forks nor in facilitating fork progression through protein-bound sites , and that a separate replisome binding site is likely to be located in the ATPase/helicase domain . The accumulation of GCRs and point mutations in the ATPase /helicase mutant , spontaneously or induced by HU or MMS , could be indicative of DNA break formation as a result of replication fork stalling . In contrast , the Orc5-binding domain mutant did not accumulate GCRs under any conditions , suggesting wildtype levels of DNA breaks , including in HU and MMS , but increasingly formed point mutations . That these point mutations formed specifically in response to HU , but not MMS , suggests that they arise during the enhanced DNA synthesis that occurs in this mutant in HU . In summary , this study has revealed a 26-residue region in Rrm3 that is critical for a novel , ATPase-independent function of Rrm3 in preventing untimely DNA replication and for binding Orc5 , which appear to be mechanistically linked . We also identified association of Rrm3 with a subset of replication origins and the dependence of this association on the N-terminal tail under replication stress , but not in unperturbed cells . Genome-wide ChIP and quantification of DNA synthesis in cells expressing the new rrm3 alleles will help to reveal the regions undergoing untimely DNA replication and provide further insight into the mechanism at replication origins underlying continued DNA synthesis under replication stress and lethality with mrc1Δ . That yeast has two DNA helicases ( Rrm3 , Pif1 ) that belong to the Pif1 family , whereas multicellular eukaryotes where the Pif 1 helicase family is conserved [2] typically have one ( e . g . Pif1 in humans ) , might be an indication that Rrm3’s role in DNA replication is highly specialized to control replication and facilitate fork progression in genomic regions that are distinctively organized in yeast and to deal with the high gene density imposed on its small genome that requires tight coordination between replication initiation and ongoing transcription .
For double isotope labeling of lysine and arginine , yeast strain KHSY5144 ( lys2Δ arg4Δ ) was grown at 30°C with and vigorous shaking for at least ten generations in “heavy” medium ( 6 . 9 g/l yeast nitrogen base without amino acids [Formedium] , 1 . 85 g/l amino acid dropout mixture without arginine and lysine [Kaiser formulation , Formedium] , 2% glucose , 15 mg/l [13C6] L-arginine and 30 mg/l [13C6] or [13C6 , 15N2] L-lysine ) . KHSY5143 ( lys2Δ arg4Δ rrm3Δ ) was grown in “light” medium , containing 15 mg/l L-arginine and 30 mg/l L-lysine at 30°C with and vigorous shaking . Chromatin was isolated using a method adapted from [73] . Approximately 4 x 109 cells were resuspended in 10 ml of 100 mM PIPES/KOH , pH 9 . 4 , 10 mM dithio- treitol ( DTT ) , 0 . 1% sodium azide , then incubated for 10 min at room temperature , followed by incubation in 10 ml of 50 mM KH2PO4/K2HPO4 , pH 7 . 4 , 0 . 6 M sorbitol , 10 mM DTT , containing 200 mg/ml Zymolyase-100T and 5% Glusulase at 37°C for 30 min with occasional mixing . Spheroplasts were washed with 5 ml of ice-cold wash buffer ( 20 mM KH2PO4/K2HPO4 , pH 6 . 5 , 0 . 6 M sorbitol , 1 mM MgCl2 , 1 mM DTT , 20 mM beta-glycerophosphate , 1 mM phenyl-methylsulfonyl fluoride ( PMSF ) , Protease inhibitor tablets ( EDTA free , Roche ) and resuspended in 5 ml of ice-cold wash buffer . The suspension was overlaid onto 5 ml of 7 . 5% Ficoll-Sorbitol cushion buffer ( 7 . 5% Ficoll , 20 mM KH2PO4/K2HPO4 , pH 6 . 5 , 0 . 6 M sorbitol , 1 mM MgCl2 , 1 mM DTT , 20 mM beta-glycerophosphate , 1 mM PMSF , Protease inhibitor tablets ) and spheroplasts were spun through the cushion buffer at 5000 rpm for 5 min to remove proteases derived from Zymolyase-100T . Pelleted spheroplasts were resuspended in 200 ml of ice-cold wash buffer and dropped into 18% Ficoll , 20 mM KH2PO4/K2HPO4 , pH 6 . 5 , 1 mM MgCl2 , 1 mM DTT , 20 mM beta-glycerophosphate , 1 mM PMSF , Protease inhibitor tablets , 0 . 01% Nonidet P-40 , with stirring . After homogenization , unbroken cells were removed by two spins ( 5000 x g for 5 min at 4°C ) . Nuclei were pelleted by centrifugation at 16 , 100 x g for 20 min and the cytoplasmic fraction removed . After washing in ice-cold wash buffer , nuclei were resuspended in 200 ml of EB buffer ( 50 mM HEPES/KOH , pH 7 . 5 , 100 mM KCl , 2 . 5 mM MgCl2 , 0 . 1 mM ZnSO4 , 2 mM NaF , 0 . 5 mM spermidine , 1 mM DTT , 20 mM beta-glycerophosphate , 1 mM PMSF , protease inhibitor tablets ) and lysed by addition of Triton X-100 to 0 . 25% , followed by incubation on ice for 10 min . Lysate was overlaid on 500 ml of EB buffer , 30% sucrose , 0 . 25% Triton X-100 , and spun at 12 , 000 rpm for 10 min at 4°C . The top layer was removed and the chromatin pellet was washed in 1 ml of EB buffer , 0 . 25% Triton X-100 and spun at 10 , 000 rpm for 2 min at 4°C . The chromatin pellet was resuspended in 40 μl of 1 . 5x Tris-Glycine SDS sample buffer and incubated for 2 min at 85°C . DTT was added to a final concentration of 5 mM and incubated for 25 min at 56°C . Iodoacetamide was added to 14 mM final concentration and incubated for 30 min at room temperature in the dark . DTT was added to a final concentration of 5 mM and incubated for 15 minutes at room temperature in the dark . The protein mixture was diluted 1:5 in 25 mM Tris-HCl , pH 8 . 2; CaCl2 was added at a final concentration of 1 mM , and trypsin was added at 4–5 ng/μl , followed by incubation at 37°C overnight . Trifluoroacetic acid was added to 0 . 4% final concentration and centrifuged at 2 , 500 x g for 10 min at room temperature . Peptides in the supernatant were desalted using reverse-phase tC18 SepPak solid-phase extraction cartridges ( Waters ) . The sample was eluted with 5 ml of 50% acetonitrile and lyophilized . The lyophilized product was resuspended in 0 . 1% formic acid prior to tandem mass spectrometric analysis on an LTQ Orbitrap XL ( Thermo ) . Scans on the Orbitrap were obtained at a mass resolving power of 60000 at m/z 400 and top 10 abundant ions were selected for fragmentation in the LTQ ion trap . Further processing of the RAW files was done in MaxQuant version 1 . 5 . 3 . 30 [74] against the Saccharomyces genome database ( SGD ) . A database of known contaminants in MaxQuant was used as well as constant modification of cysteine by carbamidomethylation and variable modification of methionine oxidation . The first search tolerance was set at 20 ppm , then 8 ppm tolerance for the main search . Fragment ion mass tolerance was set to 0 . 5 Da and the minimum peptide length was 6 amino acids . Unique and razor peptides were used for identification and the false discovery rate was set to 1% for peptides and proteins [74 , 75] . Statistical analysis of the data was carried out with Perseus software using an approach by Benjamini and Hochberg [76] . Yeast strains used in this study are derived from S288C background . Strains for SILAC labeling and chromatin fractionation were derived from KHSY5036 ( MAT ɑ , ura3-52 , trp1Δ63 , his3Δ200 ) . Yeast strains used in DNA-damage sensitivity and mutation assays were derived from KHSY802 ( MAT ɑ , ura3-52 , trp1Δ63 , his3Δ200 , leu2Δ1 , lys2Bgl , hom3-10 , ade2Δ1 , ade8 , hxt13::URA3 ) . Gene deletions were carried out as described using HR-mediated integration of a selectable cassette [77] . Mutants containing more than one gene deletion or mutation were obtained by random spore isolation from diploids heterozygous for the desired mutations . Point mutations were introduced in plasmids by site-directed mutagenesis ( QuickChange , Stratagene ) and verified by sequencing . RRM3 truncations were created in plasmid pKHS137 and plasmid pJG4-5* using HR-mediated integration in KHSY2331 ( lig4Δ ) and verified by sequencing . Yeast was grown at 30°C in yeast extract ( 10g/l ) , peptone ( 20g/l ) , dextrose ( 20g/l ) , media ( YPD ) with or without Bacto agar , or in synthetic complete ( SC ) media ( yeast nitrogen base 6 . 7g/l , dextrose 20g/l ) supplemented with the appropriate amino acid mix . All yeast strains and plasmids used in this study are listed in S2 and S3 Tables , respectively . Gross-chromosomal rearrangement ( GCR ) rates were determined by fluctuation analysis by taking the median rate of at least 15 cultures from at least two isolates and are shown with 95% confidence intervals [78–80] . Cells with GCRs were identified by their resistance to canavanine and 5-fluoro-orotic acid ( Canr 5-FOAr ) , which is indicative of simultaneous inactivation of CAN1 and URA3 on chromosome V , on selective media as previously described [80] . To observe GCRs after exposure to DNA-damaging agents HU and MMS , cells were grown to an OD600 = 0 . 5 , released into medium containing the drug and were cultured for 2 hours at 30°C . Cells were then released into fresh YPD medium and grown for 24 hours before being plated . Forward mutation rates were determined by fluctuation analysis by method of the median as previously described and are shown with 95% confidence intervals [78 , 81 , 82] . Briefly , fifteen cultures from at least two different isolates were grown overnight at 30°C in 3 ml of YPD media . Dilutions were plated on YPD agar to determine the viable cell count , and 500 μl was plated on synthetic media supplemented with 60 μg/ml canavanine , but lacking arginine to select for can1 mutants . To obtain forward mutation rates and GCR rates after exposure to MMS and HU , cells were grown to OD600 = 0 . 5 , released into medium containing the drug and cultured for 2 hours at 30°C . Cells were then released into fresh YPD medium and grown for 24 hours before being subjected to fluctuation analysis . Cell cultures were grown as indicated either in YPD or selective media to maintain plasmids ( SC-Leu ) to OD600 = 0 . 5 , and 10-fold serial dilutions were spotted on YPD or SC-Leu supplemented with methyl methanesulfonate ( Sigma Aldrich ) or hydroxyurea ( US Biological ) at the indicated concentrations . Colony growth was recorded after 2 to 3 days of incubation at 30°C . Cells were prepared for DNA content analysis as previously described [83] . Briefly , cells were washed and fixed in 70% ethanol for 1 hour at room temperature , sonicated in 50 mM sodium citrate ( pH 7 ) , washed in 50 mM sodium citrate ( pH7 ) , and RNAse A was added to a final concentration of 250 μg/ml . After overnight incubation at 37°C , cells were washed in 50 mM sodium citrate and stained with Sytox green ( Molecular Probes ) at a final concentration of 1 μM in the dark at room temperature for 1 hour immediately prior to fluorescence-activated cell sorting ( FACS ) on a BLD LSR II analyzer . The distribution of cells throughout the cell cycle phases was quantified with the FlowJo v8 . 3 . 3 software . Cells were grown to OD600 = 0 . 5 in YPD at 30°C , synchronized in G1 with α-factor ( 15 μg/ml ) , released into fresh , pre-warmed YPD . Cells were harvested after 30 min , immediately put on ice and adjusted for cell number . Whole cell extracts were prepared with 20% trichloroacetic acid as previously described [84] and separated by 10% SDS-PAGE for Western blot analysis . Phospho-specific Rad53 antibody EL7 was a gift from A . Pellicioli ( FIRC Institute of Molecular Oncology Foundation , Milan , Italy ) . Antibody sc-6680 ( SCBT ) was used for Mcm2 , ab34680 ( Abcam ) for Adh1 , ab46765 ( Abcam ) for histone H3 , AS07214 ( Agrisera ) for Rfa1 , and MMS-150P ( Covance ) for the myc-epitope . Cells expressing Orc5-V5-6xHis and either Rrm3-myc , rrm3-ΔN186-myc , or rrm3-ΔN212-myc were grown to an OD600 ~ 1 . 0 and cells were harvested by centrifugation . Cells were re-suspended in cell lysis buffer ( 50 mM HEPES pH 7 . 5 , 10% v/v glycerol , 140 mM NaCl , 1 mM EDTA , 0 . 5% Igepal , 1 mM PMSF , EDTA-free Protease inhibitors ( Pierce ) ) and vortexed in a cell disruptor with glass beads for 5 minutes . Cell lysates were cleared by centrifugation at 4°C and lysates were split in half . 1 mM PMSF and 20 mM MgCl2 was added to all samples , and 300 U of Benzonase ( Sigma ) was added to one half of the sample whereas the other half was not treated . Cell lysates were placed on ice for 30 minutes , added magnetic beads coated with to C-MYC antibody and incubated with mixing for 2 h at 4°C . Beads were washed thoroughly 10 times in 1 ml of cell lysis buffer , re-suspended in protein sample buffer , and boiled for 5 min . Samples were separated on a 7% SDS-PAGE gel and presence of Rrm3 . myc , rrm3-ΔN186 . myc , rrm3-ΔN212 . myc , Orc5 . V5 . 6xHIS and GAPDH was determined by western blotting using monoclonal antibodies against myc ( Covance ) , V5 ( Sigma ) , and GAPDH ( Pierce ) . Fifty millilters of cells corresponding to 2 . 0 x 107 cells/ml were incubated for 15 min at room temperature with or without 1% formaldehyde and harvested . Cell pellets were washed twice with PBS , re-suspended in 600 μl of cell lysis buffer ( 50 mM HEPES pH 7 . 5 , 140 mM NaCl , 1 mM EDTA , 1% Triton X-100 , 0 . 1% sodium deoxycholate , 1 mM PMSF , 1 mM benzamidine , 1 μg/ml aprotinin , 1 μg/ml leupeptine , and 1 ug/ml pepstatin ) , lysed in a cell disruptor with glass beads at 4°C , and sonicated four times for 20 s each at 4°C . The lysate was clarified by centrifugation and whole cell extract was added to magnetic beads coated with c-myc antibody or V5 antibody , followed by incubation for 60 min at 4°C . Beads were washed four times in cell lysis buffer , three times in TE buffer ( 10 mM Tris-Cl , pH 7 . 5 , 1 mM EDTA ) , and re-suspended in TE/1% SDS buffer . Sample was incubated at 65°C for 10 min and processed for DNA purification as previously described [85] . Sequences of primers used for PCR are available upon request . Cells grown to stationary phase were transferred to acidic media ( pH 3 . 5 ) and grown to logarithmic phase . Cells were synchronized in G1 phase over two hours with the addition of 2 μg/ml alpha-factor ( Genscript ) every hour . Cells were washed twice with sterile water . 2 . 5×108 yeast cells were pelleted , resuspended in 1 ml 60% methanol , and disrupted by 5 consecutive freeze and thaw cycles using liquid nitrogen and warm water , followed by incubation at -20°C for 90 minutes and boiling at 100°C for 3 minutes . The lysate was centrifuged at 19000×g for 15 minutes and the supernatant frozen in liquid nitrogen . Methanol was evaporated in a SpeedVac ( Thermo Scientific ) and the residue was rehydrated in 100 μl Ultra-pure H2O ( Invitrogen , GIBCO ) . Determination of cellular dNTP concentration was performed as earlier described [86] . Each extraction was performed at least in triplicate . Yeast strain EGY48 containing reporter plasmid pSH18-34 was co-transformed with a lexA-fusion bait vector ( pEG202 ) and a B42-tagged prey vector ( pJG4-5* ) , and transformants were selected on synthetic complete ( SC ) medium plates lacking histidine and tryptophan ( SC-Trp-His ) . Single transformants were purified and resuspended in liquid SC-His-Trp medium containing 2% galactose and 1% raffinose , and grown overnight at 30°C . Cultures were diluted to OD600 = 0 . 2 , grown to OD600 = 0 . 8 , and 2-fold serial dilutions were spotted onto SC-His-Trp-Leu containing either 2% glucose or 2% galactose with or without hydroxyurea ( HU ) or methyl methanesulfonate ( MMS ) . Colony growth was recorded after 72 hours . Diploids heterozygous for the desired mutant alleles ( rrm3::TRP1 , mrc1::HIS3 ) transformed with plasmid-borne alleles of RRM3 ( linked to LEU2 ) were sporulated by nitrogen starvation in 0 . 1% potassium acetate for 5 days at 30°C with vigorous shaking . Asci were incubated in the presence of 500 μg/ml of zymolase in 1M sorbitol for 20 min at 30°C , enriched for meiotic products as previously described [87] and plated on nonselective media ( YPD ) . After incubation for 2 days at 30°C , colonies were spotted on SC-Leu media and 100 leu+ colonies genotyped further by spotting on SC-Leu-Trp , SC-Leu-His , and SC-Leu-Trp-His to identify rrm3Δ , mrc1Δ and rrm3Δ mrc1Δ mutants , respectively , all harboring various plasmid borne RRM3 alleles . Cells were grown to early log phase and incubated with 100 mM hydroxyurea for 2 hours at 30°C with shaking . Cells were fixed with 3 . 7% formaldehyde , permeablized with ethanol , and mounted in Vectashield medium with DAPI ( Vectorlabs ) . Cells were examined using an EVOS fluorescence microscope and grouped into unbudded ( G1 phase ) , small-budded ( S phase , bud is up to one-fourth of the diameter of mother cell ) , or large-budded ( G2/M , bud is equal to or greater than one-fourth of the diameter of the mother cell ) . 200 cells for each yeast strain were scored .
|
When cells duplicate their genome , the replication machinery is constantly at risk of encountering obstacles , including unusual DNA structures , bound proteins , or transcribing polymerases and transcripts . Cells possess DNA helicases that facilitate movement of the replication fork through such obstacles . Here , we report the discovery that one of these DNA helicases , Rrm3 , is also required for restricting DNA synthesis under replication stress . We find that the site in Rrm3 critical for this new replication function is also required for binding a subunit of the replication origin recognition complex , which raises the possibility that Rrm3 controls replication by affecting initiation . This is supported by our finding that Rrm3 associates with a subset of replication origins . Rrm3’s ability to restrict replication does not require its helicase activity or the phosphorylation site that regulates this activity . Notably , cells need error-free bypass pathways and homologous recombination to deal with DNA lesions that arise when the helicase function of Rrm3 is disrupted , but not when its replication function is disrupted . This indicates that the DNA lesions that form in the absence of the two distinct Rrm3 function are different , although both activate the DNA-damage checkpoint and are toxic to cells that lack the mediator of the replication checkpoint Mrc1 .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"nucleic",
"acid",
"synthesis",
"enzymes",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"enzymology",
"dna",
"damage",
"mutation",
"dna",
"replication",
"dna",
"epigenetics",
"synthesis",
"phase",
"chromatin",
"dna",
"synthesis",
"chemical",
"synthesis",
"research",
"and",
"analysis",
"methods",
"chromosome",
"biology",
"proteins",
"gene",
"expression",
"biosynthetic",
"techniques",
"biochemistry",
"point",
"mutation",
"helicases",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences"
] |
2016
|
A Novel Rrm3 Function in Restricting DNA Replication via an Orc5-Binding Domain Is Genetically Separable from Rrm3 Function as an ATPase/Helicase in Facilitating Fork Progression
|
Myeloid dendritic cells ( mDC ) are lost from blood in individuals with human immunodeficiency virus ( HIV ) infection but the mechanism for this loss and its relationship to disease progression are not known . We studied the mDC response in blood and lymph nodes of simian immunodeficiency virus ( SIV ) -infected rhesus macaques with different disease outcomes . Early changes in blood mDC number were inversely correlated with virus load and reflective of eventual disease outcome , as animals with stable infection that remained disease-free for more than one year had average increases in blood mDC of 200% over preinfection levels at virus set-point , whereas animals that progressed rapidly to AIDS had significant loss of mDC at this time . Short term antiretroviral therapy ( ART ) transiently reversed mDC loss in progressor animals , whereas discontinuation of ART resulted in a 3 . 5-fold increase in mDC over preinfection levels only in stable animals , approaching 10-fold in some cases . Progressive SIV infection was associated with increased CCR7 expression on blood mDC and an 8-fold increase in expression of CCL19 mRNA in lymph nodes , consistent with increased mDC recruitment . Paradoxically , lymph node mDC did not accumulate in progressive infection but rather died from caspase-8-dependent apoptosis that was reduced by ART , indicating that increased recruitment is offset by increased death . Lymph node mDC from both stable and progressor animals remained responsive to exogenous stimulation with a TLR7/8 agonist . These data suggest that mDC are mobilized in SIV infection but that an increase in the CCR7-CCL19 chemokine axis associated with high virus burden in progressive infection promotes exodus of activated mDC from blood into lymph nodes where they die from apoptosis . We suggest that inflamed lymph nodes serve as a sink for mDC through recruitment , activation and death that contributes to AIDS pathogenesis .
Myeloid dendritic cells ( mDC ) are professional antigen-presenting cells that are critical for the induction of acquired immune responses to pathogens [1] . Depletion of mDC from blood in human immunodeficiency virus ( HIV ) infection has been well described and shown to be inversely correlated with virus load and absent from long-term non-progressors , suggesting a relationship between mDC and disease control [2]–[8] . A proposed mechanism to account for mDC loss from blood is their activation and subsequent recruitment to inflamed lymph nodes [9] . Increased expression of costimulatory molecules on blood mDC indicative of activation has been reported in HIV-infected individuals [3] , [5] , [10] , as has accumulation of mDC in peripheral lymph nodes during acute infection [11] . However , findings relating to mDC in lymph nodes during chronic HIV infection are inconsistent , with both accumulation [12] , [13] and substantial loss of mDC [14] being reported . mDC are depleted from both blood and lymph nodes of simian immunodeficiency virus ( SIV ) -infected rhesus macaques during AIDS [15] but data are lacking from earlier stages of infection . Few studies have evaluated mDC dynamics in both blood and lymph node in the same individuals [12] , [15] and no longitudinal studies of mDC kinetics in both compartments have been reported . As such , the relationship between mDC loss and recruitment in infection remains ill-defined , and whether differences in mDC dynamics predict disease outcome is not known . The impact of antiretroviral therapy ( ART ) on mDC loss and recovery in HIV infection is also unclear , as several studies indicate that ART is not effective at increasing blood mDC [6] , [7] , [16] while others suggest that ART significantly restores blood mDC numbers [3] , [8] , [17] , [18] . ART rapidly resolves immune activation in lymphoid tissues [19] and may have beneficial effects on lymph node mDC activation and function [13] , although this has not been well characterized . In the present study we followed the mDC response in blood and lymph nodes over time in two cohorts of SIV-infected animals that received ART and adenovirus ( Ad ) -based immunotherapy with different disease outcomes . We find that loss of blood mDC at virus set-point is predictive of disease progression , whereas an increase in blood mDC is predictive of long-term absence of disease , and that even relatively short periods of ART are beneficial to mDC homeostasis . In animals that progress to AIDS the early loss of mDC from blood is associated with evidence of increased CCR7-CCL19-mediated recruitment to lymph nodes and increased apoptosis within these tissues .
Animals in this study were enrolled in an immunotherapy protocol using Ad-based vectors the majority of which has been previously described [20] . Animals were infected with the pathogenic isolate SIVmac251 by intravenous inoculation and received ART consisting of a combination of two reverse transcriptase inhibitors , 9-[2- ( phosphonyl-methoxy ) propyl]adenine ( PMPA ) and 2′-deoxy-5-fluoro-3′-thia-cytidine ( FTC ) , from weeks 12 to 24 and weeks 32 to 44 , depending on survival . Immunotherapy consisted of priming with Ad serotype 5 ( Ad5 ) -based vectors expressing SIV Gag , Env and Nef with or without IL-15 at weeks 16 and 22 followed by boosting with Ad35-based vectors expressing the same transgenes at weeks 36 and 42 . Control-treated animals were given the same regimen of Ad5-ψ5 and Ad35-ψ5 vectors that lacked transgenes [20] . Ad-based immunotherapy boosted T cell responses to SIV but had no effect on virus load , progression to disease or survival [20] ( and data not shown ) . However , when analyzed independent of immunotherapy , animals in the cohort could be readily separated into two groups based on disease progression , with one group remaining healthy until elective sacrifice at a mean of 60 weeks post infection ( n = 11 , ‘stable’ group ) , and the other succumbing to AIDS with a mean survival time of 32 weeks ( n = 10 , ‘progressor’ group , Table 1 ) . AIDS was defined clinically by lymphadenopathy , persistent weight loss and anorexia , with or without opportunistic infections [15] . Equal numbers of animals in the stable and progressor groups received Ad-based immunotherapy with the remainder receiving control vectors or no treatment , confirming the lack of association between immunotherapy and disease outcome ( Table 1 ) . The MHC class I molecule Mamu-A*01 was expressed by 3/11 and 0/10 animals in the stable and progressor groups , respectively , consistent with an association of this molecule with control of SIV infection ( Table 1 ) [21] . Peak plasma virus loads in stable and progressor animals at 2 weeks post infection were similar at around 2×107 RNA copies/ml plasma; however virus loads began to diverge by week 4 and at virus set point virus loads differed by ∼1 log between groups ( Table 1 and Figure 1 ) . Survival time was inversely correlated with virus load at set-point ( Table 1 and Figure 1B ) , consistent with previous reports [22] , [23] . ART beginning at week 12 had parallel although modest effects on virus load in both groups , with an immediate decrease of ∼1 . 5 logs that fluctuated over the course of therapy ( Figure 1 ) . Virus load persisted at ∼0 . 5 logs below set-point levels after discontinuation of ART and then decreased by ∼1 . 5 logs with initiation of the second cycle of ART at week 32 , again with fluctuations over the course of therapy ( Figure 1 ) . Only animals in the stable group survived beyond the second cycle of ART and in these animals virus load remained at ∼1 . 5 logs below set-point until sacrifice ( Table 1 and Figure 1 ) . These data show that disease progression and survival in this cohort of animals correlated with virus load at set-point prior to initiation of ART and not with Ad-based immunotherapy , and that ART was effective at inducing modest but similar decreases in virus load in both groups . The characteristics of this cohort allowed us to ask whether differences in eventual disease outcome were reflected in earlier changes in the mDC response and whether short-term exposure to ART was beneficial to this response . Blood mDC were identified in peripheral blood mononuclear cells ( PBMC ) as CD45+ lineage− HLA-DR+ CD11c+ cells ( Figure 2A ) and were enumerated based on the ratio of mDC to CD4+ T cells [20] , [24] . Staining of blood cells with antibody to CD11c was inconsistent in animals R487 ( stable group ) and M5406 ( progressor group ) making it difficult to delineate mDC at all time points ( data not shown ) , and as a result these animals were not studied further . The median number of blood mDC in the remaining 19 animals prior to infection was 51 cells/ul with a relatively large range from 16 to 202 cells/ul , consistent with our previous findings ( Figure 2C , D ) [24] . Blood mDC were reduced at 2 weeks post infection relative to baseline levels when all animals were analyzed together ( P = . 03 ) , although when each group was analyzed separately this decrease was not significant . However , in the post-acute period the mDC response diverged , as mDC in progressor animals continued to decline to week 12 when they were significantly reduced in number relative to preinfection time points . In contrast , mDC in stable animals significantly increased from weeks 2 to 12 ( Figure 2B–D ) . The relative change in the number of blood mDC in individual animals over the first 12 weeks of infection was significant , as mDC dropped to around 30% of preinfection levels in some progressor animals ( mean for the group 60% ) but increased to nearly 500% in some stable animals ( mean for the group 206% ) ( Figure 2E ) . This change was inversely correlated with virus load at week 12 post infection , revealing a relationship between viral burden and mDC homeostasis ( Figure 2E ) . Exposure to the first round of ART in progressor animals resulted in an increase in mDC number from weeks 12 to 20 , when virus load was near its lowest point , and appeared to stabilize the number of blood mDC in the stable group ( Figure 2B–D ) . However , after ART was discontinued at week 24 the number of mDC in stable but not progressor animals rose markedly reaching a mean increase of 3 . 5 fold over baseline at week 32 , with individual increases approaching 10-fold in some animals ( Figure 2B–D ) . The magnitude of the mDC response after ART was not influenced by the vaccination regimen received during ART , as a comparison of mean mDC counts for all animals from weeks 28 to 32 ( the period of greatest response ) based on the type of immunotherapy received revealed no statistically significant differences ( data not shown , Kruskal-Wallis test , P = 0 . 3 ) . Initiation of the second round of ART again reduced mDC number in the majority of stable animals , concurrent with the reduction in virus load , after which the number of mDC remained relatively constant ( Figure 2B , D ) . The divergent mDC response contrasted with changes in CD4+ T cells , which did not statistically differ between groups at any time before or after infection ( Figure 2F ) . These data indicate that differences in eventual disease outcome in SIV infection are reflected by differences in the blood mDC response that are apparent relatively early in infection . They also indicate that short-term ART may be effective at transiently restoring blood mDC in animals with the most severe disease . We next asked whether differences in disease progression were reflected in earlier differences in activation of circulating mDC in SIV-infected macaques . For these and subsequent analyses we focused on the first 32 weeks of infection incorporating one 12-week cycle of ART and one 8-week period of treatment interruption , as after this time the number of animals surviving in the progressor group rapidly diminished ( Table 1 ) . Expression of the costimulatory molecules CD80 and CD86 was markedly increased in all animals at 2 weeks post infection indicative of rapid mDC activation ( Figure 3A–D ) . However , by 12 weeks post infection differences in mDC activation were evident between groups particularly with respect to the chemokine receptor CCR7 , which was expressed by a significantly greater proportion of mDC in animals that progressed to AIDS relative to animals with stable infection ( Figure 3A ) . A majority of CCR7+ mDC in progressor animals also expressed CD86 with a smaller proportion expressing CD80 , consistent with activation ( Figure 3E ) . The 12-week course ART was effective at reducing blood mDC activation , particularly with respect to CD80 , and in animals in the stable group expression of all markers of activation returned to preinfection levels during ART ( Figure 3A–C ) . The increase in CD80 at week 32 suggested mDC were again activated during the period of ART discontinuation , although no increase in CCR7 or CD86 expression was noted at this time ( Figure 3A–C ) . These findings indicate that during chronic infection animals that progress to AIDS have increased blood mDC activation relative to animals with stable infection . They also confirm that ART has the beneficial effect of reducing mDC activation , consistent with findings in HIV-infected humans [3] , [12] . The finding that loss of blood mDC in progressor animals occurs as the proportion of mDC expressing CCR7 increases is consistent with excessive mDC recruitment to lymph nodes via the CCR7/CCL19/CCL21 pathway , as has been suggested by in vitro studies [9] . To examine this potential in vivo , we used flow cytometry to identify mDC in lymph node cell suspensions taken prior to infection and at intervals after infection in our two groups of animals . mDC were defined as lineage− HLA-DR+ CD11c+ cells ( Figure 4A ) and enumerated as a proportion of all cells in the lineage− HLA-DR+ gate , which we have previously shown to be an accurate indicator of the absolute number of mDC per unit of weight [15] . Surprisingly , we found no significant difference in the number of lymph node mDC as a result of SIV infection regardless of disease progression , indicating a lack of mDC accumulation ( Figure 4B ) . However , the phenotype of mDC within lymph nodes was significantly different as a function of disease , as animals with stable but not progressive infection had a lower percentage of mDC expressing CCR7 , CD40 and CD86 and reduced mDC expression of MHC class II at 12 weeks relative to preinfection time points , reflecting reduced mDC activation ( Figure 4C , D ) . To address the issue of mDC recruitment further , we next used real time PCR to determine the relative expression of CCR7 ligands in lymph node tissues . We found that CCL19 but not CCL21 mRNA was increased 8-fold in lymph nodes at 12 weeks post infection , but only in animals that progressed to AIDS ( Figure 4E ) . Together with our findings in blood , these data suggest that mDC are recruited to lymph nodes in progressive disease via an enhanced CCR7/CCL19 pathway , but that expanded mDC recruitment fails to result in mDC accumulation . The lack of mDC accumulation in lymph nodes despite evidence for enhanced CCR7/CCL19-mediated recruitment in progressive infection led us to suspect that lymph node mDC were dying at an increased rate in these tissues . To examine this possibility we identified live mDC in lymph node cell suspensions as being lineage− HLA-DR+ CD11c+ cells that lacked staining with a fixable dead-cell dye , and then identified cells undergoing early apoptosis within this gate using an antibody to active caspase-3 ( Figure 5A ) . At week 12 post infection , 15% of lymph node mDC in animals that eventually progressed to AIDS were entering apoptosis , representing a 3-fold increase from preinfection levels , whereas lymph node mDC from animals with stable infection had no significant change in apoptosis ( Figure 5B ) . ART given from week 12 to 24 post infection decreased the frequency of apoptotic mDC in progressor animals , although this did not reach statistical significance ( Figure 5B ) . To determine whether apoptosis was mediated by extrinsic or intrinsic pathways we exposed lymph node cells from progressor animals taken at week 12 post infection to small molecule inhibitors of caspase-8 or caspase-9 , respectively . The presence of caspase-8 inhibitor Z-IETD-FMK reduced apoptosis by more than 50% relative to a control peptide whereas the caspase-9 inhibitor had minimal effect ( Figure 5C ) , suggesting that cell-extrinsic mediators of apoptosis were predominant . Consistent with this finding , lymph node mDC taken from progressor but not stable animals at week 12 post infection showed a significant increase in the proportion of mDC expressing CD95 relative to preinfection samples ( Figure 5D ) . Together , these data suggest that increased mDC apoptosis in lymph nodes during chronic infection in animals that progress to AIDS offsets the increase in mDC recruitment from blood , resulting in no net accumulation of mDC . Changes in mDC activation and apoptosis within lymph nodes during SIV infection could impact the capacity of these cells to respond to microbial stimuli and subsequently induce adaptive T cell immune responses . To investigate the functional capacity of mDC following SIV infection in our two groups of animals we stimulated lymph node cell suspensions taken at intervals before and after infection with 3M-007 , a small molecule synthetic agonist of TLR7/8 , which , like HIV and SIV RNA , activates mDC through their engagement of TLR8 [25]–[27] . We analyzed mDC for expression of two key immunoregulatory cytokines , TNF-α and IL-12 ( p40/p70 ) . Interestingly , lymph node mDC taken prior to infection responded relatively poorly to short-term stimulation with a small proportion of cells producing TNF-α and IL-12 ( Figure 6 ) . In contrast , stimulation of mDC taken at 12 weeks post infection resulted in 20 to 30% of cells producing TNF-α and a smaller but significant percentage producing IL-12 , representing a 4- to 5-fold increase above preinfection levels regardless of disease outcome ( Figure 6 ) . ART reduced mDC responsiveness to TLR8 stimulation although this did not reach statistical significance ( Figure 6B ) . These data indicate that mDC resident in lymph nodes of SIV-infected rhesus macaques are functional capable of responding to stimulation , and may even be hyperresponsive as a consequence of SIV infection .
In this study we examined the relationship between mDC dynamics and disease progression over time in pathogenic SIV infection of rhesus macaques . We show for the first time that mDC are preferentially lost from blood in animals that progress to AIDS but are increased in blood of animals with long-term stable infection . This divergent mDC response was apparent at virus set-point , indicating that changes in blood mDC number over the first 3 months of infection are predictive of eventual disease progression . mDC are recruited from blood to lymphoid tissues through upregulation of CCR7 , the ligand for chemokines CCL19 and CCL21 that are expressed in the lymph node paracortex [28] . In animals with progressive infection , mDC loss from blood was associated with an increase in the frequency of blood mDC expressing CCR7 and an increase in expression of CCL19 in lymph nodes , consistent with increased extravasation to lymph nodes that exceeded the rate of mDC production from bone marrow . Expression of CCL19 has been shown previously to be markedly increased in lymph nodes during the acute phase of SIV infection [29] , and our data suggest that expression in lymph nodes remains high into chronic infection as a function of virus load . Indeed , recent studies have shown that increased levels of CCL19 and CCL21 in blood correlate with higher virus loads and disease progression in HIV infected humans [30] . In vitro exposure to CCL19 and CCL21 also promotes an inflammatory response in PBMC from HIV-infected individuals with high virus loads [31] . We now provide evidence of a functional link between CCL19 upregulation in lymph nodes and increased expression of CCR7 on circulating mDC that promotes mDC recruitment to lymph nodes in progressive SIV infection . While not examined in this study , the potential exists for proinflammatory factors to promote differential emigration of mDC to lymph nodes in progressive relative to stable SIV infection . In particular , lipopolysaccharide is increased in the circulation during chronic HIV and pathogenic SIV infection as a consequence of microbial translocation through increased gut permeability [32] . Lipopolysaccharide activates mDC via engagement of TLR4 [33] and is a potent activator of DC migration in vivo [34] , [35] . In contrast to progressive infection , we found that mDC in animals that controlled SIV infection had significant increases in blood mDC over time , with increases of up to 5-fold by virus set-point and nearly 10-fold in some cases at 32 weeks of infection . Studies in HIV infected individuals have indicated that mDC loss is inversely proportional to virus load , as we have shown , and is not observed in long-term non-progressors [2] , [5] , [8] , but such cross sectional studies have by design not revealed changes over time . Increased blood mDC may arise from increased hematopoiesis in bone marrow in response to inflammatory cytokines such as TNF-α and IL-1 that are elevated during HIV infection and promote DC generation [36] , [37] . The lack of an upregulated CCR7-CCL19 axis in this group would exacerbate the impact of enhanced DC production and mobilization into blood by limiting mDC exodus into tissues . Paradoxically , there was no net increase in mDC within lymph nodes in monkeys with progressive SIV infection , associated with an increase in mDC apoptosis , suggesting that increased recruitment to lymph nodes is offset by increased cell death in severe infection . mDC apoptosis was caspase-8-dependent and associated with increased CD95 expression , similar to the findings for plasmacytoid DC in HIV and SIV infection [38] , [39] , consistent with a cell-extrinsic mechanism of apoptosis involving CD95 ligation . Apoptosis through virus infection of mDC is unlikely to be a significant factor , as previous studies indicate that only a minor fraction of lymph node mDC contain incorporated viral DNA during peak viremia [38] . HIV and SIV clearly affect mDC in the absence of productive infection , in particular through interactions of viral RNA with endosomal TLR8; however this interaction tends to promote cell survival rather than apoptosis [9] , [27] . While the increase in mDC recruitment appears to keep pace with apoptosis in tissues during the chronic stages of SIV infection studied here it is clear that mDC are ultimately lost from lymph nodes as AIDS is established , as previously reported [15] . This eventual decline may be associated with a similar decline in lymph node expression of CCL19 in the final stages of disease [29] . Several reports have described the presence of semimature mDC with reduced expression of costimulatory molecules and/or CD83 in lymph node and spleen of HIV-infected humans [11] , [13] , [40] and SIV-infected macaques [41] , [42] . Our data now suggest that these cells may have a beneficial function in vivo , as lymph node mDC with significantly lower expression of CCR7 and costimulatory molecules consistent with a semimature state were present only in animals with long-term stable infection . In vitro , semimature DC with tolerogenic function are derived from exposure to immunoregulatory cytokines including IL-10 and transforming growth factor-β [43] , however whether these factors modulate DC maturation and function in progressive versus stable SIV and HIV infection is not known . Semimature mDC from HIV-infected lymph nodes have been shown to promote regulatory T cell function [13] . While we were not able to examine the effect of these cells on regulatory T cells in this study , the prevalence of semimature mDC in stable but not progressive infection might suggest a role for enhanced regulatory T cell responses in disease control . The role of regulatory T cells in pathogenic and nonpathogenic SIV infection is currently controversial [44]–[46] , and the interplay between mDC and regulatory T cells in control and progression to disease deserves attention . In contrast to stable infection , mDC in lymph nodes of animals with progressive infection showed essentially no difference in expression of CCR7 and activation markers relative to naïve animals , although the proportion of cells expressing these markers was substantially greater than in blood . It is possible that activated mDC undergo apoptosis immediately upon entering lymph nodes , or alternatively that other newly identified costimulatory molecules from the CD28 and TNFR families not examined here may be differentially expressed in progressively infected lymph nodes [47] . In our study the two short courses of ART had only modest although similar effects on virus load in both groups of animals , reducing virus levels in plasma by ∼1 . 5 logs . This may be due to the fact that ART was initiated in chronic as opposed to acute infection and that therapy was limited to PMPA and FTC which both target the same viral protein , reverse transcriptase . Similarly limited effects of ART on virus load in SIV infection have been reported by others [48] , [49] . Despite this , ART had noticeably beneficial effects on mDC homeostasis . In blood , ART reduced mDC activation and transiently restored mDC numbers in monkeys with progressive infection , consistent with reports in HIV-infected individuals [8] , [17] , [18] . Most strikingly , discontinuation of ART in stable animals led to a marked increase in the number of mDC in blood . In lymph nodes , ART resulted in a decrease in mDC apoptosis in animals with progressive infection and a reduction in mDC responsiveness overall . Consistent with this finding , expression of proinflammatory factors and CD95L that likely induce functional activation and apoptosis of mDC are substantially reduced in SIV- and HIV-infected lymph nodes in response to ART [19] , [50] , [51] . Animals in the progressive infection group died at a median time of 34 weeks post infection and did not receive the full second course of ART initiated at week 32 . We do not believe this difference in treatment interval was a determining factor in survival , as disease status and time to sacrifice were correlated with virus load at set-point , before initiation of any therapy , and thus were independent of ART . HIV-infected individuals with higher baseline virus loads and immune activation have poorer reconstitution of innate immune cells in response to ART [16] , [52] . In our study , differences in baseline virus load influenced the response to ART , as animals with stable and progressive infection had transient increases and decreases , respectively , in the number of blood mDC , although this could clearly be influenced by the differences in virus load in the two groups while on ART . It will be important to determine the impact of improved antiretroviral drug regimens on mDC dynamics in SIV infection , including the orally available integrase inhibitors that are highly active in monkeys [53] . Our data indicate that mDC present in lymph nodes in SIV infected monkeys remain functionally responsive to exogenous stimulation regardless of disease outcome . The finding that ex vivo stimulation through TLR8 induced a five-fold increase in expression of TNF-α relative to naïve animals suggests that these cells may in fact be hyperresponsive , although testing with a more extensive panel of agonists targeting different TLR ligands is needed to confirm this . CCL19 induces terminal activation of DC and promotes DC production of proinflammatory cytokines within lymph nodes [54] , although this effect would not explain the finding of increased responsiveness of mDC in animals with stable infection that had normal levels of CCL19 in our study . It is possible that other proinflammatory factors such as IFN-γ that induce DC activation [55] and are markedly increased in lymph nodes during pathogenic SIV infection [56] are responsible for mDC increased responsiveness in SIV infection . An increasing emphasis in HIV and SIV pathogenesis is now placed on the role of gut mucosa in disease , as this is a major site of virus replication and CD4+ T cell depletion [57]–[59] . mDC are recruited to inflamed respiratory mucosal surfaces in children with respiratory viral infections [60] , and it is likely mDC and other DC subsets are similarly recruited to gut and vaginal mucosa in SIV infection [61] . It will be important to evaluate the mDC response in gut mucosa and its relationship to disease progression in SIV infection . However , such quantitative studies are technically difficult to perform as the DC is a relatively rare cell that can only be isolated in sufficient numbers through gut resection surgeries as opposed to the more commonly performed endoscopic biopsies . Collectively , these data suggest that the inflammatory response associated with increased virus load during progressive SIV infection leads to an increase in the CCR7-CCL19 chemokine axis that serves to accelerate mDC recruitment to lymph nodes . Apoptosis of mDC within tissues during this chronic phase , which was found only in animals with progressive infection , would compromise the innate and adaptive immune response to opportunistic pathogens promoting disease progression . It is currently not clear whether recently recruited and activated mDC produce increased levels of proinflammatory cytokines in vivo that may mediate immune activation characteristic of HIV and pathogenic SIV infection [62] . Interestingly , increased turnover of blood monocytes associated with apoptosis of tissue macrophages has been shown to correlate with progression to disease in SIV-infected macaques and is a better predictive marker than viral load or lymphocyte activation [63] , [64] . This response is not likely limited to lymph nodes , as evidenced by the fact that increased monocyte turnover and recruitment to brain correlates with the severity of SIV encephalitis [65] . These findings point to a broad-based dysregulation of mDC and monocytes in blood and tissues as a significant factor in the pathogenesis of AIDS .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal Care and Use Committee at the University of Pittsburgh ( Assurance Number A3187-01 ) . Surgeries were performed under anesthesia induced and maintained with ketamine hydrochloride and medetomidine hydrochloride , and all efforts were made to minimize suffering . Twenty one Indian-origin rhesus macaques ( Macaca mulatta ) used in this study were housed at the University of Pittsburgh Primate Facility for Infectious Disease Research . All animals were infected by intravenous inoculation with 1 , 000 TCID50 of uncloned , pathogenic SIVmac251 ( provided by Christopher J . Miller , California National Primate Research Center ) . Virus load in plasma was determined as described previously [66] . ART consisted of PMPA ( 20 mg/kg/d , subcutaneous injection ) and FTC ( 30 mg/kg/d , subcutaneous injection; both provided by Michael Miller , Gilead Sciences ) and was given from weeks 12–24 and from weeks 32–44 as described [20] . All animals except R189 ( sacrificed at week 11 post infection ) received one or more administrations of Ad-based vectors during the study depending on survival . Priming injections of separate Ad5-based vectors expressing codon-optimized SIVmac239 gag , env and nef with and without rhesus IL-15 . FLAG or empty Ad-ψ5 were given by intramuscular injection at week 16 ( 5×1010 total viral particles ) and week 22 ( 1×1011 total viral particles ) , and boosting injections of the same quantity of Ad35-based vectors expressing the same transgenes were given at week 36 and 42 . All Ad vectors were E1/E3-deleted with the exception of Ad35 containing the env gene which was E3 deleted . Lymph nodes were taken from the axillary or inguinal regions prior to infection and at weeks 12 , 24 and 32 post infection and single cell suspensions were generated by disruption and digestion with collagenase D , as described [67] . Identification of mDC was performed as previously described with some modifications [15] , [24] . Briefly , PBMC or lymph node cell suspensions were stained with fluorescently-labeled antibodies to Lineage markers [CD3 ( clone SP34-2; all antibodies from BD Bioscience unless otherwise noted ) , CD14 ( M5E2 ) , and CD20 ( 2H7 ) ] , HLA-DR ( G46-6 ) and CD11c ( S-HCL-3 ) , with and without antibodies to CD45 ( D058-1283 ) , CD80 ( L307 . 4 ) , CD86 ( FUN-1 ) , CCR7 ( 150503 , R&D Systems ) , CD40 ( 5C3 ) and CD95 ( DX2 ) . An amine-reactive fixable dead-cell dye ( Invitrogen ) was used to discriminate live from dead cells . mDC were defined as Lineage− HLA-DR+ cells expressing CD11c . In lymph nodes a broad Lineage− HLA-DR+/++ gate was used to include all mDC as described previously The number of blood CD4+ T cells was quantified using a precise volume of blood stained with antibodies in the absence of any wash step in TruCOUNT tubes ( BD Biosciences ) that contained a known number of fluorescent beads to provide internal calibration , as previously reported [20] . The number of blood mDC was then calculated based on the ratio of mDC to CD4+ T cells in PBMCs at the same time point [24] . All analyses were done on an LSR II flow cytometer with FACSDiva software ( BD Bioscience ) . Intracellular cytokine production by lymph node mDC was measured as described previously for plasmacytoid DC with minor modifications [38] . Briefly , cell suspensions were cultured for 5 hours with 10 µM of the TLR7/8 agonist 3M-007 ( 3 M Pharmaceuticals ) with and without the addition of 10 µg/mL brefeldin A ( Sigma ) after 1 hour . Cells were stained with surface-labeling antibodies as above and fixed and permeabilized prior to incubation with antibody to TNF-α ( MAb11 ) and IL-12 ( 8 . 6 , Mitenyi Biotec ) and analysis by flow cytometry . To detect apoptosis in mDC , lymph node cell suspensions were cultured in media for 3 hours with and without caspase-8 inhibitor Z-IETD-FMK , caspase-9 inhibitor Z-LEHD-FMK or irrelevant peptide Z-Fa-FMK ( BD Biosciences ) . Cells were stained with surface-labeling antibodies as above and fixed and permeabilized prior to incubation with antibody to active caspase-3 ( C92-605 ) and analysis by flow cytometry . Total lymph node RNA was extracted and purified from cell suspensions generated from biopsies taken prior to or 12 weeks after infection using the RNAeasy kit ( Qiagen ) after treatment with DNAse I ( Invitrogen ) . cDNA was synthesized using random primers and Superscript II reverse transcriptase ( Invitrogen ) . Primers and probes from Taqman human gene expression arrays ( Applied Biosystems , Foster City , CA ) were utilized for real time PCR analysis of CCL19 , CCL21 and β-glucuronidase expression as previously described [68] . mRNA expression levels for each gene were calculated with the 2−ΔCT method using β-glucuronidase as the endogenous control [69] . Comparisons between two groups were carried out using the Mann-Whitney U test . Comparison of DC numbers across different time points was carried out using the Wilcoxon signed-rank test . Correlations were determined using the non-parametric Spearman rank test . Graphpad Prism 5 ( Graphpad Software ) was used for statistical analysis . All P values are two-sided with significance considered to be P<0 . 05 . The identification of genes analyzed in this paper as defined by Entrez-Gene are 574386 ( CCL19 ) , 574183 ( CCL21 ) and 677692 ( β-glucuronidase ) .
|
Myeloid dendritic cells ( mDC ) are essential innate immune system cells that are lost from blood in human immunodeficiency virus infection through an ill-defined mechanism . We studied the kinetics of the mDC response in blood and lymph nodes of rhesus macaques infected with the closely related simian immunodeficiency virus . We found that differences in the number of blood mDC correlated with eventual disease outcome , as at virus set-point mDC were increased in blood in animals remaining disease free but lost from blood in animals that progressed rapidly to AIDS . mDC loss was linked to an increase in the chemokine axis responsible for mDC recruitment to lymph nodes; however , mDC did not accumulate in tissues but rather died from apoptosis . Lymph node mDC remained responsive to stimulation with a TLR7/8 agonist during infection . Importantly , mDC dysregulation was partially reversed by antiretroviral therapy . These data indicate that chronic mDC recruitment , activation and death within lymph nodes precede development of disease in SIV infected monkeys and may play a role in AIDS pathogenesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"immunology/immune",
"response",
"immunology/innate",
"immunity",
"infectious",
"diseases/hiv",
"infection",
"and",
"aids",
"pathology/immunology"
] |
2010
|
Early Myeloid Dendritic Cell Dysregulation is Predictive of Disease Progression in Simian Immunodeficiency Virus Infection
|
Two modes of plant immunity against biotrophic pathogens , Effector Triggered Immunity ( ETI ) and Pattern-Triggered Immunity ( PTI ) , are triggered by recognition of pathogen effectors and Microbe-Associated Molecular Patterns ( MAMPs ) , respectively . Although the jasmonic acid ( JA ) /ethylene ( ET ) and salicylic acid ( SA ) signaling sectors are generally antagonistic and important for immunity against necrotrophic and biotrophic pathogens , respectively , their precise roles and interactions in ETI and PTI have not been clear . We constructed an Arabidopsis dde2/ein2/pad4/sid2-quadruple mutant . DDE2 , EIN2 , and SID2 are essential components of the JA , ET , and SA sectors , respectively . The pad4 mutation affects the SA sector and a poorly characterized sector . Although the ETI triggered by the bacterial effector AvrRpt2 ( AvrRpt2-ETI ) and the PTI triggered by the bacterial MAMP flg22 ( flg22-PTI ) were largely intact in plants with mutations in any one of these genes , they were mostly abolished in the quadruple mutant . For the purposes of this study , AvrRpt2-ETI and flg22-PTI were measured as relative growth of Pseudomonas syringae bacteria within leaves . Immunity to the necrotrophic fungal pathogen Alternaria brassicicola was also severely compromised in the quadruple mutant . Quantitative measurements of the immunity levels in all combinatorial mutants and wild type allowed us to estimate the effects of the wild-type genes and their interactions on the immunity by fitting a mixed general linear model . This signaling allocation analysis showed that , contrary to current ideas , each of the JA , ET , and SA signaling sectors can positively contribute to immunity against both biotrophic and necrotrophic pathogens . The analysis also revealed that while flg22-PTI and AvrRpt2-ETI use a highly overlapping signaling network , the way they use the common network is very different: synergistic relationships among the signaling sectors are evident in PTI , which may amplify the signal; compensatory relationships among the sectors dominate in ETI , explaining the robustness of ETI against genetic and pathogenic perturbations .
Pattern Triggered Immunity ( PTI ) and Effector Triggered Immunity ( ETI ) are forms of plant immunity defined by different modes of pathogen recognition [1]–[3] . In PTI , pattern recognition receptors ( PRRs ) in the plant plasma membrane recognize molecular structures characteristic of microbes ( microbe-associated molecular patterns; MAMPs ) [4] , [5] . In ETI , a resistance ( R ) gene product , usually inside the plant cell , recognizes a corresponding virulence-promoting effector protein ( s ) delivered by a pathogen [6]–[9] . Recognition of pathogen attack in PTI and ETI leads to activation of partly overlapping sets of signaling sectors and defense responses [10] , [11] . Whereas pathogens well adapted to a particular host plant can prevail over PTI by effector interference with PTI signaling [12]–[15] , they typically overcome ETI by evading recognition , not by attacking ETI signaling [16] . In addition , most efforts to obtain mutants defective in ETI yielded only mutations in genes encoding R proteins and proteins required for R protein function . These observations suggest that the ETI signaling network is robust against pathogenic and genetic perturbations . However , the mechanism leading to robust immunity in ETI is yet to be determined . The salicylic acid ( SA ) signaling sector is generally important for immunity to biotrophs such as the bacterial pathogen , Pseudomonas syringae , while the jasmonic acid ( JA ) and ethylene ( ET ) signaling sectors are generally important for immunity to necrotrophs including the fungal pathogen , Alternaria brassicicola [17] . The SA and JA/ET sectors are mutually inhibitory in many cases [17] . Exogenously applied SA suppresses the JA sector and promotes susceptibility to A . brassicicola [18] . Conversely , P . syringae promotes virulence by suppressing the SA sector [19] by producing coronatine [20] , a molecular mimic of the active form of JA , JA-isoleucine [21] . In a highly interconnected network , analysis of mutants in which a single signaling sector is blocked provides only a limited view of network structure since phenotypes of single mutants reflect the effects of loss of the disrupted signaling sector as well as loss of interactions with other signaling sectors . As a result , network models derived from such data are incomplete and likely inaccurate . Here we use multiple-mutant analysis to reveal network properties responsible for robust immunity in plants .
To improve understanding of the signaling network in plant innate immunity , we constructed an Arabidopsis dde2/ein2/pad4/sid2-quadruple mutant . DDE2 , EIN2 , and SID2 are essential for JA biosynthesis [22] , most of the known ET responses [23] , and SA biosynthesis in response to pathogen attack [24] , respectively . PAD4 is important for SA accumulation in response to some SA-inducing stimuli [25] and has an SA-independent immune function [26] . Thus , the JA , ET , SA , and PAD4 signaling sectors are all blocked in the quadruple mutant . In the Arabidopsis accession Col-0 , the bacterial effector AvrRpt2 is recognized by the R protein , RPS2 , resulting in ETI [6] , [7] . For the purposes of this study , we defined immunity as the relative level of bacterial growth allowed within leaves . We quantified the level of AvrRpt2-ETI by subtracting log10-transformed bacterial number at two days post inoculation ( dpi ) of a P . syringae pv . tomato ( Pto ) DC3000 strain expressing AvrRpt2 from that of a Pto DC3000 strain carrying the empty vector pLAFR ( EV ) . We observed ∼80% reduction of AvrRpt2-ETI in the quadruple mutant compared to wild-type and reductions of 13% , 1% , 14% or 20% in dde2 , ein2 , pad4 or sid2 single mutants , respectively ( Figure 1A and 1B ) . Therefore , the effect of the quadruple mutation was larger than the sum of the effects of the four single mutations . We also conducted a similar experiment using a ten-fold lower inoculation dose and observed an indistinguishable effect of the quadruple mutation on AvrRpt2-ETI while the overall bacterial numbers at 2 dpi were ten-fold lower ( Figure S1 ) . This observation with a lower dose inoculum indicates that the effect of the quadruple mutation on AvrRpt2-ETI ( Figure 1A and 1B ) did not result from growth saturation of Pto DC3000 EV in the quadruple mutant . Therefore , we conclude that the network defined by the four genes is responsible for 80% of the bacterial growth restriction observed in wild-type plants , which we term AvrRpt2-ETI . ETI triggered by two other bacterial effectors , AvrRpm1 and AvrPphB , which are recognized by the R proteins RPM1 and RPS5 in Col-0 [8] , [9] , was reduced by 20% and 50% in the quadruple mutant , respectively ( Figure 1C and 1D , and Figure S2 ) . These results indicate that the signaling network defined by the four genes is required to variable extents in the three cases of ETI even though all these effectors originated from P . syringae and are recognized by the CC-NB-LRR class of R proteins [27] . In the corresponding R gene mutants , ETI was negligible or slightly negative; the latter is likely due to effector virulence functions [28] ( Figure 1A–1D and Figure S2 ) . Since the quadruple mutant was morphologically normal , developmental differences are unlikely to explain its immune phenotype ( Figure 1E ) . ETI is often associated with a hypersensitive response ( HR ) [29] , a form of rapid plant cell death . All three effectors elicited a strong macroscopic HR in wild-type plants . In the quadruple mutant , AvrRpm1-induced HR was similar to the wild type , but HR induced by AvrRpt2 or AvrPphB was undetectable or weak ( Figure S3A and S3B ) . We quantified the strength of the HR triggered by AvrRpt2 by monitoring electrolyte leakage . The quadruple mutant inoculated with Pto DC3000 AvrRpt2 clearly displayed less ion leakage than wild-type and was indistinguishable from the quadruple mutant inoculated with Pto DC3000 EV ( Figure S3C ) , indicating that the signaling network defined by the four genes controls AvrRpt2-triggered HR . We observed more ion leakage in the quadruple mutant inoculated with Pto DC3000 AvrRpm1 than wild-type ( Figure S3C ) . The signaling network defined by the four genes may negatively regulate AvrRpm1-triggered HR . These results indicate that the mechanisms that trigger HR differ between these two cases of ETI . MAMP-induced resistance against virulent Pto DC3000 is considered to be a good measure of PTI induced by the MAMP [4] , [11] . Flg22 is a MAMP derived from bacterial flagellin [30] . As previously reported [4] , [11] , pretreatment with flg22 induces immunity in plants observable as suppression of bacterial multiplication ( Figure 2A ) . Here we refer to flg22-induced resistance as flg22-PTI . Flg22-PTI was defined as the difference in log10-transformed bacterial number of Pto DC3000 between water-pretreated ( mock ) and flg22-pretreated leaves at two dpi . The bacterial growth suppression by flg22 pretreatment may only represent limited aspects of PTI in natural situations . However , we employed this experimental system because it allows quantification of the specific effect of the well-defined MAMP flg22 in respect to the state of no induced resistance in mock-pretreated plants . Consequently , we were able to define the percentage of the flg22-specific effect on bacterial growth that was abolished in particular plant mutants . Previous analysis of single mutants suggested that the JA and ET sectors have no effect on flg22-PTI , while the SA sector has a modest effect [4] , [11] , [31] . Our results were very similar to these previous observations ( Figure 2A and 2B ) . In contrast , analysis of the quadruple mutant revealed that the signaling network defined by the four genes accounted for ∼80% of flg22-PTI ( Figure 2A and 2B ) . Thus , multiple mutant analysis revealed the fundamental importance of the SA , JA , and ET sectors in flg22-PTI , which was not clear from analysis of single mutants . We also measured elf18-PTI to Pto DC3000 . Elf18 is derived from the bacterial elongation factor Tu [32] . Elf18-PTI is weaker than flg22-PTI in the wild-type ( Figure 2C and 2D ) . In the quadruple mutant ∼50% of elf18-PTI was lost ( Figure 2D ) . However , the level of immunity remaining in the quadruple mutant was very similar in flg22- and elf18-PTI ( Figure 2B and 2D ) . These results indicate that the stronger effect of flg22 than elf18 on PTI is entirely dependent on the signaling network defined by the four genes . For both flg22 and elf18 , no PTI was detected in the corresponding PRR mutants , fls2 and efr , respectively ( Figure 2A–2D ) , indicating that the PTI observed was completely dependent on specific recognition by the PRRs . Flg22-induced MAP kinase ( MPK3/6 ) activation ( Figure 3A ) and expression of early flg22-responsive genes such as a Chitinase ( Figure 3B ) and FRK1 ( Figure 3C ) were comparable between the quadruple mutant and wild type ( Col-0 ) . While these responses may contribute to flg22-PTI , they do not require the signaling network defined by the four genes . Inhibition of seedling growth by flg22 [33] was only slightly reduced in the quadruple mutant ( Figure 3D and 3E ) , indicating that seedling growth inhibition by flg22 is mediated mainly by mechanisms other than the signaling network defined by the four genes . Chitin from fungal cell walls is recognized by a LysM receptor kinase and contributes to induction of immunity to the fungal necrotroph A . brassicicola [34] , [35] . Although the JA sector is important for immunity against this necrotroph [17] , the plant immune mechanisms involved are still largely unclear . The quadruple mutant showed greater susceptibility than dde2 when damaged plant tissue was visualized by trypan blue staining ( Figure 4A ) . The quadruple mutant was also more susceptible than a pad3 mutant , which is known to be highly susceptible , presumably due to lack of the phytoalexin , camalexin [36] ( Figure 4A ) . In the quadruple mutant , camalexin accumulation was comparable to Col-0 ( Figure 4B ) , indicating that the susceptibility in the quadruple mutant is not due to loss of camalexin . To quantify disease severity , we extracted DNA from infected leaves and determined the ratio of the fungal genome copy number to the plant genome copy number using quantitative PCR . We call the log2-transformed ratio value the disease index . We confirmed that disease severity correlates well with the disease index ( Figure 4C ) . The susceptibility in ein2 , pad4 and sid2 single mutants was indistinguishable from Col-0 , suggesting that complex interactions among the sectors are involved in the great susceptibility in the quadruple mutant . Figure 5 illustrates a hypothetical network consisting of three signaling sectors A , B , and C . The output of the network , immunity , is determined by a combination of the effects of each sector A , B , and C and their interactions , A:B , A:C , B:C and A:B:C . If signaling sector A is depleted by mutation a , the output loses not only the contribution from sector A but also those from interactions A:B , A:C and A:B:C . Therefore , if the interactions are significant , the sole effect of sector A cannot be estimated based on comparison of the phenotypes of wild type and mutant a . To determine the sole effect of sector A , a comparison of the phenotypes of an a/b/c triple mutant in which none of the sectors are functional and a b/c double mutant , in which only sector A is functional , is required . To expand this concept to the four-sector ( gene ) situation , we constructed all combinations of double and triple mutants for the four genes , DDE2 , EIN2 , PAD4 , and SID2 . These double and triple mutants were morphologically normal ( data not shown ) . We measured levels of AvrRpt2-ETI in all the quadruple , triple , double and single mutants and the wild-type ( Figure S4 , Table S2 , and Table S3 ) . Quantitative measurement with a large number of replicates allowed us to estimate the effects of the wild-type genes ( not the effects of the mutations ) and of their interactions on immunity ( Figure 6A ) by fitting the following mixed general linear model: −log10 ( Bacterial number ) ∼a1 ( DDE2 ) + a2 ( EIN2 ) + a3 ( PAD4 ) + a4 ( SID2 ) + a5 ( remainder ) + b1 ( DDE2:EIN2 ) + b2 ( DDE2:PAD4 ) + b3 ( DDE2:SID2 ) + b4 ( EIN2:PAD4 ) + b5 ( EIN2:SID2 ) + b6 ( PAD4:SID2 ) + c1 ( DDE2:EIN2:PAD4 ) + c2 ( DDE2:EIN2:SID2 ) + c3 ( DDE2:PAD4:SID2 ) + c4 ( EIN2:PAD4:SID2 ) + d ( DDE2:EIN2:PAD4:SID2 ) + genotype +1|replicate/pot The values shown in Table 1 were assigned to the variables for the single wild-type gene effects and the 2- , 3- , and 4-wild type gene interactions after inoculation with Pto DC3000 AvrRpt2 . The remainder represents the remaining immunity in the quadruple mutant . Interactions among the four genes and this remainder of an unknown origin cannot be determined as the remainder cannot be manipulated by the mutations . Note that not all the values assigned are 0 or 1 but that some are fractions ( hence , it is a general linear model ) . The reason for the fractional values is explained in Text S1 . When a plant was inoculated with Pto DC3000 EV , 0 was assigned to all these variables . In this way , the genotype factor of fixed effects captures the log10-bacterial number differences among the genotypes when the plants were inoculated with Pto DC3000 EV . The random factors , 1|replicate/pot , indicate that for each independent replicated experiment , plants were grown in multiple pots . The negative of the log10-bacterial number was used for the analysis , so obtained positive coefficients indicated positive contributions to immunity . Thus , AvrRpt2-ETI was dissected into effects of single genes a1–4 , two-gene interactions b1–6 , three-gene interactions c1–4 , the four-gene interaction d , and the remainder a5 . We call this procedure signaling allocation analysis . In the wild-type , effects of each gene and the remainder and their interactions contribute to AvrRpt2-ETI ( Figure 7A ) . In sid2 , effects of DDE2 , EIN2 , PAD4 and the remainder and DDE2:EIN2 , DDE2:PAD4 , EIN2:PAD4 and DDE2:EIN2:PAD4 interactions contribute to AvrRpt2-ETI ( Figure 7A ) . In the quadruple mutant , only the remainder effect contributes to AvrRpt2-ETI ( Figure 7A ) . In other genotypes , the contributions of the effects of genes and their interactions can be assigned similarly . Figure 7B illustrates interpretation of an interaction term in a hypothetical network consisting of two sectors A and B . Effect A is the level of immunity in mutant b , and effect B in mutant a . If the sum of effects A and B is lower than the immunity level in WT , the A:B interaction is positive , meaning that there must be a synergistic effect between sectors A and B . If the sum of effects A and B is equal to the immunity level in WT , the A:B interaction is zero , indicating that sectors A and B are independent . If the sum of effects A and B is higher than the immunity level in WT , the A:B interaction is negative , meaning that each of the sectors A and B can compensate for loss of the other . Note that this compensation in the immunity does not necessarily imply mechanistic similarity , i . e . , sectors A and B may regulate completely separate sets of defense components that nevertheless compensate for one another in restricting pathogen growth . The signaling allocation for AvrRpt2-ETI indicates that all single gene effects were positive ( Figure 6A ) . In fact , the level of AvrRpt2-ETI in the quadruple mutant was lower than in any of the triple mutants ( Figure S4A ) , meaning that each single signaling sector , including the JA and ET sectors , positively contributed to AvrRpt2-ETI , as defined by the suppression of bacterial growth . These positive contributions to immunity of the JA and ET sectors appear to contradict previous general conclusions based on analyses of single mutants or responses to exogenously applied hormones [17] . The signaling allocation analysis revealed the positive contributions of the JA and ET sectors by dissecting complex interactions among the sectors . Most of the interaction terms were negative ( Figure 6A ) , indicating that the signaling sectors can compensate each other ( Figure 7B ) . For instance , the PAD4 and SID2 single gene effects and the PAD4:SID2 interaction were 1 . 2 , 0 . 8 and −0 . 8 ( Figure 6A ) , meaning that PAD4 and SID2 positively contribute to AvrRpt2-ETI and that they can compensate each other's function in AvrRpt2-ETI , suggesting that PAD4 has a role as a backup in case of loss of the SA ( SID2 ) sector . Similarly , the two-gene interactions DDE2:EIN2 , DDE2:PAD4 , EIN2:PAD4 and EIN2:SID2 , were all negative , indicating that most of the signaling sectors can compensate each other in case of losses of sectors in AvrRpt2-ETI . This network compensation among positively contributing signaling sectors should make AvrRpt2-ETI highly resistant to genetic and pathogenic perturbations since when one sector is disrupted , other signaling sectors compensate for its loss . The signaling allocation analysis was also applied to AvrRpm1-ETI ( Figure 6A and Figure S5 , Table S4 and Table S5 ) . The overall tendency up to the 2-gene interactions was similar to that in the AvrRpt2-ETI , although the overall contribution of the signaling network defined by the four genes was weaker and the contribution of the remainder variable was higher in the AvrRpm1-ETI than the AvrRpt2-ETI . For signaling allocation analyses , we measured the levels of MAMP-PTI in all combinatorial mutants ( Figure S6 and Figure S7 , Tables S6 , S7 , S8 , S9 ) . The signaling allocation analysis of flg22-PTI revealed that all single signaling sectors contribute positively to the immunity and that some interactions , such as the PAD4:SID2 interaction , were positive ( Figure 6B ) . A positive feedback loop comprised of PAD4 and SID2 that amplifies the SA signal [37] likely explains this synergistic interaction . The three-gene interactions DDE2:PAD4:SID2 , DDE2:EIN2:SID2 and DDE2:PAD4:SID2 and the four-gene interaction were also positive ( Figure 6B ) , indicating that there are synergistic interactions between the SA and JA/ET sectors . There are some differences in the signaling allocation between flg22- and elf18-PTI . For example , most of the EIN2-related interactions in elf18-PTI were negative ( Figure 6B ) and single EIN2 gene effect was much smaller than that in flg22-PTI , which explains the observation that elf18-PTI in ein2 plants is stronger than that in wild-type plants but not in the case of flg22-PTI ( Figure S6B and Figure S7B ) . However , overall the signaling allocation of elf18-PTI is similar to that of flg22-PTI although the overall effect size is smaller with elf18 ( Figure 6B ) . These results indicate that unlike ETI , PTI depends heavily on synergistic interactions among the signaling sectors . For a signaling allocation analysis , we measured the disease indexes for A . brassicicola infection with all the combinatorial mutants ( Figure S8 and Table S10 ) . Since the disease index for no immunity cannot be determined , the remainder term and the genotype factor were eliminated from the signaling allocation model . All the single gene effects were positive although the contribution from DDE2 was clearly the largest ( Figure 6C ) . All gene interactions were negative ( Figure 6C ) , indicating all signaling sectors can compensate each other . The results of the signaling allocation analysis imply that while the JA sector is the primary signaling sector for immunity to A . brassicicola , the ET , SA and PAD4 sectors contribute to robust immunity . For example , although the reduction of immunity in the dde2 single mutant was 2 . 9 disease index units ( Figure 4C ) , it was smaller than the DDE2 single gene effect of 4 . 3 units ( Figure 6C ) , indicating that other sectors partly compensate the JA sector .
We quantified immunity triggered by MAMPs and effectors by relative growth of Pto DC3000 strains within leaves in the Arabidopsis quadruple mutant dde2/ein2/pad4/sid2 , all the combinatorial mutants and wild-type plants . We have termed the measured restriction of bacterial growth due to a MAMP or an effector MAMP-PTI and effector-ETI , respectively . A striking revelation is that the same network defined by the four genes accounted for ∼80% of both flg22-PTI and AvrRpt2-ETI ( Figure 1A and Figure 2A ) , indicating that the signaling machinery is likely highly conserved between PTI and ETI . If the notion that PTI evolved before ETI [1]–[3] is correct , this observation may suggest that while acquisition of a new class of recognition mechanism ( i . e . , R proteins ) was necessary for evolution of ETI , ETI adapted the rest of the defense machinery mostly from the already existing PTI . However , use of the overlapping signaling network in PTI and ETI is very different as interactions among the signaling sectors are very different ( Figure 6A and 6B ) . Note that the general signaling allocation patterns within the network are characteristic for different types of immunity: strong positive interactions of PAD4:SID2 and DDE2:PAD4:SID2 stand out in flg22- and elf18-PTI ( Figure 6B ) ; most gene interactions are negative in AvrRpt2- and AvrRpm1-ETI ( Figure 6A ) . In all cases , the immunity was measured by growth suppression of Pto DC3000-derived strains . These differences between immunity types suggest that while immunity is enhanced by synergistic sector interactions in PTI , robust immunity to perturbations of signaling sectors is achieved by compensation functions among the sectors in ETI . In other words , PTI can be substantially impaired by a pathogen effector that disrupts one of the synergistically interacting sectors , while ETI is robust against such disruption . It will be interesting to investigate whether or not the bacterial effectors that suppress PTI signaling [12]–[15] are able to overcome ETI . While overall signaling allocation patterns were similar between AvrRpt2-ETI and AvrRpm1-ETI , levels of the dependency on the signaling network defined by the four genes were very different , ∼80% and ∼20% , respectively . The difference in the dependency on the signaling network may be explained by a difference in the effectiveness of early immune responses . In flg22-PTI , the signaling network defined by the four genes did not have substantial roles in induction of early responses ( Figure 3A and 3B ) . By extending this observation , we speculate that also in ETI the signaling network defined by the four genes mainly controls late immune responses , In AvrRpm1-ETI , early immune responses may be more effective , and consequently the contribution from late immune responses controlled by the signaling network may be less important . In contrast , early immune responses may not contribute as much to AvrRpt2-ETI , and consequently the contribution from the signaling network may be more important . Consistent with this hypothesis , it is well documented that HR induced by AvrRpm1 is more rapid than that induced by AvrRpt2 [38] . The contribution of early immune responses to AvrPphB-ETI may be somewhere between those in AvrRpm1-ETI and AvrRpt2-ETI , as the level of dependency of AvrPphB-ETI on the network ( ∼50% ) was intermediate . In considering a network to explain plant immunity , it is necessary to include a pathogen ( Figure 8A ) : The input of the network is MAMPs or effectors; this input is fed into the complex signaling network in the plant; multiple outputs from the signaling network result in induction of a battery of plant defense components; the plant defense components attack a variety of targets in the pathogen; and the effects on the targets are combined to influence the growth of the pathogen . The network output is restriction of pathogen growth . In this way , robust immunity is understood as the network in which blockages of a small number of sectors ( i . e . , parts of the network ) on the plant side do not have much effect on the output . When each sector influences many other sectors in a network ( i . e . , it is highly interconnected ) and there is no particular specialization among the sectors , the network is called a democratic network [39] because each sector has a similar level of importance . An extreme analogy of a democratic network is percolation through a mesh of water paths: many sectors need to be blocked before the output of water is significantly reduced . On the other hand , networks comprised of signaling sectors that lack interconnections are called autocratic networks [39] . A network in which specific pathways control specific functions is a typical autocratic network . Autocratic networks can be studied as individual signaling sectors in isolation from the others , and are easily studied using reductionist approaches to biological processes . However , biological networks often have architectures somewhere between the democratic and autocratic . We think that the plant ETI network has a democratic property to some extent: when one or two network sectors are blocked , the signal flows through some other sectors can compensate for the loss of the blocked sector ( s ) . As a result , the level of the output , restriction of pathogen growth , does not change much ( Figure 8B ) . However , there is a clear difference between the ETI network and a perfectly democratic network . The SA and JA/ET sectors are mutually inhibitory in many cases [17] . Such mutually inhibitory regulatory relationships between signaling sectors result in a steady state in which a dominant , primary sector suppresses the activities of the others ( Figure 8A ) . Other sectors get heavily used only when the primary sector is blocked ( Figure 8B ) . Thus , other sectors effectively buffer blockages in the primary sector . As both the SA and JA/ET sectors positively contribute to immunity ( Figure 6A ) , the loss of signaling flow through the SA sector is compensated by rerouting signal through the JA/ET sectors . Such compensation between positively contributing sectors explains the robustness of AvrRpt2-ETI as suggested previously [40] . Disruptions of multiple signaling sectors are required to compromise AvrRpt2-ETI ( Figure 8C ) . Since there are clear differences in importance among signaling sectors , the ETI network is not really a democratic network although it has some democratic properties . General conclusions based on analyses of single mutants and responses induced by exogenous application of hormones are that the SA sector is inhibitory to immunity to necrotrophs and that the JA/ET sectors are inhibitory to immunity to biotrophs [17] . Our observations show that the SA , JA , and ET sectors can contribute positively to immunity to both biotrophs and necrotrophs as single sectors . Mutual inhibition between the SA and JA/ET sectors and differences in the levels of contribution can reconcile the apparent contradiction between the general conclusions and our observations . An important point is that the general conclusions were made without data from plants blocked in all three sectors . In immunity to a biotroph , for example , immunity mediated by the JA/ET sectors is not as strong as that mediated by the primary sector , SA , although the JA/ET sectors can still contribute positively to immunity . When the JA sector is activated by exogenous JA or coronatine from pathogens ( a JA-Ile mimic ) , the JA sector seems to have negative effects on immunity to biotrophs . This is because the backup immunity mediated by the JA sector is not as strong as that mediated by the SA sector , which is inhibited by the strongly activated JA sector . Another important aspect of a highly interconnected network with some democratic properties is that the concepts of upstream , downstream , and parallel signal flows , which can be clearly defined in an autocratic network , become somewhat obscure . Figure 8C and 8D illustrate this situation: two different combinations of triple mutations result in loss of most resistance , but the relationships among these sectors cannot be clearly defined as upstream , downstream , or parallel . In addition to the signaling sectors we explored in this study , other signaling sectors including the MAP kinases [41] and other hormones such as abscisic acid , auxins and gibberellins [42]–[44] , are involved in plant immunity . It is conceivable that different combinations of simultaneous sector disruptions among all these sectors could also lead to loss of most immunity in AvrRpt2-ETI . This view correlates with a property of the extreme analogy of percolation: as long as a sufficient number of sectors are blocked , the water flow becomes nearly zero when the blocked sectors are randomly chosen . It will be necessary to expand this multiple sector disruption approach to a larger network including other signaling sectors in order to elucidate properties of the larger plant immune network . We took advantage of the model plant Arabidopsis in this study since mutants deficient in defense responses were available . Given that the defense-related genes such as DDE2 , EIN2 , PAD4 and SID2 which are used in this study are highly conserved among different plant species ( data not shown ) , immune network properties we observed with Arabidopsis may be conserved in other plant species . We observed similar signaling allocations in the network defined by the four genes in two cases of ETI or two cases of PTI although the network defined by the four genes is used to variable extents in each case of ETI or PTI ( Figure 1A–1D and Figure 2A–2D ) . Therefore the network properties we revealed may be common features in ETI and PTI . Further investigation will be required to determine whether these network properties generally apply to various cases of ETI and PTI , including those involving different species of pathogens and hosts . Network compensation among positively contributing signaling sectors can explain the robustness of plant immunity , particularly AvrRpt2-ETI . We have demonstrated that quantitative analysis of all combinations of multiple mutations that together deplete almost all of the network function provides a powerful approach for elucidation of quantitative relationships among highly interconnected signaling sectors . Similar approaches should benefit analysis of other complex biological networks .
Arabidopsis thaliana accession Col-0 was the background of all mutants used in this study . Arabidopsis atrbohD [45] , dde2-2 [46] , ein2-1 [23] , efr-2 [5] , fls2 ( SAIL_691C4 ) [4] , mpk3 ( SALK_151594 ) [47] , mpk6-2 [48] , npr1-1 [49] , pad3-1 [50] , pad4-1 [25] , pmr4-1 [51] , rpm1-3/rps2 101C [52] , rps5 ( SALK_015294 ) [53] and sid2-2 [24] were previously described . We generated the double ( dde2-2/ein2-1 , dde2-2/pad4-1 , dde2-2/sid2-2 , ein2-1/pad4-1 , ein2-1/sid2-2 and pad4-1/sid2-2 ) , the triple ( dde2-2/ein2-1/pad4-1 , dde2-2/ein2-1/sid2-2 , dde2-2/pad4-1/sid2-2 and ein2-1/pad4-1/sid2-2 ) and the quadruple ( dde2-2/ein2-1/pad4-1/sid2-2 ) mutants by standard genetic crosses , tracking the mutations by PCR product length difference or cleaved amplified polymorphic sequence ( CAPS ) markers . Primers and restriction enzymes used for screening of the mutants are listed in Table S1 . The mutants containing dde2-2 required spraying with MeJA for seed production due to loss of JA biosynthesis [46] . The ein2-1 and pad4-1 mutations create premature Stop codons [23] , [25] . The sid2-2 mutation is a deletion [24] . Therefore , mutations used for constructing the quadruple dde2/ein2/pad4/sid2 mutant are considered null . Arabidopsis plants were grown in a controlled environment at 22°C with a 12 h photoperiod and 75% relative humidity . Bacterial growth assays were performed as described previously [11] . In brief , bacterial suspensions were infiltrated into leaves of 4 to 5 week-old plants using a needleless syringe . Log10-transformed colony-forming units ( cfu ) per cm2 leaf surface area were calculated and the models described in Materials and Methods were fit to the data . Flg22 and elf18 peptides were purchased from EZBiolab Inc ( Westfield , IN , USA ) . Indicated concentrations of flg22 and elf18 solutions were applied to leaves of 4 to 5 week-old plants by infiltration using a needleless syringe one day before infiltration with bacteria . The lme function in the nlme package in the R environment was used . Pto DC3000 was grown overnight at room temperature in King's B medium supplemented with 25 µg/ml of rifampicin . Pto DC3000 strains carrying the empty vector ( pLAFR ) , AvrRpt2 , AvrRpm1 or AvrPphB were grown overnight at room temperature in King's B medium supplemented with 25 µg/ml of rifampicin and 10 µg/ml of tetracycline . The bacteria were harvested by centrifugation , washed , and diluted to the desired density with water . The following models were fit to the log10-transformed bacterial number data: Sgtrp = GTgt+R/Prp+εgtrp ( Figure 1A–1D , Figures S1 , S2 , S4 , S5 ) ; Sgtrfp = GTgt+R/F/Prfp+εgtrfp ( Figure 2A–2D , Figures S6 , S7 ) , where S , log10-transformed bacterial number; GT , genotype:treatment interaction , and random factors: R , replicate; F , flat; P , pot; and ε , residual . The mean estimates of the genotype:treatment interaction were used as the modeled log10-transformed bacterial number . The modeled log10-transformed bacterial number values were compared using two-tailed t-tests . ETI and PTI values were compared using two-tailed t-tests . For the t-tests , the standard errors appropriate for each comparison were calculated using the variance and covariance values obtained from the model fitting . MAP kinase assays were performed as described previously [54] . Arabidopsis seedlings were grown for 11 days on a medium solidified with 0 . 8% agar that contained 0 . 5×MS salts with Gamborg's vitamins ( M0404; Sigma ) and 1% ( w/v ) sucrose and then transferred to 12-well plates ( six seedlings per well ) in which each well contained 3 ml of liquid medium containing 0 . 5×MS salts with Gamborg's vitamins and 1% ( w/v ) sucrose with 1 µM flg22 peptide . After 0 to 40 min as indicated , the seedlings were frozen in liquid nitrogen . The frozen seedlings were ground in liquid nitrogen and homogenized in 100 µl of extraction buffer ( 100 mM HEPES , pH 7 . 5 , 5 mM EDTA , 5 mM EGTA , 2 mM dithiothreitol , 10 mM Na3VO4 , 10 mM NaF , 50 mM ß-glycerolphosphate , 1 mM phenylmethylsulfonyl fluoride , 1 tablet/10 ml extraction buffer of proteinase inhibitor cocktail ( 11 836 153 001; Roche Applied Science , Indianapolis , IN , USA ) and phosphatase inhibitor cocktail ( 04 906 845 001; Roche Applied Science ) , 10% glycerol , 1% ( w/v ) polyvinylpolypyrrolidone ) . After centrifugation at 13 , 000 rpm for 30 min at 4°C , supernatants were frozen and stored at −20°C . The protein concentration was determined using a Bradford assay ( BIO-RAD , Hercules , CA , USA ) with IgG as a standard . Twenty micrograms of protein was separated in an 8% polyacrylamide gel . Immunoblot analysis was performed using anti-phospho-p44/42 MAPK ( 1∶2000 , Cell Signaling Technology , Danvers , MA , USA ) and anti-AtMPK3 ( 1∶2000 , Sigma ) as primary antibody , and peroxidase-conjugated goat anti-rabbit IgG ( 1∶15 , 000 , A 6154; Sigma ) . Water ( mock ) or 1 µM flg22 ( flg22 ) were infiltrated into 4-week-old Col-0 , dde2/ein2/pad4/sid2 or fls2 plants . Six leaves from 3 plants per sample were collected 3 hpi , frozen in liquid nitrogen and stored at −80°C . Total RNA isolation and real-time PCR analysis was carried out as described previously [11] . Two independent experiments ( biological replicates ) were performed . The following model was fit to the Ct value data using the lme function in the nlme package in the R environment: Ctgytr = GYTgyt+Rr+εgytr , where GYT , gene:genotype:treatment interaction , and random factors; R , replicate; ε , residual . The mean estimate of the gene:genotype:treatment interaction was used as the modeled Ct value . The relative log2 expression values were obtained by subtracting the Ct value of the genes from the Ct value of the Actin2 gene and compared for each gene using two-tailed t-tests . For the t-tests , the standard error appropriate for each comparison was calculated using the variance and covariance values obtained from the model fitting . Flg22-induced seedling growth inhibition assays were performed as described previously [55] . Approximately twenty Arabidopsis seedlings per treatment were grown on a medium solidified with 0 . 8% agar that contained 0 . 5×MS salts with Gamborg's vitamins and 1% ( w/v ) sucrose for 5 days and then transferred to 24-well plates ( one seedling per well ) in which each well contained 800 µl of liquid medium containing 0 . 5×MS salts with Gamborg's vitamins and 1% ( w/v ) sucrose with and without 1 µM flg22 peptide . Seedling fresh weight was recorded 10 days later . From this data , the log10-transformed seedling weights were calculated . The following model was fit to the log10-transformed seedling weight data using the lme function in the nlme package in the R environment: Sgtrfp = GTgt+R/Prp+εgtrp , where S , log10-transformed seedling weight; GT , genotype:treatment interaction , and random factors; R , replicate and P , plate , ε , residual . The mean estimate of the genotype:treatment interaction was used as the modeled log10-transformed seedling weight . Flg22-induced seedling growth inhibition values were obtained by subtracting the value of seedling weight in flg22-treated samples from the value of seedling weight in mock-treated samples and compared using two-tailed t-tests . For the t-tests , the standard error appropriate for each comparison was calculated using the variance and covariance values obtained from the model fitting . Alternaria brassicicola strain ATCC96836 was grown on potato dextrose agar ( BD , Franklin Lakes , NJ , USA ) for 10 days at room temperature with a 12 h photoperiod . Subsequently , the spores were washed from the surface of the plate with 0 . 02% Tween-20 , and hyphae were removed from the suspension by filtering through four layers of cotton cheesecloth . Concentration of spores was determined using a hemocytometer and adjusted to 1×105 spores/ml with 0 . 02% Tween-20 . Three-week-old plants were inoculated by placing a 10 µl droplet of spore suspension ( 1×105 spores/ml ) onto the leaf surface . Inoculated plants were kept at 100% RH at 22°C with a 12 h photoperiod for 3 days . DNA was isolated and the relative amount of A . brassicicola ( CutinA . 1 ) DNA to plant ( iASK ) DNA was determined by qPCR as described previously [56] . qPCR analysis was carried out using an ABI7500 Real Time PCR system ( Applied Biosystems , Foster city , CA , USA ) and the SYBR Green JumpStart Taq ReadyMix ( Sigma , Saint Louis , MO , USA ) . Disease index [log2 ( CutinA . 1/iASK ) ] values were obtained by subtracting the Ct value of the CutinA . 1 from the Ct value of the iASK . The following model was fit to the disease index data: Sgytr = GTgt+Rr+εgytr , where S , disease index; GT , genotype:time interaction , and random factors: R , replicate; ε , residual . The mean estimate of the genotype:time interaction was used as the modeled disease index value . The disease index values were compared using two-tailed t-tests . For the t-tests , the standard error appropriate for each comparison was calculated using the variance and covariance values obtained from the model fitting . Inoculated leaves were boiled in 2 ml of a staining solution ( 14 ml of 95% ethanol , 2 ml of water-saturated phenol , 2 ml of Glycerol , 2 ml of lactic acid , 1 ml of water and 0 . 02 g of trypan blue ) and incubated overnight at room temperature . Then samples were destained in destaining solution ( 2 . 5 g/ml of chloral hydrate ) for 2 days . Plants were inoculated by placing one droplet of 10 µl of fungal spore suspension onto the leaf surface . Inoculated plants were kept at 100% RH at 22°C with a 12 h photoperiod for 3 days . Each sample consisted of one leaf . Two independent experiments were conducted with 12 replicates per sample per experiment . Camalexin was extracted from Arabidopsis leaves infected with A . brassicicola and quantified as described previously [57] . From this data , the log10-transformed camalexin amount ( ng per leaf ) was calculated . The following model was fit to the log10-transformed camalexin data using the lme function in the nlme package in the R environment: Sgp = Gg+Rr+εgr , where S , log10-transformed camalexin amount; G , genotype , and random factors; R , replicate; ε , residual . The mean estimate of the genotype was used as the modeled log10-transformed camalexin . The modeled log10-transformed camalexin values were compared using two-tailed t-tests . For the t-tests , the standard error appropriate for each comparison was calculated using the variance and covariance values obtained from the model fitting . We estimated the effect of each wild-type gene and the interactions among the wild-type genes , which we call the defense signaling allocation , by fitting mixed general linear models . For signaling allocation analysis of flg22 or elf18-PTI: −log10 ( Bacterial number ) ∼a1 ( DDE2 ) + a2 ( EIN2 ) + a3 ( PAD4 ) + a4 ( SID2 ) + a5 ( remainder ) + b1 ( DDE2:EIN2 ) + b2 ( DDE2:PAD4 ) + b3 ( DDE2:SID2 ) + b4 ( EIN2:PAD4 ) + b5 ( EIN2:SID2 ) + b6 ( PAD4:SID2 ) + c1 ( DDE2:EIN2:PAD4 ) + c2 ( DDE2:EIN2:SID2 ) + c3 ( DDE2:PAD4:SID2 ) + c4 ( EIN2:PAD4:SID2 ) + d ( DDE2:EIN2:PAD4:SID2 ) + genotype+1|replicate/flat/pot The values shown in Table 1 were assigned to the variables for the single wild-type gene effects and the 2- , 3- , and 4-wild type gene interactions according to the genotype of the plant when the plant was pretreated with a MAMP . The remainder variable represents the remaining immunity in the quadruple mutant . When plants were mock-pretreated , 0 was assigned to all these variables . In this way , the genotype factor captures the log10-bacterial number differences among the genotypes when the plants were mock-pretreated . The random factors 1|replicate/flat/pot indicate that for each independent replicated experiment , plants were grown in multiple flats each of which is divided into multiple pots . The negative of the log10-bacterial number was used for the response , so obtained positive coefficients indicated positive contributions to immunity . Thus , PTI was dissected into effects of single genes a1–4 , two-gene interactions b1–6 , three-gene interactions c1–4 , the four-gene interaction d , and the remaining immunity a5 . For signaling allocation analysis of immunity against A . brassicicola: From the model design for PTI , the remainder term was removed because the null immunity value was unknown and the genotype factor term was removed because mock treatment in each genotype cannot be defined . −log2 ( CutinA . 1/iASK ) ∼a1 ( DDE2 ) + a2 ( EIN2 ) + a3 ( PAD4 ) + a4 ( SID2 ) + ;b1 ( DDE2:EIN2 ) + b2 ( DDE2:PAD4 ) + b3 ( DDE2:SID2 ) + b4 ( EIN2:PAD4 ) + b5 ( EIN2:SID2 ) + b6 ( PAD4:SID2 ) + c1 ( DDE2:EIN2:PAD4 ) + c2 ( DDE2:EIN2:SID2 ) + c3 ( DDE2:PAD4:SID2 ) + c4 ( EIN2:PAD4:SID2 ) + d ( DDE2:EIN2:PAD4:SID2 ) +1|replicate The values shown in Table 1 were assigned to the variables for the single wild-type gene effects and the 2- , 3- , and 4-wild type gene interactions according to the genotype of the plant , except that no remainder variable was in this model . Additional Materials and Methods are provided in Text S2 .
|
Plants sense molecules originating from pathogens and turn on a battery of immune responses . Activation of immune responses is controlled by a complex network of signaling mechanisms . A traditional approach , knocking out one mechanism at a time , has revealed little about major parts of the signaling network involved in two forms of immunity , Effector-Triggered Immunity ( ETI ) and Pattern-Triggered Immunity ( PTI ) . ETI and PTI are triggered by different types of pathogen molecules . By simultaneously knocking out four major signaling mechanisms in the network , we demonstrated that a common network comprised of the four signaling mechanisms accounts for most of ETI and PTI triggered by particular molecules . The common network was also important for another form of immunity . We also precisely measured how much each signaling mechanism contributes to ETI and PTI and studied how the signaling mechanisms work together . We found that signaling mechanisms work together synergistically in PTI , which may amplify the signal , while they back up one another in ETI to make the immune signaling highly resistant to pathogen attack ( pathogens produce molecules that interfere with immune signaling ) . Therefore , different forms of plant immunity share the same signaling mechanisms , but they use the same mechanisms in very different ways .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"plant",
"biology/plant-biotic",
"interactions",
"mathematics/statistics",
"genetics",
"and",
"genomics/plant",
"genetics",
"and",
"gene",
"expression"
] |
2009
|
Network Properties of Robust Immunity in Plants
|
Strongyloidiasis is a much-neglected soil born helminthiasis caused by the nematode Strongyloides stercoralis . Human derived S . stercoralis can be maintained in dogs in the laboratory and this parasite has been reported to also occur in dogs in the wild . Some authors have considered strongyloidiasis a zoonotic disease while others have argued that the two hosts carry host specialized populations of S . stercoralis and that dogs play a minor role , if any , as a reservoir for zoonotic S . stercoralis infections of humans . We isolated S . stercoralis from humans and their dogs in rural villages in northern Cambodia , a region with a high incidence of strongyloidiasis , and compared the worms derived from these two host species using nuclear and mitochondrial DNA sequence polymorphisms . We found that in dogs there exist two populations of S . stercoralis , which are clearly separated from each other genetically based on the nuclear 18S rDNA , the mitochondrial cox1 locus and whole genome sequence . One population , to which the majority of the worms belong , appears to be restricted to dogs . The other population is indistinguishable from the population of S . stercoralis isolated from humans . Consistent with earlier studies , we found multiple sequence variants of the hypervariable region I of the 18 S rDNA in S . stercoralis from humans . However , comparison of mitochondrial sequences and whole genome analysis suggest that these different 18S variants do not represent multiple genetically isolated subpopulations among the worms isolated from humans . We also investigated the mode of reproduction of the free-living generations of laboratory and wild isolates of S . stercoralis . Contrary to earlier literature on S . stercoralis but similar to other species of Strongyloides , we found clear evidence of sexual reproduction . Overall , our results show that dogs carry two populations , possibly different species of Strongyloides . One population appears to be dog specific but the other one is shared with humans . This argues for the strong potential of dogs as reservoirs for zoonotic transmission of S . stercoralis to humans and suggests that in order to reduce the exposure of humans to infective S . stercoralis larvae , dogs should be treated for the infection along with their owners .
Soil-transmitted helminthiasis ( STH ) affects up to one in four individuals in the world , disproportionately impacting impoverished populations with less access to clean water , sanitation , and opportunities for socioeconomic development [1] . Strongyloidiasis is one of the most neglected tropical diseases [2 , 3] . Estimates of the number of people infected with the causative agent Strongyloides stercoralis vary and go up to 370 million worldwide [2 , 4 , 5] . The local prevalence can reach more than 40% in some tropical and subtropical countries [3 , 6] . Factors such as high temperature , high moisture , poor sanitation and sharing premises with domestic animals may contribute to high prevalence of S . stercoralis [3 , 7 , 8] . S . stercoralis is the major causative agent of human strongyloidiasis [9] but there are also reports of people infected with Strongyloides fuelleborni and Strongyloides fuelleborni kellyi , in Africa and in Papua New Guinea [9] . Based on molecular data , S . fuelleborni kellyi should probably be considered an independent species rather than a subspecies of S . fuelleborni , [10] . Although S . stercoralis infection frequently remains asymptomatic , immuno-compromised patients can develop a systemic infection , which may lead to fatal forms of strongyloidiasis . The medical relevance of this parasite has probably been grossly underestimated due to difficulty of diagnosis [4 , 5 , 11] . Also , it should be noted that Strongyloides is not limited to tropical and underdeveloped areas , and the presence of S . stercoralis and fatal cases caused by it have also been reported from well-developed regions with temperate climates such as the European Union and North America [12–18] . S . stercoralis has a complex , rather unique life cycle ( Fig 1 ) consisting of parasitic and free-living generations [19–21] . In brief: infective third stage larvae ( L3i ) , which are all females , invade a new host by skin penetration and , after migrating through the blood and the lungs , are coughed up and swallowed and eventually establish in the small intestine of the host . The parasitic adult females reproduce by parthenogenesis . The progeny of the parasitic females have four developmental options: 1 ) Firstly , they may become female , and develop into infective third stage larvae ( iL3 ) within the host and re-infect the same host individual ( autoinfective cycle ) ; 2 ) Secondly , they may become female , but this time leave the host as first-stage larvae , develop into iL3 and search for a new host ( direct/homogonic development ) ; 3 ) Thirdly , they may become female and leave the host , but this time develop into free-living , non-infective third stage larvae and subsequently into adult females ( indirect/heterogonic development ) ; 4 ) Or fourthly , they become male and leave the host and develop into free-living adult males ( indirect/heterogonic cycle ) . The free-living adults mate and reproduce in the environment and all their progeny are females and develop to iL3s . No male iL3s have been reported in any Strongyloides species . For two species of Strongyloides ( Strongyloides ratti and Strongyloides papillosus ) , it has been shown that the reproduction in the free-living generation is sexual , in spite of some earlier literature that had described it as pseudogamic ( by sperm dependent parthenogenesis ) [22–24] . For S . stercoralis prior to this report no genetic analysis of the mode of reproduction had been conducted and non-sexual ( pseudogamic ) reproduction as proposed based on cytological observations remained an option [25 , 26] . Whilst all species of Strongyloides may undergo homogonic or heterogonic development , the autoinfective cycle ( option 1 ) appears to be specific for S . stercoralis and maybe a few other less well-investigated species [19] . This autoinfective cycle allows the parasite to persist in a particular host individual for many years , much longer than the life expectancy of an individual worm . Usually , healthy individuals tolerate such long lasting infections well and control them at very low levels [5] . These people have no clinical symptoms and the infection is unlikely to be detected . However , if such a chronically infected person becomes immunodeficient due to disease or immunosuppressive treatment ( i . e . cancer chemotherapy or organ transplantation ) , this may lead to failure to control the infection and consequentially to self-enhancing progression of strongyloidiasis ( hyperinfection syndrome and disseminated strongyloidiasis ) , which is lethal if not treated [5] . Parts of the 18S rDNA ( small Subunit , SSU ) sequence , in particular the hypervariable regions ( HVR ) I and IV , are widely used as nuclear markers for molecular taxonomy of nematodes in general ( e . g . [27–33] ) and Strongyloides spp . in particular [8 , 34–36] . Whilst some sequence variation in HVR I within S . stercoralis was reported [35 , 37] , HVR IV appears virtually invariable within this species . To our knowledge , there is only one report to date of a single nucleotide difference within this region [38] ( accession number M84229 ) . Whilst humans are their natural hosts , dogs , cats , and non-human primates have also been proposed to be suitable hosts for S . stercoralis [39–42] . To what extent strongyloidiasis is a zoonotic disease has been the subject of controversy in the literature for several decades . Originally Brumpt ( 1922 ) [43] , later supported by Augustine ( 1940 ) [44] split the Strongyloides of dogs from S . stercoralis and described it as a separate species , called Strongyloides canis . Recent comparative analyses of the mitochondrial locus cytochrome c oxidase subunit 1 ( cox1 ) [6 , 36 , 38 , 45] and the whole genome sequence of 33 individual S . stercoralis from Japan and Myanmar [46] indicated that there is substantial genetic diversity among S . stercoralis isolated from human hosts and [38] suggested that there might exist human- and dog-specialized subpopulations . On the other hand , dogs have long been known to be suitable experimental hosts for human derived S . stercoralis [39 , 41] , and many authors consider Strongyloides in dogs and humans to belong to the same species , i . e . S . stercoralis . While the more recent literature appears to favor separation , it remains unclear whether S . stercoralis naturally infecting dogs and humans belong to the same populations or not , and correspondingly , what the potential is for dogs to serve as a source for human S . stercoralis infections ( recently reviewed by Thamsborg and colleagues [42] ) . In order to address this question , we compared individual Strongyloides isolated at the same time and location from humans and dogs , which , to our knowledge , had never been done . In our study area , rural communities in Northern Cambodia , people share their premises closely with their dogs , and the prevalence of strongyloidiasis is high [3] . We compared the sequences of the nuclear SSU HVR I and HVR IV and , for a selected subset of worms , the mitochondrial cox1 gene and whole genome sequences . Further , we characterized reproduction in the free-living generations of wild and laboratory isolates of S . stercoralis . Our results show that dogs carry S . stercoralis genetically indistinguishable from the ones in humans in addition to a dog specific population . Further , we demonstrate that reproduction in the free-living generation of both wild and laboratory isolates of S . stercoralis is sexual and not pseudogamic . Overall , our observations strongly support the hypothesis that dogs are a potential source for human S . stercoralis infection and suggest that in order to reduce the exposure of people to infective Strongyloides larvae , dogs should be treated along with their owners in settings where people are exposed to dog excrement .
Fecal samples were collected from humans and dogs of the same households in the villages Anlong Svay ( AS ) and Chom Long ( CL ) in May 2013 and in Damnak Chin ( DC ) and Kampot ( KP ) in June 2016 . All villages are in the Rovieng District ( 13°21′N 105°07′E ) in Preah Vihear province in Northern Cambodia . Human stool samples were collected and S . stercoralis larvae were isolated according to established methods [47] . In brief , stool samples were collected for two consecutive days from each member of the household who agreed to participate in this study . In May 2013 all the fecal samples collected from humans were analyzed within 3 hours after collection using Baermann and Kato-Katz methods . The sediments of positive Baermann funnels were preserved and transported to our laboratory in Tübingen in 70% ethanol at ambient temperature . In June 2016 the fecal samples were mixed with an approximately equal volume of sawdust , moisturized and cultured at ambient temperature for 24–48 hours and analyzed using the Baermann method . From positive Baermann funnels a portion of the worms were transferred individually into 10 μl of water or , for those intended for whole genome sequencing , 10 μl of Tissue and Cell Lysis Solution ( component of the MasterPure DNA Purification Kit , Epicenter MC85201 ) and the remaining worms were preserved as batches in 70% ethanol . While the work was ongoing the samples were stored in the hotel freezer . For transport to our laboratory the samples were refrigerated using wet ice but not frozen . In the majority of cases , worms from the 2016 sample come from those that had been picked individually whilst alive into water or Tissue and Cell Lysis Solution . If any ethanol preserved specimen from 2016 was used , this is explicitly stated . Fecal samples were also collected from dogs found in the proximity of S . stercoralis positive households . The samples were taken directly from the rectum of the animals with the help of the owners and the field assistants . The samples were further processed like the human samples except that for some samples , 3 g of feces were placed on NGM agar plates [48] and incubated for 24–48 hours at ambient temperature and emerging S . stercoralis were picked directly from the plates instead of setting up saw dust cultures followed by baermanization . For ethanol fixed samples , single worms were picked and washed twice with Phosphate-buffered saline ( PBS ) and then incubated in 20 μl 1X lysis buffer ( 20 mM Tris-HCl pH 8 . 3 , 100 mM KCl , 5 mM MgCl2 , 0 . 9% NP-40 , 0 . 9% Tween 20 , 0 . 02% Gelatine , 240 μg/ml Proteinase K ) at 65°C for 2h , followed by incubation at 95°C for 15 min . 2 μl ( for SSU ) or 4 μl ( or single copy loci ) of this lysate were used as template for PCR amplification . For worms stored in 10 μl water , 10 μl 2x lysis buffer were added , after which the samples were treated as described above . For samples preserved in Tissue and Cell Lysis Solution , single worm DNA was prepared using the MasterPure DNA Purification Kit ( Epicenter MC85201 ) according to the manufacturer’s protocol , and the DNA was stored frozen in 10 μl of TE buffer . 1 μl was used for SSU amplification , and the reminder for sequencing library construction . PCR reactions were done in a total volume of 25 μl ( up to 25 μl nuclease-free water , 2 . 5 μl 10X ThermoPol Reaction buffer ( New England BioLabs ) , 0 . 5 μl dNTP’s ( 2mM each ) , 0 . 5 μl 10 mM forward primer ( Table 1 , Fig 2 ) , 0 . 5 μl 10 mM reverse primer ( Table 1 , Fig 2 ) , 0 . 3 μl Taq DNA polymerase ( New England BioLabs ) , 1 μl to 4 μl template as specified above ) . Thermocycling program: 94 °C for 2 min , followed by 35 cycles of denaturing ( 94 °C for 30 sec ) , annealing ( temperature given in Table 1 for 15 sec ) , extension ( 72 °C for time given in Table 1 ) , and a post amplification final extension ( 72°C for 10 min ) and cooling to 4 °C . 1 μl of the PCR reaction and either one of the PCR primers or , in the case of SSU HVR I , a designated sequencing primer were used in sequencing reactions using the BigDye Terminator v3 . 1 Cycle Sequencing Kit ( Applied Biosystems , ) according to the manufacture’s protocol . The reactions were submitted to the in-house sequencing facility at the Max Planck Institute for Developmental Biology at Tübingen for electrophoresis and base calling . Sequences were analyzed with SeqMan Pro version 12 ( Lasergene package; DNAStar , Inc . , Madison , WI USA ) . Chromatograms were visually inspected to detect ambiguous signals indicating mixed sequences ( heterozygous worms ) . For comparison with published sequences , we used BLAST against the NCBI nucleotide database ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) . For the SSU , we used the GenBank entry AF279916 as reference sequence . All position numbers refer to this entry . For S . stercoralis cox1 , we used the sequence LC050212 as reference . 552 base pairs , of which 63 were polymorphic , were considered ( for the full sequences of the different haplotypes see S1 Text ) . For ytPxxx markers , position 1 is the first base of the ( non-nested ) forward primer ( for the sequences of the markers see S1 Text ) . Phylogenetic analysis of the cox1 sequences was done using MEGA7 [49] with default settings . As an outgroup species we used Necator americanus ( AJ417719 ) . The Maximum Likelihood tree is shown in Fig 3 . For comparison we also reconstructed Maximum Parsimony and Neighbor Joining trees , which resulted in the same tree topology as far as well-supported nodes are concerned . DNA of 23 single free-living males S . stercoralis from different hosts ( 11 males from 9 different humans and 12 males from 10 different dogs ) was prepared as described above . Whole-genome sequencing libraries were prepared with Clontech Low Input Library Prep Kits ( Takara Bio , USA ) following the manufacturer's protocol . Samples were submitted to in-house sequencing on an Illumina HiSeq 3000 instrument ( 150 bp paired-end ) . An approach similar to a previous S . stercoralis population study [46] was used to analyze the whole genome data . In brief , raw reads were trimmed with skewer [50] ( version 0 . 1 . 123; -q 30 -Q30 -l 60 ) . Trimmed reads were mapped to the S . stercoralis reference genome ( GCA_000947215 . 1 ) and the small subunit ribosomal RNA ( SSU rRNA; AF279916 ) using bwa mem [51] ( version 0 . 7 . 12; MEM algorithm with defaults ) . Duplicate reads were marked with Picard tools ( http://broadinstitute . github . io/picard ) ( version 2 . 2 . 1; MarkDuplicates with defaults ) . Variants were called with GATK [52] ( build 2016-09-27-g026f7e8; HaplotypeCaller and GenotypeGVCFs both with defaults ) after Indel realignment ( GATK IndelRealigner with defaults ) and freebayes [53] ( version 1 . 0 . 2-6-g3ce827d , defaults ) . The intersection of both variant calls was imported into GNU R using SNPRelate [54] ( version 1 . 8 . 0; method = "biallelic . only" ) . Single-nucleotide polymorphism ( SNPs ) were recursively removed using a sliding window approach and a Linkage Disequilibrium ( LD ) threshold of 0 . 05 and or a minimum allele frequency ( MAF ) of 0 . 05 . The resulting SNP set was used to calculate the fraction of identity by state ( IBS ) for each pair of samples . Results were stored in a genomic identity-by-state relationship matrix and used to estimate a phylogeny with the BIONJ algorithm implemented in ape [55] ( version 4 . 1; defaults ) . A second phylogeny was estimated with a reference-free approach . Trimmed short reads were used to generate a k-mer count graph with khmer ( version 0 . 2 . 0; load-into-counting . py with a k-mer size of 19 otherwise defaults ) . k-mer counts were used to calculate pairwise distances between samples using kWIP [56] . The resulting distance matrix was imported into GNU R to estimate a phylogeny; again using the BIONJ algorithm . Free-living females were isolated after two days of culture as described above and placed individually onto NGM plates seeded with E . coli OP50 [48] . The plates were inspected daily . Females that had produced progeny but presumably had ceased reproduction because they no longer contained embryos in their uterus were picked and prepared for genotyping as described above . One day later , larvae from these females were also isolated and processed . For 17 of these families the mothers and the progeny were picked individually and alive into 10 μl of water and processed as described above . Five more families ( Females 7 , 8 , 14 , 15 , 16 ) were preserved in Ethanol ( one tube per family ) and the individuals were only separated when they were genotyped . The mothers and the progeny were genotyped at ytP289 and ytP290 as described above . ytP289 contains single nucleotide polymorphisms at positions 308 ( A/G ) , 359 ( A/G ) , 416 ( C/T ) and 566 ( C/T ) . Three different combinations ( alleles ) existed in our samples . Allele 1 has the combination G+A+C+C , allele 2 A+G+T+T , and allele 3 A+A+C+C . ytP290 contains single nucleotide polymorphisms at positions 291 ( A/C ) and 310 ( A/G ) . Two different combinations ( alleles ) existed in our samples . Allele 1 has the combination A+A and allele 2 C+G . For the full sequences see S1 Text . The UPD strain and PV001 line of S . stercoralis were maintained in immuno-suppressed dogs and cultured in fecal cultures with charcoal as previously described [41] . Free-living L4 larvae and young adults ( males and virgin females ) were isolated from 1-day fecal cultures at 22 °C using Baermann funnels as described [41] . Single virgin females and males were handpicked and transferred in male-female pairs onto NGM plates spotted with 30 μl of OP50 [48] and 20 μl of water from a Baermann funnel and incubated at 22 oC for 24 hrs . On the next day , from pairs where the female contained developing embryos in the uterus , the males were transferred into PCR tubes containing 10 μl of lysis buffer ( see above ) and frozen for later use . Once the females contained no embryos any longer ( after three days ) they were processed like the males . After all the eggs had hatched , all the L1/L2 were transferred individually into PCR tubes as described for the parents . Single worm lysis was performed as described above for ethanol fixed specimens without the PBS washing step . The parents and eight progeny per cross were genotyped at the marker ytP274 as described above . ytP274 has a single SNP ( T/C ) at position 236 . For the full sequence of ytP274 see S1 Text . The sampling of material in Cambodia was approved by the National Ethics Committee for Health Research ( NECHR ) , Ministry of Health , Cambodia and the ethics committee of the cantons of Basel-Stadt and Basel-Land ( EKBB ) , Switzerland . All participants were informed of the study procedures and provided written informed consent prior to enrolment . All data handled were strictly confidential . All individuals infected with S . stercoralis were treated with Ivermectin ( single oral dose of 200 μg/kg BW ) . Co-infections with other intestinal helminths were treated according the Cambodian treatment guidelines . The experiments requiring culture of S . stercoralis in host animals were all done at the University of Pennsylvania with the approval of the University of Pennsylvania Institutional Animal Care and Use Committee ( IACUC ) . The S . stercoralis UPD strain and PV001 isolate were maintained in prednisolone-treated dogs according to IACUC protocols 702342 , 801905 , and 802593 . All IACUC protocols , as well as routine husbandry care of the animals , were conducted in strict accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . No human subjects were used in this part of the study . The sequences obtained from this study are available from GenBank ( accession numbers KU724124-KU724129 and KX226367-KX226384 and KY548505 ) . The whole genome data are available permanently from the FTP server of the Max Planck Institute for Developmental Biology ( ftp://ftp . tuebingen . mpg . de/pub/PLOS_NTD_Jaleta_2017_whole_genome_data ) . They have also been submitted to the European Nucleotide Archive ( accession number PRJEB20999 ) .
In May 2013 , we collected stool samples from a total of 537 persons from 128 households . Of these , 177 individuals ( 32 . 96% ) living in 95 different households were positive for S . stercoralis . From positive households we obtained rectal fecal samples from a total of 88 dogs . Of these , 78 ( 88 . 63% ) originating from 44 different households were positive . In June 2016 , 20 of 169 ( 11 . 8% ) people from 17 out of 62 households tested positive for S . stercoralis . Of the 29 dogs from 14 households tested , 22 ( 75 . 9% ) living in 12 different households were positive . It should be noted that the prevalence in the two hosts cannot be compared directly because of the biased sampling of dogs . Firstly , we sequenced the region around the HVR I of the SSU from individual S . stercoralis isolated from humans and dogs [10 , 34–36 , 27–33] . Overall , we found the five different haplotypes described in Fig 2 . The two most dissimilar haplotypes ( III and IV ) show four differences ( 2 indels and 2 SNPs ) . Earlier studies [27–30 , 35 , 36] found this fragment to be largely invariable within species and two or more differences to be fairly reliable indicators of different species . Therefore , a within species variability as we observed was rather unexpected but not entirely implausible . In order to obtain additional information , we next sequenced the SSU HVR IV[35 , 36] . Among all Strongyloides individuals from dogs , we found two sequence variants in the SSU HVR IV as defined in [35] . The two haplotypes differed at three positions ( two indels , one base substitution , Fig 2 ) . One of these variants , from now on referred to as HVR IV haplotype A , is the one previously described as the HVR IV sequence of S . stercoralis [35] , and was found ( Table 2 ) in 11 ( 11 . 5% ) out of 96 dog derived worms in the 2013 sample and 39 ( 31% ) out of 126 dog derived worms in the 2016 sample ( in total 50 ( 22 . 5% ) out of 222 ) . This variant was also found in all human derived worms of both samples ( in total 521 worms from a total of 85 host individuals , S1 Table ) . The other variant , referred to hereafter as HVR IV haplotype B , has , to our knowledge , not been described previously and was present in 85/96 ( 88 . 5% ) dog derived worms in the 2013 sample , and in 87/126 ( 69% ) dog derived worms in the 2016 sample . Notably , this haplotype was not found in any of the 521 human derived worms . If the two haplotypes indeed indicate two reproductively isolated groups , which have been separated for some time in evolution , this should also be reflected in their mitochondrial and nuclear genomes . Therefore , from the 2013 sample we sequenced a portion of the mitochondrial cox1 gene of 21 dog-derived worms ( 17 with SSU HVR IV haplotype B and 4 with SSU HVR IV haplotype A ) and 57 human derived worms ( all with SSU HVR IV haplotype A ) [6 , 36 , 38] . In total we identified 17 different cox1 haplotypes . Seven were associated with SSU HVR IV haplotype A and 10 with SSU HVR IV haplotype B ( Fig 3 , S1 Table ) . No haplotypes were shared between worms with the two SSU HVR IV haplotypes . In a phylogenetic analysis , the cox1 haplotypes associated with the two HVR IV haplotypes were well separated from each other . From the 2016 sample we selected 23 worms for whole genome sequencing and compared them over their entire genome using two different approaches ( Fig 4 ) . Both methods robustly separated the worms with SSU HVR IV haplotype B form all worms with SSU HVR IV haplotype A . The two methods yielded different results on the exact topology of the tree within the two SSU HVR IV defined groups but provided no indication that the dog derived worms with SSU HVR IV haplotype A form a group different from the human derived worms . From this we conclude that dogs in our study area carry two populations of S . stercoralis , which are distinguishable by their SSU HVR IV haplotype . One population , to which the majority of the Strongyloides belong , was not found humans . The other population , however , is shared with humans , strongly indicating that S . stercoralis with SSU HVR IV haplotype A can shuttle between the two vertebrate host species . When we analyzed the region around HVR I of S . stercoralis with HVR IV haplotype A , we found the same three variants ( haplotypes I , II , III c . f . Fig 2 ) as described by Schär , Guo and colleagues [37] at various frequencies ( Table 3 ) . In combination with HVR IV haplotype B , which occurs only in dogs , we also found three variants of HVR I ( haplotypes I , IV , V c . f . Fig 2 ) . It is noteworthy that only HVR I haplotype I occurred in combination with both HVR IV haplotypes ( c . f . Fig 2 ) [37] . If the different HVR I haplotypes reflect at least partially separated populations this should be reflected in the whole nuclear genome and the separation should also be visible in the mitochondrial genome . However , the mitochondrial cox1 phylogeny is not correlated with the HVR I differences ( Fig 3 ) , indicating that there is mixing of the nuclear genome between the different mitochondrial matrilinages . In addition , data from whole genome sequencing ( Fig 4 ) are not consistent with the null hypothesis that worms with the same HVR I haplotype are more closely related to each other than to individuals with a different HVR I haplotype . The two methods lead to different tree topologies , none of which correlates with the HVR I haplotypes . Therefore this analysis provides no evidence for the existence of genetically isolated subpopulations within the worms with HVR IV haplotype A . The same is true for the worms with HVR IV haplotype B and different HVR I haplotypes . These results show that among the worms analyzed , other than SSU HVR IV haplotypes , not every SSU HVR I haplotype appears to represent its own separate genetically isolated population . As a possible explanation for the absence of hybrids , Schär , Guo and colleagues [37] proposed that S . stercoralis might not reproduce sexually , either because the free-living generation undergoes pseudogamy ( sperm dependent parthenogenesis ) as proposed earlier [25] , or because in the study area the population propagates only through the non-sexual homogonic cycle . We showed above that the nuclear genome and the mitochondrial genomes , which are normally maternally inherited , do not evolve in parallel . This argues strongly for at least occasional sexual reproduction . Large numbers of free-living S . stercoralis of both sexes were present in our field and laboratory isolates and males were clearly required for reproduction in the free-living generation . Of a total of 480 females ( 96 from field isolates and 384 from a laboratory isolate ) placed individually on plates before they had a chance to mate , none produced progeny . This result indicates that female autonomous reproduction by parthenogenesis or self-fertilization occurs rarely or not at all . However , the result is not an argument against pseudogamy , because , although they do not contribute genetic material to the progeny , males and their sperm are required to activate embryogenesis in this mode of reproduction . Further , although mitochondria are normally inherited only from the mother , in other nematodes male derived mitochondria occasionally fail to be degraded and are incorporated in zygotes [58 , 59] . This could also happen upon pseudogamic interaction between oocyte and sperm and lead to a recombination of nuclear and mitochondrial genomes . Therefore , we sought to demonstrate sexual reproduction more directly . To this end we genotyped 22 individual mothers ( all derived from humans ) and several of their progeny at two single-copy loci ( Table 4 ) . These results demonstrate clearly that the progeny were not the product of clonal reproduction because we found larvae with genetic material absent from the mother but presumably derived from the male and/or larvae that did not have alleles present in the mother , indicating that the mother passed on only half of its genetic material as expected for sexual but not for clonal reproduction . We sought to further solidify this in a fully controlled experiment with known males and a larger number of progeny . To this end , we set up crosses with single females and males using a laboratory isolate of S . stercoralis . The results ( Table 5 ) were fully consistent with Mendelian inheritance with equal genetic contribution by males and females but not with clonal reproduction . Given the results above , one would expect to observe animals that are hybrids between the different SSU HVR I haplotypes . However , like in an earlier study [37] , we failed to find any hybrids among the 436 larvae from 97 different host individuals ( 68 humans 29 dogs ) isolated in 2013 . The SSU locus is on the X chromosome in S . stercoralis ( information extracted from a previously published dataset [57] ) . Therefore , hybrid males could not be detected because they only contain one X chromosome . However , for [37] and in our 2013 sampling , young larvae that were the progeny of parasitic worms were analyzed . The sex of these larvae was unknown but given that the field isolates produced both sexes , it is very likely that a substantial number of females were among them . In 2016 we tested adult worms and used predominantly males because , unlike females , they may not contain genetic material from other individuals ( sperm , embryos ) . Only 30 of the 307 worms from 2016 listed in Table 2 were pre-reproductive females . None of them was a hybrid . However , of the gravid females described in Table 4 one ( Female 17 ) and one of its progeny showed a mixture of HVR I haplotypes II and III . While the mixed signal in the female might have been caused by sperm and embryos derived from a male with a different haplotype than the mother , the larva was most likely a true hybrid . This prompted us to test fully mature females that had been isolated from cultures , after they had a chance to mate , preserved in ethanol and separated by host individuals . We analyzed 9 to 23 worms from each of six host individuals ( four humans , two dogs ) , from which we had already isolated S . stercoralis of different HVR I haplotypes . In five of the cases we found worms with mixed haplotypes , along with individuals with only one SSU haplotype ( Table 6 ) . Two observations are noteworthy . Firstly , both dogs had worms with HVR IV haplotypes A and B . In total , 11 out of 21 worms had mixed HVR I haplotypes but only a single one showed a mixture between the two HVR IV haplotypes , which might indicate that males and females with different HVR IV haplotypes tend to avoid each other . Secondly , from host individual KP57 ( human ) , 9 of 14 worms had mixed haplotypes . From the same host individual we had also genotyped 12 pre-reproductive females , none of which was a hybrid . This finding suggests that the females genotyped in this experiment were not true hybrids but contained sperm and developing embryos derived from males of the other haplotype detected .
In rural communities in Cambodia , many people share their premises with domestic animals and the general hygenic , water and sanitation infrastructures are precarious [60 , 61] . Therefore , the conditions appear very favorable for human to animal and animal to human transmission of STH including S . stercoralis [3] . In order to find evidence for or against zoonotic transmission of S . stercoralis under such circumstances , we isolated large numbers of S . stercoralis from humans and dogs at the same time and in the same households and analyzed individual worms using molecular genetic markers . To our knowledge there had been no such study of S . stercoralis of comparable scale undertaken anywhere . It should be noted that our experimental strategy aimed to sample individuals with a large potential for transfer of Strongyloides spp . between the two hosts ( e . g . , only dogs found close to households with positive people were sampled ) . Therefore our study was not designed to yield accurate estimates of haplotype frequencies in the entire population . We also point out that we did not directly demonstrate transmission from dogs to humans and therefore cannot exclude that the transmission is mostly or exclusively from human to dog . Nevertheless , our results strongly suggest that there is a considerable risk for dog to human transmission . This would not be in agreement with conclusions by Takano and colleagues [62] who found that humans in households with Strongyloides-infected dogs were not more likely to be parasitized by S . stercoralis than those with parasite free dogs and concluded that natural transmission does not occur between humans and dogs . However , this study was conducted in Japan , in areas with presumably much better sanitary conditions than in the Cambodian villages where the present study was conducted . Consequently , only five Strongyloides positive dogs were found and none of their owners was infected . Likewise , a study conducted in Southern China , in a setting probably more comparable to our study area [47] , did not identify the presence of animals as a statistically significant risk factor for human strongyloidiasis . However , this conclusion was based on only 21 infected individuals ( 11 . 7% of the tested ) , and no details about the exposure to dogs are given . In rural settings dogs are usually semi-domesticated and roam freely such that the risk of exposure to contamination by canine feces among people who do not own a dog themselves might be approximately equal to the risk among people who do . Therefore , the lack of statistical significance cannot be taken as evidence against zoonotic transmission . Interestingly , a later study in a similar setting [63] revealed that anthelmintic treatment of people alone was not sufficient to significantly reduce the prevalence of S . stercoralis . Overall , the present findings strongly suggest that dogs must be seriously considered as sources for human strongyloidiasis . Whether dogs are the only non-human carrier of concern or if other animals also have the same potential remains to be determined . However , our results also show that at least in our study area , the majority of the Strongyloides present in dogs is of a genotype that we never found humans . Therefore , the number of Strongyloides spp . detected in dogs by coproscopic diagnosis might be an inaccurate index of the risk of exposure to S . stercoralis that dogs pose to humans . The 18S rDNA HVR IV appears to be diagnostic for the two separate Strongyloides populations in dogs . This agrees with the findings of Hasegawa and colleagues [35] who found this region to be invariable within S . stercoralis . Both , the mitochondrial cox1 and whole genome sequence analyses confirmed that the two HVR IV variants represent separate phylogenetic groups . Therefore , our results support the proposal by Brumpt and Augustine [43 , 44] of a separate species , Strongyloides canis . Consistent with [37] we found three different 18S rDNA HVR-I genotypes in human derived S . stercoralis . Worms of different haplotypes sometimes co-existed in the same host individual . In comparison with other nematodes [27–33] it is unusual that such differences in this region of the SSU occur within one species . However , species status can never be inferred from sequence information alone . Nevertheless , in those examples in nematodes where the same fragment around the HVR I was used and more rigorous criteria for species separation ( e . g . mating experiments ) could be applied , a sequence difference of more than one position was a safe indicator of a distinct species [28–33] . Most of the time , this region appears completely invariable within a species and there are several examples where even separate species do not differ in their HVR I . However , our comparative analyses of the mitochondrial cox1 locus and of whole genome data suggest that in S . stercoralis , different HVR I genotypes do not indicate separate species , but rather that in S . stercoralis the HVR I is more variable than in other nematodes . It should be noted that all the well-studied cases mentioned above involve obligate sexual nematode species , some of which are capable of self-fertilization . Asexual reproduction through the homogonic cycle , which may be a frequent mode of reproduction in S . stercoralis , might contribute to a higher variability within the species . Nevertheless , although we show that the different HVR I genotypes are not diagnostic for different species , our results do not exclude the existence of cryptic species among S . stercoralis . It is striking that , like Schar et al . [37] , we failed to detect hybrids between different HVR I haplotypes among all worms that were the progeny of parasitic mothers and that were definitely unmated . However , in mature free-living females and in their progeny , we frequently found mixed signals . While , for reasons described above , we doubt that these females were true hybrids , their offspring presumably were . We think that this indicates that adults of different HVR I haplotype do mate , at least in laboratory fecal cultures , and that at least some of the progeny develop to larval stages . We can only speculate about why such hybrids are not found in the progeny of the parasitic generation . It might be that the hybrids , or even the progeny of the free-living generation in general , are sub-viable and only rarely develop into successful fertile parasitic females . In this case genetic mixing between subpopulations would occur only rarely . Over long periods of time , even rare exchange might be significant and cause enough mixing of the genomes that we were not able to detect genetic differentiation between subpopulations . Alternatively , in the S . stercoralis populations in our study area there might be very high inbreeding ( brother sister mating ) under natural conditions . This would lead to a very high degree of homozygosity in the population . Both scenarios described above would lead to a very low number of SSU haplotype heterozygotes in the population , and we may have simply missed the rare hybrids . In order to address this , controlled crosses and experimental infections are required . In the context of this study we had no opportunity to do such experiments because , for logistic and for legal reasons , we were not able to bring live worms into our laboratory . Our results demonstrate , however , that such crossing experiments are possible . We demonstrate that in contrast to earlier claims [25] , sexual reproduction in the free-living generation of S . stercoralis does occur and is likely the predominant , if not the only the mode of reproduction in this generation . With this , the number of species of Strongyloides for which genetic analyses demonstrated that the reproduction in the free-living generation is sexual rises to four out of four tested ( S . ratti [23 , 24] , S . papillosus [22] , S . vituli [64] , S . stecoralis , this study ) . Sexual reproduction by S . stercoralis has also been confirmed recently by experimental crossing of free-living male and female worms harboring discrete reporter transgenes[65] . Although these findings do not exclude that asexual species or strains might exist , even within what is currently referred to as S . stercoralis , they do suggest strongly that sexual reproduction rather than pseudogamy is the prevailing mode of reproduction in free-living Strongyloides spp .
Our results provide a compelling solution for the long-standing controversy about whether the Strongyloides sp . of dogs is identical to the S . stercoralis of humans or not . In fact , both scenarios appear to be true . Dogs , at least in our study area , host two different populations . These either represent separate species or well-separated sub-species of Strongyloides spp , and only one of them is shared with humans . It remains to be determined if the different types of Strongyloides we observed in humans and dogs also occur in other regions of the world . Among S . stercoralis in humans there is variability in the rDNA sequence . While we did not find further genomic evidence supporting multiple genetically separate populations in humans the absence of hybrids between the different SSU HVR I haplotypes is striking . It will be most interesting to ascertain whether different SSU HVR I types indeed interbreed and , even more importantly , if they might be associated with different clinical outcomes . Therefore , we suggest using molecular diagnostics for Strongyloides spp . wherever possible . In order to generate comparable data , we propose following the lead of Hasegawa and colleagues [35 , 38] , and using the SSU HVRs I and IV and the mitochondrial cox1 locus as primary markers as was done in this study . With respect to strongyloidiasis control and prevention , this study suggests that dogs should be seriously considered as a source for human S . stercoralis infection at least in settings similar to our study area . Prevention of human contact with dog feces and of dog contact with human excrement as well as anthelmintic treatment of dogs are likely to reduce the exposure of humans to infective S . stercoralis larvae .
|
Infections of humans with the nematode Strongyloides stercoralis can persist for a very long time , due to the capacity of this pathogen to undergo an autoinfective life cycle and re-infect the same host over and over again . Clinical manifestation , known as human stongyloidiasis , may be fatal and can arise many years and generations of worms after the initial infection occurred . Although Strongyloides stercoralis has been known as the causative agent of strongyloidiasis for a very long time , some key questions about its biology and epidemiology remain open . Here we address two of them . Firstly , it has long been known that dogs can serve as experimental hosts for S . stercoralis but it is a matter of debate whether Strongyloides spp . found in dogs in the wild are human pathogenic S . stercoralis , and whether dogs therefore are a source of zoonotic transmission of this parasite . Here we show that dogs carry two genetically different populations of Strongyloides spp . one of which is shared with humans . This demonstrates that dogs represent a possible reservoir for zoonotic strongyloidiasis . Secondly , the all female , parthenogenetic parasitic generations may alternate with single facultative free-living generations , which consist of both females and males . In spite of the presence of both sexes , it had been postulated that males do not contribute genetic material to the progeny and that sperm are merely required to trigger parthenogenetic embryonic development . Here we show that the free-living adults of S . stercoralis reproduce sexually .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion",
"Conclusions",
"and",
"outlook"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"vertebrates",
"dogs",
"animals",
"genetic",
"mapping",
"mammals",
"invertebrate",
"genomics",
"energy-producing",
"organelles",
"mammalian",
"genomics",
"bioenergetics",
"mitochondria",
"strongyloides",
"stercoralis",
"cellular",
"structures",
"and",
"organelles",
"strongyloides",
"pathogenesis",
"animal",
"genomics",
"biochemistry",
"haplotypes",
"cell",
"biology",
"host-pathogen",
"interactions",
"heredity",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"genomics",
"amniotes",
"organisms"
] |
2017
|
Different but overlapping populations of Strongyloides stercoralis in dogs and humans—Dogs as a possible source for zoonotic strongyloidiasis
|
Comprehensive information on the timing and location of gene expression is fundamental to our understanding of embryonic development and tissue formation . While high-throughput in situ hybridization projects provide invaluable information about developmental gene expression patterns for model organisms like Drosophila , the output of these experiments is primarily qualitative , and a high proportion of protein coding genes and most non-coding genes lack any annotation . Accurate data-centric predictions of spatio-temporal gene expression will therefore complement current in situ hybridization efforts . Here , we applied a machine learning approach by training models on all public gene expression and chromatin data , even from whole-organism experiments , to provide genome-wide , quantitative spatio-temporal predictions for all genes . We developed structured in silico nano-dissection , a computational approach that predicts gene expression in >200 tissue-developmental stages . The algorithm integrates expression signals from a compendium of 6 , 378 genome-wide expression and chromatin profiling experiments in a cell lineage-aware fashion . We systematically evaluated our performance via cross-validation and experimentally confirmed 22 new predictions for four different embryonic tissues . The model also predicts complex , multi-tissue expression and developmental regulation with high accuracy . We further show the potential of applying these genome-wide predictions to extract tissue specificity signals from non-tissue-dissected experiments , and to prioritize tissues and stages for disease modeling . This resource , together with the exploratory tools are freely available at our webserver http://find . princeton . edu , which provides a valuable tool for a range of applications , from predicting spatio-temporal expression patterns to recognizing tissue signatures from differential gene expression profiles .
Spatio-temporal gene expression information is fundamental for understanding the intricate mechanisms of embryonic development , which involves both precisely controlled differentiation into dozens of tissue types , as well as dramatic temporal transitions such as gastrulation and segmentation . In the model organism Drosophila melanogaster , decade-long efforts in systematically characterizing developmental and tissue-specific expression patterns have greatly advanced the understanding of embryonic development and tissue formation [1–3] . Large-scale projects involving in situ hybridization and imaging , most representatively the Berkeley Drosophila Genome Project ( BDGP ) [1] , have resulted in thousands of gene expression patterns now manually annotated to very specific tissues with controlled vocabularies [1] . These measurements render Drosophila embryonic development one of the most comprehensively characterized systems of spatio-temporal transcriptional regulation . However , the current compendium of embryonic expression is not without limitations , even for this deeply investigated model organism . For many tissues only a small number of gene’s have been annotated to be expressed , with a median of 80 genes for each tissue , which is likely an underrepresentation of the true number of tissue-expressed genes . In addition , more than 4 , 000 protein-coding genes , as well as many non-coding genes , have not been annotated at all [1] . Moreover , the expression measurements via classical in situ hybridization and imaging are typically qualitative or grossly semi-quantitative [4] . Lastly , the limited dynamic range of standard in situ hybridization protocols that are typically used in large-scale efforts restricts the detection of lowly expressed genes . To address these limitations , in situ data can be complemented by integrating them with the ever-growing transcriptome data measured by microarray or RNA-seq . With their whole-genome coverage and quantitative nature , the advantages of transcriptome profiles are highly complementary to those of in situ data . Even for cases where the transcriptome profiles are not tissue- or stage-specific , they are an information-rich source of co-regulated gene expression patterns , such as those of genes within the same regulatory or biochemical pathways . In these datasets , as gene expression is measured in a diversity of genetic or environmental conditions , functionally related genes are often co-expressed or perturbed together [5 , 6] . Using machine learning algorithms and known tissue-specific gene standards , these co-regulatory patterns can be extracted to infer tissue specific expression of genes , which combine advantages of both data types . Our previous work has shown that such an approach can accurately predict cell type-specific gene expression in the human kidney [6] and worm tissues [5] from non-tissue specific data , using manually curated cell type gene markers . In D . melanogaster embryonic development , the well-annotated resource of gene expression across tissues and developmental time , provide a unique opportunity to systematically predict not only tissue-specific expression but also the temporal stage of gene expression . Prior work has demonstrated the feasibility of making such predictions in a small set of 15 tissue-stages [7] . Moreover , considering the established tissue lineage information for D . melanogaster development [1] , which has been curated into a tissue developmental ontology , the prediction algorithms can be further improved by integrating information across tissue-stages while considering the ontological relations between cell types . Here , we present an approach for the prediction of gene expression during embryonic development in D . melanogaster based on integrating the tissue specificity of in situ hybridization data with the quantitative and comprehensive measurements from public genomic data compendium . The algorithm , which we refer to as structured in silico nano-dissection ( SIND ) , improves individual tissue predictions with a cell lineage-aware prediction integration stage . This stage learns and applies transcriptional dependencies between developmentally and anatomically related tissue-stages , with lineage knowledge encoded in the structure of a probabilistic graphical model based on the tissue developmental ontology . Using our SIND algorithm , together with in situ gene annotations and a compendium of 6 , 378 genome-wide expression and chromatin immunoprecipitation ( ChIP ) datasets , we predicted global expression patterns in 282 tissue-developmental stages , including >4000 genes without prior in situ data . The predictions were validated both computationally , via hold-out validation , and experimentally in the case of new predictions without prior literature evidence . We verified predictions in muscle and brain for 13 novel genes with no prior in situ data or literature evidence to the extent of our knowledge , for both tissue-specificity and temporal specificity , including co-expression patterns in multiple tissues , demonstrating the accuracy of our predictions . The complete and quantitative nature of the predictions enables previously infeasible analysis of tissue specificities , such as analyzing tissue-specific signatures in whole-animal differential expression datasets and mapping human diseases to D . melanogaster tissue/stages for aiding the development of human disease models in the fly . We provide all our predictions and exploratory tools in a user-friendly web interface FIND ( fly in silico nano-dissection , find . princeton . edu ) that allows the user to query predicted gene expression pattern and perform tissue-specificity prediction task .
As an overview ( Fig 1 , see also Methods ) , our approach first predicts gene expression probabilities for each tissue-stage term individually by learning linear models of transcription and chromatin signals that predict tissue-stage specific expression levels using a structural support vector machine ( SVM ) algorithm , and then integrates individual tissue-stage term predictions with a global developmental lineage-aware probabilistic graphical model . We named the Drosophila tissue-developmental stage gene expression prediction system FIND . Conceptually , in the first stage , our method discovers and exploits the co-regulatory patterns of cell lineage-specific genes from high-throughput measurements of samples from diverse tissue origins and under a wide range of genetic , chemical , or environmental perturbations [5 , 6] . The cell lineage-specific genes were prioritized based on a weighted subset of conditions found to be informative by the algorithm . Notably , the high-throughput datasets we used as input do not have to be as tissue-specific as the predicted target tissue-stage , because we only require the co-regulatory patterns of tissue-specific genes to be shared across a subset of datasets [6] . In fact , most of the tissues we make predictions for are fine embryonic structures without existing high-throughput microarray or RNA-seq data specific for those structures . In the second stage , we use prior developmental cell lineage knowledge to help integrate individual tissue-stage specific predictions with a conditional random field ( CRF ) probabilistic graphical model . Developmental lineage knowledge is expected to be informative for improving spatial-temporal predictions . This is both because gene expression in a precursor tissue usually provides informative insight into gene expression patterns of the tissue it develops into , and because related tissues , e . g . different cell types in the nervous system , are likely to share expression patterns for many genes . Specifically , we estimate the quantitative gene expression dependencies between developmentally related tissues empirically from in situ annotations and use it for integrating predictions . For each gene , the probabilistic graphical model treats expression in each tissue-stage term as a binary random variable . The value of each tissue-stage variable is dependent on both the input to the model , which are the SVM predictions from individual classifiers , and the values of other tissue-stage variables that are connected in the tissue developmental ontology . As the expression for each tissue is globally predicted considering dependencies between all tissue-stage terms , this approach allows for borrowing information from relevant tissues . Spatio-temporal gene expression during Drosophila embryogenesis has been extensively studied via large-scale in situ hybridization projects , such as the BDGP in situ database [1] and Fly-FISH database [2] . These studies provide a very rich resource of spatio-temporal specificity of gene expression and are used as training standards for our models . Our genome-wide data integration approach complements the current in situ hybridization data in multiple ways: our approach makes predictions for plenty of genes ( >4 , 000 ) unmeasured or undetected by in situ , computes quantitative scores comparable across genes and tissues and predict expression probabilities across all tissues , reducing the bias of manual annotation towards the most prominent or easy to recognize tissues . We first obtained a compendium of 6 , 378 high-throughput transcriptomic and chromatin profiling experiments , in addition to training standards consisting of in situ annotations for 282 tissue-stage terms spanning 6 time periods of fly embryonic development ( outlined in S1 File ) . The developmental cell lineage information is encoded by a tissue developmental ontology based on an earlier work [6] ( Supplementary File 1 ) . We then used this data to train support vector machine classifiers and global integration model . We obtained a performance of median area under receiver-operating characteristics ( AUROC ) 0 . 87 across 282 tissue-stage terms ( Fig 2a and 2b , S2 File ) . The predictions for each gene were made by five-fold cross-validation ( see Methods ) . Integrating individual tissue predictions with our conditional random field ( CRF ) model is a major factor for enhancing performance , in addition to utilizing a wide range of datasets and data types ( S1 Fig ) . Using the ontology-based CRF method that we developed provides a consistent improvement across almost all tissues ( Fig 2b ) , demonstrating the importance of the integrative approach allowing sharing information cross tissue-stages with developmental cell lineage . The global topological structure of the predicted tissue-stage-specific transcriptomes , as visualized by LDA and PCA , shows organization by both developmental stage order ( S2 and S3a Figs ) and by embryonic layer of origin ( Fig 2c and S3b Fig ) . For each gene , our predictions specify the temporal range of expression and the specific cell lineages at multiple time points in the developmental trajectory . This is demonstrated for CG12792 , which is a ubiquitously expressed genes at early developmental stages , with more restricted expression at 13 hours after egg laying ( AEL ) ( Fig 2d ) . This developmental trajectory is well captured by the cross-validated predictions when all expression labels for these genes were held out ( Fig 2e ) . Based on our predictions , the probabilities for ubiquitous expression decrease during development , while the probabilities for specific cell lineages ( muscle and midgut ) increase over time . Importantly , these increasingly restricted expression patterns in those specific tissues are an accurate recapitulation of the expression dynamics detected by the in situ experiments ( Fig 2d ) . To systematically measure the performance of spatial and temporal pattern predictions , we compared the predictions of each gene to the expected tissues and temporal stages defined by in situ and to the tissue-stages with no detectable in situ expression . This is another important dimension of prediction performance , besides correctly ranking genes for each tissue . The distribution of AUROC for genes is shown in ( S4 Fig ) . Even for genes with very complex expression patterns ( > 20 annotations ) , an average AUROC of 0 . 89 is obtained ( S4 Fig ) . We examined the ability of the model to predict expression patterns ( outside of the training set ) by both collecting gene expression information from the literature and performing fluorescent in situ hybridization ( FISH ) experiments for genes where no prior tissue-developmental mapping was available . We started by examining gene expression patterns in two tissues ( BDGP in situ ontology categories ) that had both a large number of genes with high scoring predictions ( all predictions above threshold 0 . 8 ) and are easy to assess in in situ images: brain primordium ( stages 11–12 ) and embryonic muscle system ( stages 13 and later ) . We found that all genes newly predicted to be expressed in theses tissues with no prior annotations were confirmed . For brain primordium , all 5 genes with detectable FISH signals as predicted by FIND were indeed highly expressed in brain at stages 11 or 12 ( S1 Table , Fig 3a ) . For 11 more genes , literature evidence corroborated the FIND predictions ( S1 Table ) . Thus 16/16 tested genes exhibit the FIND- predicted expression pattern for brain primordium . For the embryonic muscle system ( stages 13 and later ) all 8 genes with detectable FISH signals showed patterns in muscle systems , including in the somatic muscle , visceral muscle and dorsal vessel ( the fly embryo heart ) ( S2 Table , Fig 3b–3d ) . For 5 more genes , literature evidence again corroborated the FIND predictions ( S3 Table ) . Thus 13/13 tested genes exhibited the FIND-predicted expression pattern for embryonic muscle system . Importantly , in all cases where we have spatio-temporal expression information , the transcript was expressed on the predicted tissue , underscoring the precision of the method . We have also tested predictions for very specific tissues and organs , such as late embryo Malpighian tubules or salivary glands . We performed FISH for four ( Malpighian tubules ) and five ( salivary glands ) genes with more than 20-fold enrichment above background , and in all cases we were able to detect expression on the respective tissues ( S5 Fig ) . Thus , we verified all high confidence predictions that we tested , indicating the utility of using FIND predictions to guide future studies . To determine how FIND predicts new multi-tissue expression patterns , we compared the predicted spatial-temporal patterns with experimentally measured gene expression across tissue and developmental stages . A particularly interesting example is the gene Ppn ( Papilin ) . The main cell type where this gene is expressed is the plasmatocyte ( Fig 3c ) , which is consistent with previous observations from Fly-FISH [2] . In addition , we could confirm the previously reported expression in the dorsal vessel [8] and also detected Ppn expression in the visceral muscle and somatic muscle primordia ( Fig 3c , bottom ) . Importantly , our model gave all of these tissues high prediction scores for ppn expression ( Fig 3c , top ) , showing that complex expression patterns can be accurately predicted from our method . We next assessed how the variation of gene expression probabilities across stages are related to temporal variation in expression levels during development for specific genes . For example , in the case of TpnC47D , the predicted expression scores ( Fig 4a ) are very high for the term embryonic larval muscle system ( stages 13–16 ) but much lower for the precursor term muscle system primordium ( stages 11–12 ) . Indeed , experimental data confirmed this sharp up-regulation of expression , as expression in muscle is only detected from stage 13 and completely absent at earlier stages ( Fig 4a ) . Regulation of bsf expression provides another complex and interesting example . bsf is predicted to have similar expression level throughout stages 11–16 in midgut but with an increase of ~3 fold in muscle for the same temporal transition . As predicted , the bsf gene has a clear pattern of co-expression in the midgut and muscle at stage 13 and later , but when looking at stage 12 , the expression in muscle is much lower than in the midgut primordia ( Fig 4b ) . In conclusion , joint analysis of predictions for multiple tissue-developmental stage terms ( either different tissues at a fixed time interval , or across time intervals for one tissue ) can be a very valuable tool to guide the investigation of complex spatial-temporal expression patterns for individual genes or groups of genes . In addition to using spatiotemporal expression pattern predictions to generate hypotheses for individual genes , these predictions can also enable new spatio-temporal specificity analysis of high-throughput experimental data . In case studies below , we showcase two new directions of such applications . High-throughput expression profiling and differential expression analysis are common and important techniques for exploratory analysis of molecular responses to different perturbations . While perturbations such as genetic ablation of a specific gene are expected to affect certain tissues preferentially , this is usually difficult to assess since , for practical reasons , expression profiling experiments are often performed on the whole body or tissue homogenates . FIND’s quantitative score for every tissue and stage for every gene in the genome can detect tissue-specificity signals from whole-organism differential expression experiments , even in the absence of reliable tissue-specific markers . This is particularly useful for tissues with few annotated marker genes . As a proof-of-concept , we used three differential gene expression datasets comparing whole-embryo mutant vs . wild type to predict the tissues most affected by these perturbations independently of any prior knowledge other than FIND predictions ( Table 1 ) . For example , Myocyte enhancer factor 2 ( Mef2 ) loss-of-function mutants have defects in myoblast fusion and terminal muscle differentiation [9] , consistent with the role of Mef2 as an essential conserved transcriptional regulator of muscle development . We use the transcriptome analysis of this mutant ( E-TABM-57 ) as input for FIND-analysis , and indeed this affected tissue is precisely detected by our computational method: all top six terms ( out of 282 ) predicted to be the most affected in the mutants are muscle tissues ( Table 1 ) . Specifically , we rank tissues based on Spearman correlation of the differential gene expression log fold changes and the prediction scores for each of the 282 tissue-stages . The second example is Twist , an upstream regulator of Mef2 that has an earlier and broader role in mesoderm development [10] . Twist loss-of-function mutants lead to a block in gastrulation and a loss of mesodermal cells ( E-TABM-162 ) . Accordingly , our method predicts the loss of a wide range of mesoderm derivatives in twist loss-of-function mutants , and at earlier stages than those predicted for Mef2 ( Table 1 ) . Finally , the third differential expression dataset analyzed a development time-course of trachealess ( trh ) mutant versus wildtype whole embryo samples . The trh gene codes for an essential transcription factor in the development of the tracheal system [11] ( E-GEOD-28780 ) . In agreement with this role , four out of six of the top predictions are tracheal tissue terms . The other two top predictions are related to the salivary gland duct , another major tissue affected by trh loss-of-function mutants [11] ( Table 1 ) . In summary , for all three mutants , we are able to make accurate and informative tissue-specificity predictions from experiments performed on whole embryos , without using any prior knowledge of the functions of these genes or phenotypes of their ablation , We therefore expect our tissue expression predictions and methods to be useful for improving detection of tissue signals in other high-throughput experiments where the available known tissue markers are insufficient . While Drosophila melanogaster provides a flexible and powerful genetic model for a wide range of human diseases , it is not always clear which tissue is the best choice for modeling a certain human disease . This is especially relevant when a direct tissue equivalent is absent in the fly . To address this challenge , we examined FIND-predicted tissue-stage patterns in the fly embryo for human disease genes . Specifically , we identified fly orthologs of human disease gene from OMIM database , using a functional mapping approach which identifies gene that are not only similar in sequence , but also share similar roles in pathways in human and fly [12] . Then , embryonic tissues were ranked based on their enrichment of genes related to each disease , thus prioritizing fly tissues by their relevance to human disease . Among the diseases with most significant tissue enrichment , muscle diseases and cardiomyotrophy was mapped to somatic muscles ( Fig 5 ) . Despite the distinct morphology of fly eyes , larval eye primordium is selected as the best tissue for modeling for eye diseases ( Fig 5 ) . Indeed , fly eye is known to share key developmental biology with mammals , for example , the Drosophila eyeless and its mammal ortholog Pax6 are both essential to eye development [13] . Auditory system diseases are best mapped to the dorsal head epidermis , as well as sensory complex ( Fig 5 ) . Even though flies lack a skeletal system , bone diseases are mapped best to mesoderm primordium tissues , suggesting that at least their early development during mesoderm stage may be modeled ( Fig 5 ) . In conclusion , the results show that genome-wide predictions of tissue-stage expression can assist in identifying and developing new human disease models . By integrating diverse expression and chromatin profiles with in situ tissue-specific gene standards , we provided accurate predictions of spatio-developmental expression patterns for all annotated genes , both coding and non-coding . Enabling our approach is a cell lineage-aware computational method , SIND , that uses a unified spatial-temporal transcription probabilistic graphical model to integrate predictions for developmentally related tissues informed by tissue developmental ontology . We have both computationally and experimentally verified the potential of these predictions , confirming the accuracy and predictive power of our method . The whole-genome coverage and quantitative resolution of our approach thus can not only provide researchers with predicted spatio-temporal expression patterns for over 4 , 000 genes that lack experimental evidence , but also power their data analyses , including detection of tissue-specificity signals in non-tissue resolved ( e . g . whole embryo ) expression datasets and identification of most relevant tissues for study of human disease . To facilitate the exploration of our spatio-developmental gene expression predictions and downstream analyses , we provide a user-friendly web interface ( http://find . princeton . edu ) . This web tool allows for querying tissue-gene predictions , discovering genes with similar tissue-specific expression patterns and predicting tissue-stage specificity from a user specified list of genes ( e . g . differential gene expression after a perturbation experiment ) . It is important to note that our approach currently covers only embryonic developmental stages due to limitations in availability of in-situ data at other stages for training; this could be addressed by applying this approach to additional temporal stages as training data becomes available . Furthermore , the ability to predict gene expression patterns in highly-specific cell types depends on the input data compendiums being informative of the gene co-expression patterns in these cell types and the availability of known expressed genes from these specific cell types as labels . Single cell data can thus potentially contribute to training both as input data and as labels in the future , although such an approach will need to address noisiness and the challenges of robust definition of cell types from single cell data . Lastly , of course , even with high accuracy and coverage , the resources that we provide are still computational predictions , and should be treated as such . Their power is in enabling analysis and directing follow-up experiments . Finally , we note that this resource is built upon , and complementary , to current genomic data , and will continually benefit from a progressively growing amounts of in situ tissue specificity standards and RNA-seq and chromatin profiling data , as well as integrating new types of genomic features like sequence features or new experimental techniques . This flexibility is an important property for such a dynamic field as functional genomics .
The data for training the Drosophila melanogaster spatio-developmental expression prediction model are obtained and processed as follows . Drosophila melanogaster raw microarray data for the two major platforms GPL1322 and GPL72 were downloaded from NCBI Gene Expression Omnibus at May 2014 and processed with the GEO2R package . Missing values in microarrays were imputed with KNNImpute [14] . Drosophila melanogaster RNA-seq and ChIP-seq raw reads data were downloaded from NCBI Sequence Read Archive ( SRA ) and aligned to BDGP5 fly genome assembly using bowtie2 with local alignment mode [15] . RNA-seq alignment data was summarized by RPKM using BDGP v5 . 78 gene annotation ( and lncRNA annotation from [16] ) . Non-transcriptomic sequencing alignment data was summarized by RPKM within ± 500bp region to TSS . modENCODE RNA-seq dataset was downloaded from [16] . The datasets used are listed in S3 File . All sequencing samples were log-transformed after adding a pseudo-count . The pseudo-count was determined as half of the value of the lowest non-zero gene for each sample . All samples were standardized to mean 0 and variance 1 . Tissue- and developmental stage- specific expression standards were obtained from BDGP in situ database ( insitu . fruitfly . org ) and FlyFISH database ( fly-fish . ccbr . utoronto . ca ) on May 2014 . Genes included in both microarray platforms were used . Since BDGP in situ annotation are more comprehensive in fine tissue structures , we use BDGP as the main standards while using FlyFISH as assistance standards that were only used to train first-stage classifiers . The subset of genes mapped to microarray platforms were used , with a holdout set strictly not used in any steps of algorithm design and the decision of hyperparameters for evaluation . Our Drosophila embryonic development tissue developmental ontology is modified from [1] with a few changes adapted from flybase tissue developmental ontology FBdv [17] . The ontology we used is provided in S1 File . Our machine learning algorithm uses a cell-lineage aware approach to predict gene expression with spatio-developmental specificity based on whole-genome scale expression data . The cell lineage information is critical for specific tissues and stages that can benefit from considering developmental context including its precursor and descendant tissues . Our model takes in situ gene expression annotations and functional genomics data as input and performs prediction in the stages . In the first stage , one individual classifier was trained for each individual tissue-stage term . Specifically , for each tissue-stage term , a linear structural SVM model was trained to predict in situ annotations ( whether a gene is annotated to a tissue-stage ) from functional genomic datasets including transcriptome profiling data ( microarray and RNA-seq ) and chromatin profiling data ( ChIP-seq ) . The model was trained using the SVMperfer tool in the Sleipnir library [18] . In the second stage , we use a conditional random field classifier , which integrates individual tissue-stage predictions using learned dependencies of gene expression between tissue-stage terms . The conditional random field model represents each tissue-stage as a node and cell-lineage relations as edges connecting the nodes . Tissue-stage nodes are connected with the following rules: The CRF model that directly provides probability output for each gene and each tissue is formulated as the following P ( Y|X ) =1Ze∑m ( WmX+αm ) Ym+∑ ( m , n ) ∈GontologyJmnYmYn P ( Y│X ) describes the conditional probability model . The conditional probability is determined by a term that describes dependencies on first-stage individual classifiers output ∑m ( WmX + αm ) Ym and a cell lineage dependency term ∑m∑n<mJmnYmYn . Note that if the second term does not exist the model reduces to multiple logistic regression models . Specifically , Y is the vector of indicator variables for expression in tissue-stages , where Ym = 1 indicates expression in the m-th tissue-stage and Ym = 0 indicates no-expression in the tissue-stage indexed by m . X represents a matrix containing all SVM predictions from the first stage classifiers , in which a column corresponds to a classifier . The model is also illustrated by a diagram in S6 Fig . W , J , a are estimated from the input ( first stage predictions and in situ annotations ) . Specifically in the first term Wm represents weight the m-th tissue-stage classifier from the first stage , which is analogous to coefficients of a logistic regression model and αm is the intercept . We allow a tissue-stage to be informed by predictions for any tissue-stage classifiers in the first term , which integrate expression signals over expression predictions for multiple tissue-stage-specific classifiers based on annotations from both BDGP in situ and FlyFISH . In the second term , Jmn represents interaction between tissue-stage m and n and only edges in the tissue developmental ontology are allowed to be non-zero . Gontology represent the ontology based graph structure which is a set of edges built with rules described above . Prediction for a tissue-stage can either increase or decrease probability of a connected tissue-stage term depending on the estimated dependency scores J . Z is the normalizing constant which ensures the sum of probabilities equals 1 , and it does not need to be explicitly computed in our learning algorithm . The model can be trained by minimizing the L1 Penalized likelihood ∑log P ( Y│X ) + λ|J| . The L1 regularization λ|J| is used to avoid overfitting to tissue-stage dependencies by encouraging J to be 0 if it contributes little to the likelihood . |J| represents the L1 norm of interactions J which equals sum of absolute values of all entries of J . λ is the L1-regularization parameter . To prevent overfitting to the in situ annotations , for the training of the second stage model , all predictions from the first stage were made on holdout genes with 5 fold cross-validation . Specifically , for each fold , 4/5 of the genes were used for training and the trained model was used to predict the rest 1/5 holdout genes . The concatenated predictions on the holdout set for each fold , which covers all the training set genes , were used as input to the second stage classifier . As the concatenated predictions were all predicted using models that have not seen the gene before , it minimizes the bias due to overfitting in the second stage . Next , we describe the optimization algorithm for minimizing the objective function . Computing the exact gradient of the CRF likelihood over the ontology-based graph is intractable since the graph is not tree or chain structured . Approximation inference algorithms , such as pseudo-likelihood-based method is biased but computationally efficient , while Markov chain Monte Carlo ( MCMC ) sampling-based inference methods converge to the exact gradient in the limit of large sample size , but is computationally expensive to obtain low variance gradient estimate . To address both speed and accuracy , we choose to use a hybrid of both . We first obtain a fast approximation of parameters by optimizing a modified pseudo-likelihood objective , ∑i∑mlogP ( Ym ( i ) |X ( i ) , Y-m ( i ) ) +λ|J|=∑i∑me∑m ( WmX+αm ) Ym+∑n<mJmnYmYne∑m ( WmX+αm ) Ym+∑n<mJmnYmYn+1+λ|J| for which the gradient can be computed efficiently because it does not involve computing Z which is intractable , and initialize the second step optimization which optimizes the true objective with the estimators from first step . The L1 regularized optimization problem was solved with Projected scaled sub-gradient ( Gafni-Bertsekas variant ) as implemented in ( https://www . cs . ubc . ca/~schmidtm/Software/thesis . html ) . For the second step , we estimate the stochastic gradient of the model objective by Gibbs sampling and employ stochastic gradient descent with momentum for optimization . For Gibbs sampling based gradient estimation , we first rewrite the likelihood as P ( Y|X ) =1Ze∑mfm ( X ) Ym+∑m∑n<mJmnYmYn To compute the stochastic gradient for the log-likelihood when conditioned on X , we used the fact ( ignored writing down conditioning on X for simplicity ) ∂log ( P ) ∂fm=Ym¯-E ( Ym ) , ∂log ( P ) ∂Jmn=YmYn¯-E ( YmYn ) where expectations E ( Ym ) and E ( YmYn ) are estimated by MCMC sampling . For each iteration , we run a separate Markov chain for each gene and average the stochastic gradients obtained from each chain over all genes . To address efficiency and burn-in time of MCMC , we initialize MCMC chains from the last sample of the previous iteration ( also called persistent MCMC sampling ) , and for each iteration , we only run one sweep of Gibbs sampling which sample each node exactly once and use the single sample Y* to compute estimator E ( Ym ) =Ym* and E ( YmYn ) =Ym*Yn* . The Markov chains for all genes are parallelized for speeding up . For each iteration , we increment model parameters based on stochastic gradient descent with specified step size . We use a learning rate annealing scheme of 100 , 000 iterations of learning rate 1e-6 , followed by 50 , 000 iterations of 1e-7 and 50 , 000 iterations of 1e-8 . Additionally , a momentum term was used for stabilizing the update and improving convergence . L1 regularization was applied to control overfitting . We set momentum to 0 . 9 and L1-regularization parameter to 8 . After the model is trained , for prediction from the second-stage CRF model we ran 80 , 000 rounds of Gibbs sampling and average over the samples . For genes with in situ annotations , we predict their expression with 5 fold cross-validation . Human disease gene sets were obtained from OMIM . We selected the functional ortholog of human genes in Drosophila melanogaster using functional knowledge transfer ( FKT ) method with p-value cutoff 0 . 1 [12] . For each fly gene the lowest p-value human ortholog was chosen , then each tissue-stage expression prediction profile is mapped to human genes accordingly ( a total of 4937 fly genes were mapped with above cutoff confidence ) . Statistical significance of gene set enrichment in each mapped tissue-stage expression prediction profile was computed with parametric analysis of gene set enrichment ( PAGE ) method [19] . Fluorescent in situ hybridizations were performed as previously described [20] . Briefly , we used cDNA clones from the DGRC collections DGC1 and 2 [21] for preparing digoxigenin-labelled or fluorescein-labeled RNA probes , except for the CG14688 gene for which we use a previously reported cDNA template [22] and CG34284 for which we amplified a cDNA fragment . In this case , we performed Biozym Red HS Master Mix ( Biozym ) using as template total cDNA obtained from mixed stage of D . melanogaster wt embryos ( an overnight embryo collection ) and the primers GTCGGTCATTGAAAGTTTCC and CCGAACATATTCCAAATCTG ) . The amplified fragment ( 495 pb ) was cloned into pGEM-T easy ( Promega ) and the positive clones were verified by sequencing . Probes were detected with peroxidase-conjugated antibodies ( Roche ) and developed using the TSA-plus Tyramide fluorescence system ( Perkin Elmer ) for fluorescein or Cy3 fluorophore deposition . Imaging was performed in a Zeiss LSM 780 confocal microscope , using a Plan-Apochromat 20x/0 . 8 objective . For fluorescein detection , we used the 488nm wavelength of a multiline Argon laser , a MSB 488/561 dichroic mirror and a spectral detector set for a 500-600nm detection range . For Cy3 detection , samples were excited with a DPSS laser at . For DAPI detection , samples were excited with a diode at 405nm and the emission fluorescence was detected with a MBS405 dichroic mirror . Other acquisition parameters , such as PMT gain and offset , were set independently for each probe , dependent on the signal . After the acquisition , images were processed using ImageJ 1 . 46r [23] . Although a Z-stack was routinely acquired , a single focal plane was manually selected for visualization , except for the images of Malpighian tubules and salivary glands , where a Z-projection was used as indicated . Post-acquisition processing consisted on cropping and orienting single embryos and adjusting minimum and maximum intensity values for maximizing contrast . The code is available at https://github . com/FunctionLab/find_scripts . The gene expression prediction dataset is available at http://find . princeton . edu/predictions/download/ or https://doi . org/10 . 5281/zenodo . 3408411 .
|
When and where a gene is expressed is fundamental information for understanding embryonic development . Current knowledge for such expression patterns is typically far from complete . Even for the long-standing model organism , Drosophila melanogaster , with large-scale in situ projects that have provided invaluable expression information for many genes , 40% of the genes still lack spatio-temporally resolved expression information . Such data is complemented by transcriptome datasets such as microarray and RNA-seq , which have whole-genome coverage and measure expression levels with greater dynamic range , but they typically lack precise spatio-temporal resolution . To bridge this gap , we developed a machine learning approach that combines the spatio-temporal resolution of in situ data with the accurate quantification and whole-genome coverage of genomic experiments , integrating information from 6 , 378 expression and chromatin profiling data sets . With this new approach , we present a genome-wide resource of spatio-temporal gene expression predictions for over 200 tissue-developmental stages during Drosophila embryogenesis . This resource is experimentally validated to have high-quality predictions , can guide the discovery of new tissue-specific genes , and provides a new tool to perform genome-wide analyses of spatio-temporal specificity .
|
[
"Abstract",
"Introduction",
"Results",
"Methods"
] |
[
"invertebrates",
"machine",
"learning",
"algorithms",
"medicine",
"and",
"health",
"sciences",
"muscle",
"tissue",
"applied",
"mathematics",
"pharyngeal",
"muscles",
"animals",
"simulation",
"and",
"modeling",
"animal",
"models",
"algorithms",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"mathematics",
"artificial",
"intelligence",
"genome",
"analysis",
"experimental",
"organism",
"systems",
"embryos",
"drosophila",
"research",
"and",
"analysis",
"methods",
"embryology",
"computer",
"and",
"information",
"sciences",
"musculoskeletal",
"system",
"animal",
"studies",
"gene",
"expression",
"biological",
"tissue",
"muscles",
"insects",
"arthropoda",
"machine",
"learning",
"eukaryota",
"anatomy",
"genetics",
"transcriptome",
"analysis",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"gene",
"prediction",
"computational",
"biology",
"organisms"
] |
2019
|
Accurate genome-wide predictions of spatio-temporal gene expression during embryonic development
|
Many daily situations require us to track multiple objects and people . This ability has traditionally been investigated in observers tracking objects in a plane . This simplification of reality does not address how observers track objects when targets move in three dimensions . Here , we study how observers track multiple objects in 2D and 3D while manipulating the average speed of the objects and the average distance between them . We show that performance declines as speed increases and distance decreases and that overall tracking accuracy is always higher in 3D than in 2D . The effects of distance and dimensionality interact to produce a more than additive improvement in performance during tracking in 3D compared to 2D . We propose an ideal observer model that uses the object dynamics and noisy observations to track the objects . This model provides a good fit to the data and explains the key findings of our experiment as originating from improved inference of object identity by adding the depth dimension .
Throughout daily life we need to monitor and track our surroundings , avoiding collisions while walking , cycling or driving . This ability is based on estimating our self-motion and the motion of objects around us using visual cues such as retinal motion , binocular disparity , relative size and motion parallax . A complexity arises because these cues are noisy and often ambiguous; for example , both a moving object and an eye movement create retinal motion . To form inferences about how objects move in the world around us , the brain must therefore disambiguate cues and integrate noisy information . Here , we focus on the complexities involved in tracking with multiple moving objects . Within the laboratory , our ability to track moving objects is typically investigated using a multiple object tracking ( MOT ) paradigm , in which a subject tracks a subset of targets out of a larger number of objects as they move on a 2D screen . These experiments have a long history showing that many factors influence our tracking capability . Tracking accuracy appears to decline with increasing object speed [1] , number of objects [2] and the relative closeness of objects [3] , but can be increased by simply coloring objects differently [4] or altering object shapes [5] . Although these findings may seem disparate , Vul et al [6] have recently provided a normative explanation . The authors view multiple object tracking as a data association or correspondence problem , referring to a problem that broadly needs to be solved in cognitive behaviors , such as in the matching of binocular images for stereovision or to prevent multiple items to be swapped when stored in memory [7 , 8] . Vul and colleagues modeled object tracking by devising an ideal observer model in which the uncertainty of position and velocity signals affects how well these signals can be associated to the objects that caused them . In the model , the position uncertainty increases with number of tracked objects , similar to other suggestions [9] . As a result of this uncertainty , noisy position measurements cannot distinguish between objects if these are close together . The model , however , also uses velocity signals to distinguish between objects , which is especially useful when they are close . However , as objects move faster , the velocity measures will become more uncertain , so that at high velocities , the ability to distinguish between objects will decline and the predictions about their future position will deteriorate . This causes performance to decline as objects move closer together and as they move faster . While multiple object tracking studies generally focus on objects moving in the two-dimensional frontoparallel plane , this is an atypically simple , special case . In real life , objects move continuously in all three dimensions [10–14] . If multiple object tracking reflects an association problem , then adding depth information may promote tracking . More specifically , two objects that move closely together from a two-dimensional frontoparallel perspective but far apart in depth , may still be correctly associated using depth cues . Indeed , Ur Rehman , Kihara , Matsumoto , & Ohtsuka [15] already reported that tracking performance improves when the moving objects are separated by moving in different depth planes . But , again , during realistic 3D object motion objects are not restricted to moving only in different depth planes . How is tracking performance affected when realistic depth cues and continuous motion in depth are present ? Thus far , only few studies have performed a direct comparison between object tracking in 2D and 3D . Both Liu et al [16] and Vidakovic & Zdravkovic [17] added monocular pictorial depth cues to the scene , but found no significant improvement in object tracking , suggesting that such cues are not precise enough to help solve the correspondence problem . Of course , this cannot be generalized to all depth cues . For example , binocular disparity is the main binocular cue for depth , and known to be more reliable than pictorial cues [18] . In this study , we investigated how tracking performance changes when objects move in continuous 3D space ( displayed using both monocular and binocular cues ) compared to moving in a single depth plane . To assess the role of depth information , we manipulated the average speed and average distance between the objects in all dimensions . Following Vul et al [6] , we constructed four versions of an ideal observer model to test how position and velocity information could be incorporated into multiple object tracking in 3D .
Ten healthy naïve subjects ( 8 female ) , aged 18–30 years , participated in this study . All subjects had normal or corrected to normal vision , including normal stereovision ( tested using the Randot Stereo test ( Stereo Optical Inc . , Chicago , USA ) ) and no known history of neurological or visual disorders . Informed written consent was obtained from all subjects prior to the experiment and the experiment was approved by the Ethics Committee of the Faculty of Social Sciences . One subject failed to comply with the task instructions , and was removed from the subject pool . Visual stimuli were projected using two digital stereo DLP®-rear projection cubes ( EC-3D-67-SXT+ -CP , Eyevis GmbH , Reutlingen , Germany ) on a 2 . 83 X 1 . 05 m ( width X height ) surface with a resolution of 2800 by 1050 pixels . Subjects were seated 1 . 75 m in front of the center of the screen , which thus subtended 77 . 9° X 33 . 4° of visual angle . Vertical retraces of the images were synchronized using an Nvidea Quadro K5000 graphics card . The visual display was updated at 60 Hz . Stereoscopic images were generated using channel separation , based on interference filter technology ( INFITEC® GmbH , Ulm , Germany ) , projecting images for the left and right eye using different wavelengths . Subjects wore a pair of glasses with selective interference filters for each eye and used a chin rest for stabilization . Visual stimuli ( referred to as objects from now on ) consisted of spheres shaded to appear 3D . The shading was constant across objects and depth , which prevented it being used to discriminate different objects . The objects subtended 0 . 5° visual angle at screen depth and were rendered in a virtual space of 3 . 00 m wide , 2 . 00 m high , and 1 . 75 m deep ( 0 . 875 m in front and 0 . 875 m behind the screen ) using their 3D position . The visual scene also contained a stationary yellow fixation cross of 0 . 2° visual angle at screen depth straight ahead of the observer . Objects were rendered in OpenGL using a realistic perspective transformation , thus providing multiple depth cues such as relative size , motion parallax and binocular disparity . During each trial the position of the objects was updated according to a modified Ornstein-Uhlenbeck process , as used by Vul et al [6] . Objects moved according to Brownian motion while being attached to a virtual spring situated at zero ( where zero is the center of the display ) : xt=xt−1+vt ( 1 ) vt=λvt−1−kxt−1+wt ( 2 ) wt∼N ( 0 , σw2 ) in which xt , vt , wt are the position , velocity and random acceleration of the object at time step t , respectively . k is a spring constant which was varied to generate desired dynamics and λ is a damping term which was fixed to 0 . 9 . These dynamics cause the objects’ position and velocity to evolve stochastically but allow their variances to be expressed in closed form . This enabled us to systematically manipulate how close the objects were to each other on average , and how fast they moved on average . Specifically , we calculated the spring constant k and the acceleration variance σw2 , to produce a desired σx: the standard deviation in object position and σv: the standard deviation of their velocity . This was done by assuming that these variances do not change across time steps . The stationary standard deviations of position and velocity of an object are as follows: σx= ( 1+λ ) σw2k ( λ−1 ) ( k−2λ−2 ) ( 3 ) σv=2σw2 ( λ−1 ) ( k−2λ−2 ) ( 4 ) These equations can be rearranged to calculate the spring constant and acceleration variance required to produce a desired σv and σx . σw2= ( λ2−1 ) σv2 ( σv2−4σx2 ) 4σx2 ( 5 ) k= ( 1+λ ) σv22σx2 ( 6 ) We used the same dynamics but independent noise for each dimension of object motion . For the frontoparallel plane these calculations were performed in visual angle and then converted into meters for display . The same value in meters was used for the depth dimension . The subject had to track three out of six moving objects ( see Fig 1 ) . Each trial began with the presentation of six stationary objects for 1 . 5 s . Three of these objects were white and the ones that had to be tracked were cued red . Next , all objects turned white and began to move according to the dynamics described above . The subject had to track the cued objects for 5 s after which all objects stopped moving and one was randomly turned red . The subject had to indicate if this was one of the originally cued objects , using a button press . Then the next trial began . The initial positions and velocities of the objects were randomly sampled using the position and velocity standard deviation for that trial , i . e . , σx and σv , respectively ( see Table 1 ) . Furthermore , objects moved either in 2D ( frontoparallel plane ) or in 3D . Each subject was tested in 60 conditions , split into 6 sessions of 45 minutes , the order of which was counterbalanced across subjects . In each session , we fixed the value of σx to reduce performance effects caused by estimation of the movement dynamics . Within each session , σv was randomly selected on each trial from a set of fixed values ( see Table 1 ) and 30 trials were performed for each value . Prior to each session , subjects performed 15 practice trials , leading to a total of 1890 trials ( 6 sessions * 10 parameter values * 30 trials + 6 practice blocks *15 trials ) . Data were analyzed using Matlab 2014b ( The MathWorks , Natick , MA , USA ) . To assess how tracking accuracy changed as a function of σv , we fit a psychometric curve to the proportion of correct responses for each session . Because of asymmetry in the data we used a cumulative Weibull distribution: p=g+ ( 1−g−γ ) * ( 1−e− ( 1aσv ) β ) ( 7 ) in which p is the proportion of correct responses , g is the guess rate , γ is the lapse rate , σv is the velocity standard deviation of the trial , a is the scale parameter and β is the shape parameter . Because a Weibull distribution requires β > 0 , we used 1σv as the stimulus because this co-varies positively with performance . We fit the parameters of the psychometric function to the data of each subject and session separately , allowing the scale , shape , and lapse rate to change across sessions and subjects . Fitting was performed using a maximum likelihood approach by computing the probability of each response given the parameter values and finding the parameter values that maximized this probability . Furthermore , γ was constrained between 0 and 0 . 2 and g was fixed to 0 . 5 in the fitting procedure . To measure the effect of distance and depth cues we compared the fitted psychometric curves by inverting the Weibull function to identify the velocity standard deviation σv value that would yield a particular correct response probability . σv=1a*log ( γ+g−1γ+p−1 ) ) 1β ( 8 ) For our comparisons , we used the 0 . 75 proportion correct as criterion level of performance . These values were submitted to a within-subject analysis of variance ( ANOVA ) to assess the influence of spatial extent ( three levels: σx = 2 , 3 , and 4° ) and dimensionality ( two levels: 2D and 3D ) . Vul et al [6] described and used a Bayesian tracking solution for multiple object tracking in 2D . Here we used and expanded this modeling approach to account for object tracking in 3D , in which depth information is added to the model and used to resolve uncertain data associations . In the model , we assume the observer represents the objects by their position and velocity in 3D , that is a position and velocity state for x , y and z ( i . e . , depth ) dimensions ( see Fig 1 ) . We used meters and meters per frame for the position and velocity units , respectively . Given the linear Gaussian dynamics of the objects and noisy observations , we estimated the state of each object using a Kalman filter . This is an approximation since the Kalman filter is a suboptimal estimator when the noise in the measurements is state dependent ( see below ) . However , the difference between the distributions is small and this approximation allows us to maintain analytical tractability . The Kalman filter incorporates two sources of noise , process noise , which is part of the object dynamics , and measurement noise , which arises in the observer during observation of the stimuli . The variance of the process noise is given by σw2 ( see Eq 5 ) . The measurement noise is specified by the sensory noise of position and velocity in each dimension . It is assumed that position noise in the frontoparallel plane ( x and y axis ) increases with eccentricity [19] . σpx=c ( 1+14|px| ) ( 9 ) σpy=c ( 1+14|py| ) ( 10 ) in which c is a free scaling parameter and px and py are the x and y position of the object in meters relative to the fixation point . The depth noise follows from stereoscopic uncertainty , which is known to modulate as a function of retinal eccentricity [20] and distance from fixation in depth [21] . We converted the scaling factors found in these studies into meters yielding: σpz=d ( 1+14py2+px2 ) ( 1+1 . 5|pz| ) ( 11 ) where d is a free scaling parameter for our stimuli and pz is the position along the depth axis ( z axis ) with zero at the fixation point , which is at the center of the screen . For modeling the velocity noise in the frontoparallel plane ( x and y axis ) , we used Weber scaling [22] . Finally , the model takes the noise in the stereomotion signals into account . Based on Cumming [23] we assume a linear relationship between stereoacuity and stereomotion thresholds , with a slope of about 1 . 66 . As a result , the standard deviation of velocity noise in the depth direction was taken as . Given the above measurement equations and the dynamics described in the stimuli section , we used the Kalman filter to estimate the state of a single object ( see S1 Text ) . Because multiple objects must be tracked , there is an additional complexity for the model , i . e . which measurement to use to update the state of which object ? The exact Bayesian solution to this problem is to estimate the state of each object given every measurement and then to sum the state estimates based on how likely this assignment is . This is computationally demanding given that the six objects in our task yield 720 possible permutations of assignments at each time step . In the model , this is resolved by selecting the assignments based on their probability [6] . Using the Kalman filter approach , the probability that a perceptual measurement originated from a particular object can be computed in closed form , which indicates how likely each permutation of assignments is . The model selects the three assignments with the highest probability and computes the state estimate based on them [24 , 25] . See S1 Text for full description of the tracking algorithm . The model uses three data assignment vectors at each time step , following previous sample based models [26–28] . The model simulated 1000 trials for each of the conditions subjects underwent . Each trial consisted of three main phases . First , the model was provided stationary objects to initialize the state estimates without velocity information . Secondly , the model tracked the moving objects for the same duration as the human observers using noisy perceptual measurements of the true states . Finally , we drew a sample from the final state of one object ( the probe ) , and corrupted it with additive measurement noise according to the above equations and computed the probability of this belonging to the estimates of each object . The model responded the probe was from a target if the sum of the target probabilities exceeded that of the non-targets . A schematic illustration of the model can be seen in Fig 2 . The model tracks the objects based on the perceptual signals it has available . Because there is no consensus in how velocity information is incorporated into object tracking [6 , 29 , 30] , we considered four variants of our model . First , we tested the full extrapolation model ( FE model ) . This is the most complete version of the model , as described above , predicting the objects future positions using the process dynamics that generated the object trajectories ( see Stimuli ) . Hence , this model represents an observer who knows , according to the dynamics , how the objects move , thereby combining extrapolation and noisy perceptual measurements . Second , we considered two models without the velocity-based extrapolation step ( NE models ) . These versions of the model represent an observer who perceives velocity and uses it to dissociate assignments , but does not use the velocity information for extrapolation . We implemented this as follows . In one version , λ was set to 0 . 9 in Eqs 5 and 6 and 0 in Eq 2 , thereby representing an observer with correct knowledge of the object dynamics but without extrapolation ( NE-cd ) . In the other version , we set λ to 0 in Eqs 2 , 5 and 6 , which means that the observer does not extrapolate , as for the NE-cd model , but also uses incorrect dynamics for tracking ( NE-id ) . Finally , we considered a version of the model in which tracking is based on position measurements only , without involving velocity information in any way ( No Velocity ( NV ) model ) . In this version of the model , we removed the velocity states from the Kalman filter , thereby modeling an observer who did not use velocity at any point in the tracking process . For illustration purposes , Fig 3A shows a simulation of the NE-cd model in a simplified tracking task with two objects . As shown , during the tracking , the model initially tracks the position of each object quite accurately but at about 3 . 7 s , a point of confusion occurs and the model swaps the two objects in the further tracking . Fig 3B , which shows the likelihood of the measurements arising from each object at the point of confusion , suggests that measurements in this case are more likely to come from the other object . Depth information may improve tracking by making this association problem easier , as illustrated in Fig 3B . As shown , a measurement that would incorrectly be assigned in one dimension , may be correctly assigned using the information from the additional dimension . In other words , the additional dimension helps to correctly infer which object generated the measurement , thus disambiguating the assignment . Of note , this disambiguation not only depends on the dimensionality of the task but also how well the future positions of the objects can be predicted . Fig 3C illustrates the predictions of the FE and NE-cd model . In contrast to the FE model , in the NE-cd model current velocity does not influence the position and velocity estimate at the next time step . Not using velocity information causes a bias towards zero velocity at the next step . A bias towards zero position is also seen due to the spring dynamics used . Accordingly , it is more difficult to accurately predict the motion of the objects and therefore assign the perceptual measurements correctly . In the model , parameters c and d are free scaling parameters . We fit these parameters to the pooled group data using a maximum likelihood approach . The fit procedure was performed by finding the values that maximize the likelihood of the data given our model . As the data takes the form of a discrete number of correct answers for each of the 60 conditions , we computed the log likelihood of our data given the model as logL ( {ci}|model ) =∑i=160log ( B ( ci;Ni , pi ) ) ( 15 ) where B is a binomial distribution evaluated for each condition i with the number of correct responses ci , number of trials Ni and the proportion correct of the model pi as the probability .
The left panels of Fig 4 show the results of a typical subject when objects were tracked in either 2D ( in blue ) or 3D ( in red ) . Data points indicate the percentage of correct responses as a function of the velocity standard deviation ( σv ) , for the three values of the position standard deviation ( σx ) . Note the reversed velocity axis ( abscissa ) –the origin is on the right of the x-axis . When objects move at the highest average speed ( σv = 0 . 2°/frame ) , the subject reports at chance level ( 50% correct ) , while for lower speeds tested ( σv < 0 . 03°/frame ) performance is nearly perfect , irrespective of the position variance . We fitted psychometric curves through these data ( see Methods , Eq 7 ) . As a performance threshold we took the velocity standard deviation at which the subject responds in 75% of the trials with a correct answer . As shown , performance thresholds are higher when objects move in 3D than in 2D ( red curve are leftward shifted relative to the blue curves ) and are also increased in the sessions with higher position variance . This suggests that this subject could track objects at a higher speed when the mean distance between the objects increased and when depth information was added . The results of this subject are exemplary for all subjects . Their average data and fitted curves are shown in the right panels of Fig 4 . The 2D results are consistent with the observations of Vul et al [6] , tracking accuracy declines as speed increases but increases with distance between objects . The 3D results show that adding depth information improves tracking performance . Threshold values were extracted based on the individual fits , then averaged , and plotted in Fig 5 as a function of position standard deviation ( black dashed and solid lines ) . A within-subject ANOVA with position standard deviation ( three levels: σx = 2 , 3 and 4° ) and dimensionality ( two levels: 2D and 3D ) as factors revealed not only significant main effects of position standard deviation ( F ( 2 , 16 ) = 78 . 52 , p < . 001 ) and dimensionality ( F ( 1 , 8 ) = 151 . 07 , p < . 001 ) , but also a significant interaction ( F ( 2 , 16 ) = 5 . 20 , p = . 018 ) . Posthoc testing showed the difference between 2D and 3D tracking is significant for all three σx values ( paired t-tests , p<0 . 01 ) . Thus tracking performance is better when objects are further apart not only in 2D but also 3D , with the depth interacting to produce a more than additive effect on performance . In order to account for the data , we specified four versions of the optimal observer model for object tracking in 2D and 3D . The model versions differ as to how the velocity information is taken into account by the observer . More specifically , the FE model tracks objects optimally by combining extrapolation with noisy position and velocity measurements . The NE models obtain noisy perceptual measurements of velocity information without using this information for extrapolation and the NV model does not take velocity information into account at all . The colored lines in Fig 6 present the predictions of the four model versions together with the subject data . The FE-model does not capture the data well , while the NE-id , NE-cd and NV-models perform reasonably well . This can also be seen in the predicted velocity standard deviation thresholds shown in Fig 5 , where the FE model underestimates some thresholds while overestimating others in 3D and underestimates them in 2D . To perform a quantitative comparison of the models , we computed the relative log likelihood of each model ( compared to most likely model ) given our data and the best fit parameters ( see Table 2 ) . The relative log-likelihoods show evidence in favor of the NE-cd model . Note , as all models include the same number of parameters , corrections such as AIC or BIC are not required for model comparison [31] . It should be noted that for computational reasons these fits were obtained through a rough grid search and as such slightly better fits may be obtainable . We verified for each model that the likelihood function had a concave shape for the grids used . Therefore , the minima of each should be a reasonable representation of the parameters .
We show that objects that move in 3D are tracked better than objects moving in 2D and that the magnitude of this improvement increases with the mean distance between objects in 3D space . We compared four ideal observer models to aid in providing a quantitative explanation behind these results . We find an ideal observer model that tracks objects optimally ( FE-model ) by extrapolating the next position and combining this with noisy perceptual measurements cannot account for our behavioral data . Instead models that assume subjects use velocity information suboptimally ( NE-cd , NE-id , and NV models ) provide a better fit to the data . Specifically , we find that our No Extrapolation with correct dynamics ( NE-cd ) model , which receives velocity measurements from the objects but cannot use velocity to predict the next position , provides the best fit . The NE-cd provides an intuitive explanation for the benefit of depth information . When we track multiple objects we must infer which objects generated which noisy perceptual measurements to both identify the targets and to generate accurate predictions . As we mentioned previously , inferring the correct data assignments is easier when we have the additional depth dimension because assignments that can be confused in 2D are likely to be disambiguated in 3D . The interaction between depth and distance can be explained in a similar manner . Although having an additional dimension provides the capability to disambiguate which objects generated which measurements , the distance between the objects in this additional dimension is crucial . If objects are close together in depth then perceptual noise could still cause objects to be confused . As we increase the distance we reduce the overlap between the predictions of one object and the measurements of another making it more difficult to confuse them . As such , our model can explain our finding that increasing speed lowers tracking performance because it increases our uncertainty [6] and depth improves tracking by also making it harder for predictions and measurements to overlap . In addition , our model also allows us to explain why objects placed in depth planes are tracked more accurately [15 , 32] . Placing objects in different planes disambiguates object to measurements assignments when they are close together in 2D thereby reducing the number of incorrect assignments . Additionally , if our model is a realistic approximation to the task then the noise parameters obtained after fitting should be consistent with other work . Indeed , the frontoparallel noise scaling ( c ) and depth noise scaling ( d ) of the best-fit model are similar to those previously reported . Bayes & Husain [33] found the precision of positional short term working memory for 3 items to be approximately 0 . 5 deg⁻¹ with the items being shown 10 deg to the left of fixation . Using Eq ( 9 ) and converting to their units produces an estimated precision from our model of 0 . 7 deg⁻¹ . For eccentricity scaling of depth noise our model predicts a standard deviation of 0 . 0202 m at fixation and 0 . 0981 m at 9 deg , similar to previously reported values which were between 0 . 0087–0 . 0195 m in the fovea and between 0 . 0479–0 . 1831 m at 9 deg [20] . Although these tasks are different from ours they do illustrate the values obtained are plausible and within the range of previous data providing some additional support of our model . Despite the NE-cd model successfully explaining our experimental observations there are still components of the tracking process that need further investigation . Firstly , the noise terms we use in our model are simplifications . Investigation into how realistic these simplifications are is needed . To illustrate this , the current model cannot explain the finding that tracking objects in two different planes is harder when the planes are separated by large distances compared to small distances [15] . One explanation is that the noise in our estimates of object position in the frontoparallel plane is affected by distance from fixation , a component that was not introduced into our model . However , it could also be that the additional distance alters the size and contrast of the retinal image thereby changing the perceptual uncertainty while maintaining the independence of frontoparallel noise and depth . Therefore , research is needed to investigate how distance from fixation affects tracking in a virtual rather than real 3D set up where these properties can be tightly manipulated . Secondly , we only compare four possible models for velocity usage , one of which predicts the next position of the object and combines this with noisy measurement ( FE ) , two of which perceive velocity information but do not use it for prediction ( NE models ) and one of which uses only position information for all the tracking ( NV ) . There are additional possibilities for how observers could use velocity information . It is possible that observers do extrapolate but that there is a difference between the true motion of the objects in experiments and the model used by subjects , an idea which has also been presented to explain findings in visual working memory [34] . Alternatively , individuals may not build models of object motion in tracking tasks , but instead make predictions only using perceived velocity and Newtonian dynamics [30] . This multitude of possibilities makes it difficult to draw too strong conclusions about the role of velocity information in MOT . However , as the perfect extrapolation model produced the worst fit it is evident that some form of under extrapolation is present . This is consistent with experiments showing that when objects are being tracked and become occluded , accuracy is higher if they reappear where they disappeared rather than at their extrapolated position [35 , 36] . An attractive way to investigate which sensory noise model and velocity model underlies our tracking ability would be to use factorial model comparison [37 , 38] . Essentially , this uses Bayesian model comparison [39 , 40] to compare sets of models . This could be used to investigate different noise models and different ways velocity is incorporated to identify which pairing best fits human tracking data . Unfortunately , modeling MOT is difficult as the task is inherently computationally intensive and model comparison requires thousands of iterations per model to integrate over the parameter space . As such it may be appealing to consider simpler tasks that still capture the elements of MOT to facilitate modeling attempts of the underlying processes . For example , Ma & Huang [9] modeled multiple trajectory tracking , a task in which observers see multiple dots moving left to right and have to report whether they deviated at the mid-point . This simple task embodies some elements of MOT such as the influence of sensory noise and solving the correspondence problem . It can be formulated in an analytical way to allow for efficient model comparison . However , this task may not be ideal to study the role of velocity information , as it does not require a large focus on extrapolation . A similar experiment that requires more positional extrapolation may prove useful to determine different noise models and how velocity is used . Additionally , in this experiment each object is relevant to the task , in contrast MOT tasks typically incorporate distracters , which may affect the tracking process . Furthermore , we made the assumption that observers track both non-targets and targets identically . Other models have been proposed that exclusively track targets [30] , however , there is experimental evidence that both targets and non-targets are tracked . That is , if subjects perform an MOT task and are asked to report when a probe is presented on a target or non-target they detect the probe more often on a target , but still detect it on non-targets as well [41] . This suggests observers track both targets and non-targets but not in identical ways . An additional extension to our model would be to consider modifications that allow tracking to differ between targets and non-targets while maintaining its current explanatory power . Our model also has implication for future work in MOT . Specifically , it makes predictions about which factors should influence the difficulty of the assignment problem and therefore which factors should affect tracking performance . For example , our model predicts that the amount of facilitation that 3D motion provides is dependent on the precision of the depth information . If the precision of our depth estimate is low then the improvement should also be low and vice versa . This has been tested somewhat indirectly , as precise depth cues such as disparity alone can improve tracking performance [32] but less precise depth cues such as relative size do not [16] . We do not know of any experiments directly testing if gradual manipulations in depth cue reliability produce the expected effect . Our model also has implications for 2D MOT . Theories have proposed that tracking performance is limited only by the distance between objects , and not to the number of objects or speed [42] . Our model suggests it is not distance alone but the relationship between distance and measurement precision . This yields the experimental prediction that objects can be close together and are still trackable if measurement precision is high but creating poorer precision should require moving objects to be further apart to produce the same performance . To our knowledge there is no work testing the role of measurement precision on tracking in either 2D or 3D MOT . Doing so would greatly improve our knowledge of the role uncertainty plays in our capability to track multiple objects . Due to the generality of the correspondence problem in visual perception and cognition , the finding that depth cues reduce correspondence errors has implications for other topics . For example , a significant source of errors within working memory experiments are so called “swap errors” [8] . These refer to errors in which an observer recalls not the item probed but another memorized item . It has been shown the number of these errors increases as objects are brought closer together [8] . This suggests that these errors result from making an incorrect correspondence between the location probed and the existing memory representation . Depth cues could play a role in reducing the occurrence of this type of errors within working memory . A recent change detection experiment provided some support for the idea that depth cues reduce swap errors [43] . In this experiment , subjects had to memorize a display of colored items whose position was either 2D or 3D . Subsequently , they were shown a second display where the colors could change and had to indicate if the display had changed . Results indicated subjects were more accurate at detecting a change when the items were presented in 3D than 2D . One reason for this improvement could be a reduction in swap errors when making the comparison between the two displays . This could be tested more directly by estimating the proportion of swap errors when items are presented either on a single plane or multiple depth planes . If a reduction in swap errors occurs , this would suggest that depth information is a crucial component in solving multiple forms of visual correspondence .
|
Many daily life situations require us to track objects that are in motion . In the laboratory , this multiple object tracking problem is classically studied with objects moving on a two-dimensional screen , but in the real world objects typically move in three dimensions . Here we show that , despite the complexity of seeing in depth , observers track multiple objects better when they move in 3D than 2D . A probabilistic inference model explains this by showing that the association of noisy visual signals to the objects that caused them is less ambiguous when depth cues are available . This highlights the role that depth cues play in our everyday ability to track objects .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"velocity",
"classical",
"mechanics",
"applied",
"mathematics",
"social",
"sciences",
"neuroscience",
"learning",
"and",
"memory",
"simulation",
"and",
"modeling",
"algorithms",
"cognitive",
"neuroscience",
"mathematics",
"cognition",
"memory",
"extrapolation",
"vision",
"research",
"and",
"analysis",
"methods",
"curve",
"fitting",
"mathematical",
"functions",
"kalman",
"filter",
"mathematical",
"and",
"statistical",
"techniques",
"physics",
"working",
"memory",
"psychometrics",
"psychology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"sensory",
"perception",
"cognitive",
"science",
"numerical",
"analysis",
"motion"
] |
2017
|
Effect of depth information on multiple-object tracking in three dimensions: A probabilistic perspective
|
Prions , the agents causing transmissible spongiform encephalopathies , colonize the brain of hosts after oral , parenteral , intralingual , or even transdermal uptake . However , prions are not generally considered to be airborne . Here we report that inbred and crossbred wild-type mice , as well as tga20 transgenic mice overexpressing PrPC , efficiently develop scrapie upon exposure to aerosolized prions . NSE-PrP transgenic mice , which express PrPC selectively in neurons , were also susceptible to airborne prions . Aerogenic infection occurred also in mice lacking B- and T-lymphocytes , NK-cells , follicular dendritic cells or complement components . Brains of diseased mice contained PrPSc and transmitted scrapie when inoculated into further mice . We conclude that aerogenic exposure to prions is very efficacious and can lead to direct invasion of neural pathways without an obligatory replicative phase in lymphoid organs . This previously unappreciated risk for airborne prion transmission may warrant re-thinking on prion biosafety guidelines in research and diagnostic laboratories .
Transmissible spongiform encephalopathies ( TSEs ) are fatal neurodegenerative disorders that affect humans and various mammals including cattle , sheep , deer , and elk . TSEs are characterized by the conversion of the cellular prion protein ( PrPC ) into a misfolded isoform termed PrPSc [1] . PrPSc aggregation is associated with gliosis , spongiosis , and neurodegeneration [2] which invariably leads to death . Prion diseases have been long known to be transmissible [3] , and prion transmission can occur after oral , corneal , intraperitoneal ( i . p . ) , intravenous ( i . v . ) , intranasal ( i . n . ) , intramuscular ( i . m . ) , intralingual , transdermal and intracerebral ( i . c . ) application , the most efficient being i . c . inoculation [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] . Several biological fluids and excreta ( e . g . saliva , milk , urine , blood , placenta , feces ) contain significant levels of prion infectivity [13] , [14] , [15] , [16] , [17] , and horizontal transmission is believed to be critical for the natural spread of TSEs [18] , [19] , [20] , [21] , [22] , [23] . Free-ranging animals may absorb infectious prion particles through feeding or drinking [24] , [25] , and tongue wounds may represent entry sites for prions [26] . PrPSc has also been found in the olfactory epithelium of sCJD patients [27] , [28] . Prion colonization of the nasal epithelium occurs in various species and with various prion strains [11] , [12] , [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] . In the HY-TME prion model , intranasal application is 10–100 times more efficient than oral uptake [29] and , as in many other experimental paradigms [38] , [39] , [40] , [41] , [42] , [43] , [44] , the lymphoreticular system ( LRS ) is the earliest site of PrPSc deposition . A publication demonstrated transmission of chronic wasting disease ( CWD ) in cervidized mice via aerosols and upon intranasal inoculation [45] , yet two studies reported diametrically differing results on the role of the olfactory epithelium or the LRS in prion pathogenesis upon intranasal prion inoculation [11] , [12] , perhaps because of the different prion strains and animal models used . These controversies indicate that the mechanisms of intranasal and aerosolic prion infection are not fully understood . Furthermore , intranasal administration is physically very different from aerial prion transmission , as the airway penetration of prion-laden droplets may be radically different in these two modes of administration . Here we tested the cellular and molecular characteristics of prion propagation after aerosol exposure and after intranasal instillation . We found both inoculation routes to be largely independent of the immune system , even though we used a strongly lymphotropic prion strain . Aerosols proved to be efficient vectors of prion transmission in mice , with transmissibility being mostly determined by the exposure period , the expression level of PrPC , and the prion titer .
Prion aerosols were produced by a nebulizing device with brain homogenates at concentrations of 0 . 1–20% ( henceforth always indicating weight/volume percentages ) derived from terminally scrapie-sick or healthy mice , and immitted into an inhalation chamber . As per the manufacturer's specifications , aerosolized particles had a maximal diameter of <10 µm , and approximately 60% were <2 . 5 µm [46] . Groups of mice overexpressing PrPC ( tga20; n = 4–7 ) were exposed to prion aerosols derived from infectious or healthy brain homogenates ( henceforth IBH and HBH ) at various concentrations ( 0 . 1 , 2 . 5 , 5 , 10 and 20% ) for 10 min ( Fig . 1A , Table 1 ) . All tga20 mice exposed to aerosols derived from IBH ( concentration: ≥2 . 5% ) succumbed to scrapie with an attack rate of 100% . The incubation time negatively correlated with the IBH concentration ( 2 . 5%: n = 4 , 165±54 dpi; 5%: n = 4 , 131±7 dpi; 10%: n = 5 , 161±27 dpi; 20%: n = 6 , 133±8 dpi; p = 0 . 062 , standard linear regression on standard ANOVA; Fig . 1A and F , Table 1 , Table S1A ) . tga20 mice exposed to aerosolized 0 . 1% IBH did not develop clinical scrapie within the observational period ( n = 4; experiment terminated after 300 dpi ) , yet displayed brain PrPSc indicative of subclinical prion infection ( Fig . 1A and 2A ) . In contrast , control tga20 mice ( n = 4 ) exposed to aerosolized HBH did not develop any recognizable disease even when kept for ≥300 dpi , and their brains did not exhibit any PrPSc in histoblots and Western blots ( data not shown ) . In the above experiments , and in all experiments described in the remainder of this study , all PrP-expressing ( tga20 and WT ) mice diagnosed as terminally scrapie-sick were tested by Western blot analysis and by histology: all were invariably found to contain PrPSc in their brains ( Fig . 2 ) and to display all typical histopathological features of scrapie including spongiosis , PrP deposition and astrogliosis ( Fig . 1H ) . We then sought to determine the minimal exposure time that would allow prion transmission via aerosols ( Fig . 1B , Table 1 ) . tga20 mice were exposed to aerosolized IBH ( 20% ) for various durations ( 1 , 5 or 10 min ) in two independent experiments . Surprisingly , an exposure time of only 1 min was found to be sufficient to induce a 100% scrapie attack rate . Longer exposures to prion-containing aerosols strongly correlated with shortened incubation periods ( Fig . 1B and G , Table 1 , Table S1A and B ) . In order to test the universality of the above results , we examined whether aerosols can transmit prions to various mouse strains ( CD1 , C57BL/6; 129SvxC57BL/6 ) expressing wild-type ( wt ) levels of PrPC . CD1 mice were exposed to aerosolized 20% IBH in two independent experiments ( Fig . 1C , Table 1 ) . After 5 or 10-min exposures , all CD1 mice succumbed to scrapie whereas shorter exposure ( 1 min ) led to attack rates of 0–50% [1 min exposure ( first experiment ) : scrapie in 0/3 mice; 1 min exposure ( second experiment ) : 2/4 mice died of scrapie at 202±0 dpi; 5 min ( first experiment ) : n = 4 , attack rate 100% , 202±12 dpi; 5 min ( second experiment ) : n = 3 , attack rate 100% , 202±0 dpi; 10 min ( first experiment ) : n = 4 , attack rate 100% , 202±0 dpi; 10 min ( second experiment ) : n = 4 , attack rate 100% , 206±16 dpi] . In CD1 mice exposed to prion-containing aerosols for longer intervals , we detected a trend towards shortened incubation times which did not attain statistical significance ( Table S1A and S1B ) . We also investigated whether C57BL/6 or 129SvxC57BL/6 mice would succumb to scrapie upon exposure to prion aerosols ( Fig . 1D and E , Table 1 ) . A 10 min exposure time with a 20% IBH led to an attack rate of 100% ( C57BL/6: 10 min: n = 4; 185±11 dpi; 129SvxC57BL/6: n = 5; 10 min: 182±15 dpi ) . Control Prnpo/o mice ( 129SvxC57BL/6 background; n = 3 ) were resistant to aerosolized prions ( 20% , 10 min ) as expected ( Fig . 1E and H , Table 1 ) . When tga20 mice were challenged for 10 min , variations in the concentration of aerosolized IBH had a barely significant influence on survival times ( p = 0 . 062; Fig . 1F ) , whereas variations in the duration of exposure of tga20 mice affected their life expectancy significantly ( p<0 . 001; Fig . 1G ) . Furthermore tga20 mice , which express 6–9 fold more PrPC in the central nervous system ( CNS ) than wt mice [46] , [47] , [48] , succumbed significantly earlier to scrapie upon prion aerosol exposure for 10 min ( 20% ) ( tga20 mice: 134±4 dpi; CD1 mice: 202±12 dpi , p<0 . 0001; C57BL/6 mice: 185±11 dpi , p = 0 . 003; 129SvxC57BL/6 mice: 182±15 dpi , p = 0 . 003; Fig . 1B–E , Fig . S1 , Table S1A and S1C ) . Incubation time was prolonged and transmission was less efficient in CD1 mice than in tga20 mice after a 1 min exposure to prion aerosols ( 20% ) . The variability of incubation times between CD1 mice was low ( 1st vs . 2nd experiment with 5-min exposure: p = 0 . 62 , 1st vs . 2nd experiment with 10-min exposure: p = 0 . 27; Fig . 1C , Table 1 ) . This suggests that 1 min exposure of CD1 mice to prion aerosols ( 20% ) suffices for uptake of ≤1LD50 infectious units . This finding underscores the importance of PrPC expression levels not only for the incubation time but also for susceptibility to infection and neuroinvasion upon exposure to aerosols . Histoblot analyses confirmed deposition of PrPSc in brains of tga20 mice exposed to prion aerosols derived from 10% or 20% IBH , whereas no PrPSc was found in brains of Prnpo/o mice exposed to prion aerosols ( Fig . 2D ) . We then performed a semiquantitative analysis of the histopathological lesions in the CNS . The following brain regions were evaluated according to a standardized severity score ( astrogliosis , spongiform change and PrPSc deposition; [49] ) : hippocampus , cerebellum , olfactory bulb , frontal white matter , and temporal white matter . Scores were compared to those of mice inoculated i . c . with RML ( Fig . 2E and F ) . Lesion profiles of terminally scrapie-sick mice ( tga20 , CD1 , C57BL/6 and 129SvxC57BL/6 ) infected i . c . or through aerosols were similar irrespectively of genetic background or PrPC expression levels ( Fig . 2E and F ) , with CD1 and 129SvxC57BL/6 hippocampi and cerebella displaying only mild histological and immunohistochemical features of scrapie regardless of the route of inoculation . We attempted to trace PrPSc in the nasopharynx , the nasal cavity or various brain regions early after prion aerosol infection ( 1–6 hrs post exposure ) and at various time points after intranasal inoculations ( 6 , 12 , 24 , 72 , 144 hrs , 140 dpi , and terminally ) with various methods including Western blot , histoblot and protein misfolding analyses . However , none of these analyses detected PrPSc shortly after exposure to prion aerosols ( 6–72 hrs post prion aerosol exposure ) whereas at 140 dpi or terminal stage PrPSc was detected by all of these methods ( Fig . S2; data not shown ) . We then investigated whether PrPC expression in neurons would suffice to induce scrapie after exposure to prions through aerosols . NSE-PrP transgenic mice selectively express PrPC in neurons and if bred on a Prnpo/o background ( Prnpo/o/NSE-PrP ) display CNS-restricted PrP expression levels similar to wt mice [50] . Prnpo/o/NSE-PrP ( henceforth referred to as NSE-PrP ) mice were exposed to prion aerosols ( 20% homogenate; 10 min ) . All NSE-PrP mice succumbed to terminal scrapie ( 216±8 dpi; n = 4; Fig . 1E , 2G , Table 1 ) , although incubation times were significantly longer than those of wt 129SvxC57BL/6 mice ( 180±15 dpi; n = 5; p = 0 . 004 ) . Histology and immunohistochemistry confirmed scrapie in NSE-PrP brains ( Fig . 2H and I ) . Histopathological lesion severity score analysis ( see above ) revealed a lesion profile roughly similar to that of control 129SvxC57BL/6 mice ( Fig . 2E , H ) . More severe lesions were observed in NSE-PrP cerebella whereas olfactory bulbs were less affected . Real time PCR analysis revealed 2–4 transgene copies per Prnp allele in Prnpo/o/NSE-PrP mice . A detailed quantitative analysis of PrPC expression levels at various sites of the CNS was performed by comparing the signals obtained by blotting various amounts of protein from NSE-PrP , wt and tga20 tissues ( Fig . S3 ) . A value of 100 was arbitrarily assigned to expression of PrPC in wt tissues; olfactory epithelia of tga20 and NSE-PrP mice expressed ≥350 and ∼30 , respectively ( Fig . S3A ) . In olfactory bulbs , tga20 and NSE-PrP mice expressed ≥150 and 30 , respectively ( Fig . S3B ) . In brain hemispheres tga20 and NSE-PrP mice expressed >250 and >150 , respectively ( Fig . S3C ) . Therefore , NSE-PrP mice expressed somewhat more PrPC than wt mice in brain hemispheres , but somewhat less in olfactory bulbs and olfactory epithelia . In many paradigms of extracerebral prion infection , efficient neuroinvasion relies on the anatomical and physiological integrity of several immune system components [40] , [42] , [43] , [44] . To determine whether this is true for aerosolic prion challenges , we exposed immunodeficient mouse strains to prion aerosols . This series of experiments included JH−/− mice , which selectively lack B-cells , and γCRag2−/− mice which are devoid of mature B- , T- and NK-cells ( Fig . 3A ) . Upon exposure to prion aerosoIs ( 20% IBH; exposure time 10 min ) both JH−/− and γCRag2−/− mice succumbed to scrapie with a 100% attack rate ( JH−/−: n = 6 , 181±21 dpi; γCRag2−/−: n = 11 , 185±41 dpi , p = 0 . 65 ) . The incubation times were not significantly different to those of C57BL/6 wt mice exposed to prion aerosols ( JH−/− mice: p = 0 . 9; γCRag2−/− mice: p = 0 . 7 ) . Histological and immunohistochemical analyses confirmed scrapie in all clinically diagnosed mice . Lesion severity score analyses ( Fig . 3A and 3E ) showed that JH−/− and γCRag2−/− mice had lower profile scores in cerebella and higher scores in hippocampi and frontal white matter than C57BL/6 mice . Slightly higher scores in temporal white mater areas and the thalamus could be detected in JH−/− and γCRag2/−/− mice , whereas γCRag2−/− mice showed lower scores in olfactory bulbs . Consistently with several previous reports , γCRag2−/− mice ( n = 4 ) did not succumb to scrapie after i . p . prion inoculation ( 100µl RML6 0 . 1% 6 log LD50 ) even when exposed to a prion titer that was twice higher than that used for intranasal inoculations ( data not shown ) . Depending on the exposure time and the IBH concentration , tga20 mice developed splenic PrPSc deposits . In contrast , none of the scrapie-sick JH−/− , LTβR−/− and γCRag2−/− mice displayed any splenic PrPSc on Western blots and/or histoblots ( Fig . S4A–D ) despite copious brain PrPSc . Follicular dendritic cells ( FDCs ) are essential for prion replication within secondary lymphoid organs and for neuroinvasion after i . p . or oral prion challenge [42] , [44] , [51] . Lymphotoxin beta receptor-Ig fusion protein ( LTβR-Ig ) treatment in C57BL/6 mice causes dedifferentiation of mature FDCs , resulting in reduced peripheral prion replication and neuroinvasion upon extraneural ( e . g . intraperitoneal or oral ) prion inoculation [52] , [53] . We therefore investigated whether FDCs are required for prion replication after challenge with prion aerosols . C57BL/6 mice were treated with LTβR-Ig or nonspecific pooled murine IgG ( muIgG ) before and after prion challenge ( −7 , 0 , and +7 days ) ( Fig . 3B ) . The effects of the LTβR-Ig treatment were monitored by Mfg-E8+/FDC-M1+ staining for networks of mature FDCs in lymphoid tissue . This analysis revealed a complete loss of Mfg-E8+/FDC-M1+ networks at the day of prion exposure and at 14 dpi ( data not shown ) . LTβR-Ig treatment and dedifferentiation of FDCs did not alter incubation times upon aerosol prion infection ( LTβR-Ig: n = 3 , attack rate 100% , 184±0 dpi; muIgG: n = 3 , attack rate 100% , 184±0 dpi ) ( Fig . 3B , Table 2 ) . The diagnosis of terminal scrapie was confirmed by histological and immunohistochemical analyses in all clinically affected mice ( Fig . 3B; data not shown ) . Histopathological lesion severity scoring revealed that LTβR-Ig treated C57BL/6 mice displayed a higher score in all regions investigated than untreated C57BL/6 mice upon challenge with prion aerosols ( 20% IBH; 10 min ) ( Fig . 2E and 3B ) . We found slightly less severe scores in the olfactory bulbs of C57BL/6 mice treated with muIgG than in untreated C57BL/6 mice upon challenge with prion aerosols ( Fig . 2E and 3B ) , and a slightly higher score in the temporal white matter ( exposure to 20% aerosol for 10 min; Fig . 2E and 3B ) . LTβR signaling is essential for proper development of secondary lymphoid organs and for maintenance of lymphoid microarchitecture , and was recently shown to play an important role in prion replication within ectopic lymphoid follicles and granulomas [40] , [41] , [44] . To investigate the role of this pathway in aerogenic prion infections , LTβR−/− mice were exposed to prion aerosols ( 20% IBH; 10 min exposure time ) . All LTβR−/− mice succumbed to scrapie ( LTβR−/−: n = 4 , 272±0 dpi ) and displayed PrP deposits in their brains ( Fig . 3C ) . Histological severity scoring of aerosol-exposed mice revealed higher scores in LTβR−/− hippocampi and lower scores in cerebellum , olfactory bulb , frontal and temporal white matter than in C57BL/6 controls ( exposure: 20%; 10 min; Fig . 2E and 3C ) . We then investigated the role of CD40 receptor in prion aerosol infection . CD40−/− mice fail to develop germinal centers and memory B-cell responses , yet CD40L−/− mice show unaltered incubation times upon i . p . prion challenge [54] . Similarly to the other immunocompromised mouse models investigated , CD40−/− mice developed terminal scrapie upon infection with prion aerosols with an attack rate of 100% ( n = 3 , 276±50 dpi ) . Lesion severity analyses of CD40−/− mice revealed a slightly higher score in the cerebellum and the temporal white matter than in C57BL/6 mice ( Fig . 2E and 3C ) . Therefore , LTβR and CD40 signaling are dispensable for aerosolic prion infection . Certain components of the complement system ( e . g . C3; C1qa ) play an important role in early prion uptake , peripheral prion replication and neuroinvasion after peripheral prion challenge [43] , [55] , [56] . We have tested whether this is true also for exposure to prion aerosols . Mice lacking both complement components C3 and C4 ( C3C4−/− ) were exposed for 10 min to 20% aerosolized IBH . All C3C4−/− mice succumbed to scrapie ( n = 3 , 382±33 dpi; Fig . 4A ) . Histopathological evaluation of all scrapie affected mice revealed astrogliosis , spongiform changes and PrP-deposition in the CNS ( Fig . 4A ) . The data reported above argued in favor of direct neuroinvasion via PrPC-expressing neurons upon aerosol administration . However , a possible alternative mechanism of transmission may be via the ocular route , namely via cornea , retina , and optic nerve [57] , [58] . In order to test this possibility , newborn ( <24 hours-old ) tga20 and CD1 mice , whose eyelids were still closed , were exposed for 10 min to prion aerosols generated from a 20% IBH . All mice succumbed to scrapie and showed PrP deposits in brains ( tga20 mice: n = 3 , 173±23 dpi; CD1 mice: n = 3 , 211±0 dpi ) ( Fig . 4B ) . Newborn tga20 mice succumbed to scrapie slightly later ( p = 0 . 0043 ) than adult tga20 mice , whereas no differences were observed between newborn and adult CD1 mice exposed for 10 min to prion aerosols generated from a 20% IBH ( p = 0 . 392 ) . The brains of all animals contained PK-resistant material , as evaluated by Western blot analysis ( data not shown ) . In addition , untreated littermates or other sentinels which were reared or housed together with aerosol-treated mice immediately following exposure to aerosols showed neither signs of scrapie nor PrPSc in brains , even after 482 dpi . This suggests that prion transmission was the consequence of direct exposure of the CNS to prion aerosols rather than the result of transmission via other routes like ingestion from fur by grooming or exposure to prion-contaminated feces or urine . We further investigated additional mice for the occurrence of PrPSc in secondary lymphoid organs upon exposure to prion aerosols . PK-resistant material was searched for in spleens , bronchial lymph nodes ( bln ) and mesenteric lymph nodes ( mln ) at terminal stage of disease . C57BL/6 , 129SvxC57BL/6 , muIgG treated C57BL/6 mice , newborn tga20 mice as well as newborn CD1 mice contained PrPSc in the LRS , whereas LTβR−/− mice and C57BL/6 mice treated with LTβR-Ig lacked PrPSc deposits in spleens ( Fig . S4A–F; Table 3 ) . To dissect aerosol-mediated from non-aerosolic contributions to prion exposure , we directly applied a prion suspension ( RML 6 . 0 , 0 . 1% , 40µl , corresponding to 4×105 LD50 scrapie prions ) to the nasal mucosa of various mouse lines ( Fig . S5 ) . Since mice breathe exclusively through their nostrils [59] , [60] we reasoned that this procedure would simulate aerosolic transmission with sufficient faithfulness although the mechanisms of prion uptake could still differ between aerosolic and intranasal administration [11] . tga20 ( n = 10 ) , 129SvxC57BL/6 ( n = 5 ) , C57BL/6 ( n = 8 ) and Prnpo/o mice ( n = 8 ) were challenged intranasally with prions ( Fig . S5 ) . To test the possibility that the inoculation procedure itself might impact the life expectancy of mice , C57BL/6 mice ( n = 4 ) were inoculated intranasally with healthy brain homogenate ( HBH ) for control ( Fig . S5E ) . None of the animals that had been inoculated with HBH displayed a shortened life span , nor did they develop any clinical signs of disease - even when kept for ≥500 dpi . In contrast , after intranasal prion inoculation all C57BL/6 , 129SvxC57BL/6 and tga20 mice succumbed to scrapie with an attack rate of 100% ( Fig . S5A–C ) , whereas Prnpo/o mice were resistant to intranasal prions ( Fig . S5D ) . After intranasal inoculation , tga20 mice ( n = 10 , 160±28 dpi ) displayed a shorter incubation time ( Fig . S5C ) than 129SvxC57BL/6 ( n = 5 , 217±20 dpi ) or C57BL/6 mice ( n = 8 , 266±33 dpi; Fig . S5A and S5B ) . Further , histological and immunohistochemical analyses for spongiosis , astrogliosis and PrP deposition pattern confirmed terminal scrapie ( Fig . S5J ) . A histopathological lesion severity score analysis revealed similar lesion profiles as detected after exposure to prion aerosols ( Fig . S5K ) . However , in the olfactory bulb of tga20 and 129SvxC57BL/6 mice the score was lower upon intranasal administration than in the aerosol paradigm ( Fig . 2E ) . Finally , we tested whether prion transmission via the intranasal route would be enabled by selective PrPC expression on neurons . For that , we inoculated NSE-PrP mice . All intranasally challenged NSE-PrP mice ( n = 6 , 291±86 dpi ) succumbed to scrapie . The incubation time until terminal disease did not differ significantly from that of 129SvxC57BL/6 control mice ( n = 5 , 217±20 dpi; p = 0 . 0868 ) . Next , we sought to determine which components ( if any ) of the immune system are required for neuroinvasion upon intranasal infection with prions . To address this question , Rag1−/− and γCRag2−/− mice were intranasally inoculated with prions ( inoculum RML 6 . 0 , 0 . 1% , 40µl , equivalent to 4×105 LD50 scrapie prions ) . Remarkably , all intranasally prion-inoculated Rag1−/− ( n = 9 , 203±6 dpi ) ( Fig . 5A and H ) and γCRag2−/− mice ( n = 16 , 243±24 dpi ) ( Fig . 5D and G ) succumbed to scrapie , providing evidence for a LRS-independent mechanism of prion neuroinvasion upon intranasal administration . Incubation times in Rag1−/− were significantly different to those of intranasally challenged control mice ( C57BL/6; attack rate 100%; n = 8 , 266±33 dpi; p = 0 . 0009 ) whereas γCRag2−/− mice were not different from those of intranasally challenged control mice ( Balb/c: attack rate 100% , n = 6 , 209±48 dpi , p = 0 . 099 ) ( Fig . 5B and Fig . 5D ) . After intranasal prion administration , PrPSc was present in the CNS of Rag1−/− or γCRag2−/− mice . WB analysis corroborated terminal scrapie ( Fig . 5G and H ) . Histopathological lesion severity scoring revealed a distinct lesion profile characterized by a high score in the temporal white matter and the thalamus in case of Rag1−/− mice . In case of γCRag2−/− mice the cerebellum , the olfactory bulb and the frontal white matter displayed lower scores ( Fig . 5I and J ) . In contrast to the CNS spleens of the affected animals did not contain PK-resistant material in terminally sick Rag1−/− and γCRag2−/− mice ( Fig . S6E ) . For control , Rag1−/− as well as γCRag2−/− mice were intranasally inoculated with HBH to test the possibility that intranasal inoculation itself impacts their life expectancy . None of the mice inoculated with HBH died spontaneously or developed scrapie up to ≥300 dpi ( n = 4 each; Fig . S6A , C–D ) . Further , Balb/c mice and C57BL/6 mice ( n = 4 each ) inoculated intranasally with HBH ( Fig . S5E and S6D ) did not develop any disease for ≥300 dpi . As a positive control , Rag1−/− mice were i . c . inoculated with 3×105 LD50 scrapie prions . This led to terminal scrapie disease after approximately 130 days and an attack rate of 100% ( n = 3 , 131±8 dpi ) ( Fig . 5B and data not shown ) . As additional negative controls , Rag1−/− and γCRag2−/− mice were i . p . inoculated with prions ( 100 µl RML 0 . 1% , 1×106 LD50 ) . Although more infectious prions ( approximately 2 fold more ) were applied when compared to the intranasal route , i . p . prion inoculation did not suffice to induce scrapie in Rag1−/− and γCRag2−/− mice ( attack rate: 0% , n = 4 for each group , experiment terminated after 400 dpi ) . The complement component C1qa is involved in facilitating the binding of PrPSc to complement receptors on FDCs [56] . Accordingly , C1qa−/− mice are resistant to prion infection upon low-dose peripheral inoculation . CD21−/− mice are devoid of the complement receptor 1 , display a normal lymphoid microarchitecture and show a reduction in germinal center size . The incubation time in CD21−/− mice is greatly increased upon peripheral prion inoculation via the i . p . route [56] . To determine whether the complement system is involved in prion infection through aerosols , C1qa−/− and CD21−/− mice were intranasally inoculated with prions . C1qa−/− mice and CD21−/− mice succumbed to scrapie with an attack rate of 100% ( C1qa−/− mice: n = 4 , 288±26 dpi; CD21−/− mice: n = 10 , 235±24 dpi ) ( Figs . 6A–C ) , with CD21−/− mice succumbing to scrapie slightly earlier when compared to C1qa−/− mice . However , survival times did not differ significantly from C57BL/6 control mice ( n = 8 , 266±33 dpi; C1qa−/− mice: p = 0 . 24; CD21−/− mice: p = 0 . 05 ) ( Fig . S5A and S5B ) . Western blot analysis of one terminally scrapie-sick C1qa−/− mouse revealed one PrPSc positive spleen ( 1/4 ) ( Fig . S6F ) . Two terminally scrapie-sick CD21−/− mice showed PK resistance in their spleens ( 2/10 ) ( Fig . S6G ) . These results indicate that the complement components C1qa and CD21 are not essential for prion propagation upon intranasal application . CXCR5 controls the positioning of B-cells in lymphoid follicles , and the FDCs of CXCR5-deficient mice are in close proximity to nerve terminals , leading to a reduced incubation time after i . p . prion inoculation [39] , [61] . Here we explored the impact of CXCR5 deficiency onto intranasal prion inoculation . CXCR5−/− mice exhibited attack rates of 100% , and incubation times did not differ significantly from those of C57BL/6 mice ( n = 5 , 313±91 dpi; p = 0 . 32 ) ( Fig . 6D ) . 3 out of 5 terminally scrapie-sick CXCR5−/− mice revealed PK resistant material in their spleens ( 3/5 ) , as detected by Western blot analysis ( Fig . S6H ) . Pharmacological inhibition of LTβR signaling strongly reduces peripheral prion replication and reduces or prevents prion neuroinvasion upon i . p . prion challenge [42] , [44] , [53] . To determine whether inhibition of LTβR signaling would affect prion transmission through the nasal cavity , we treated C57BL/6 mice with 100µg LTβR-Ig and for control with 100µg muIgG/mouse/week pre- and post-prion challenge ( −7 days , 0 days , +7 days; 14 days ) . LTβR-Ig-treated mice were then inoculated intranasally with prions . 100% of the intranasally challenged mice died due to terminal scrapie ( C57BL/6 mice treated with LTβR-Ig: n = 8 , 476±200 dpi; Fig . 7A ) . MuIgG treated mice served as controls and showed an insignificantly shortened incubation time ( attack rate: 100% , n = 9 , 246±29 dpi ) ( Fig . 7B and C; C57BL/6 LTβR-Ig treated vs . muIgG treated mice: p = 0 . 014; C57BL/6 untreated vs . LTβR-Ig treated C57BL/6 mice: p = 0 . 021; C57BL/6 untreated vs . C57BL/6 muIgG treated mice: p = 0 . 22 ) . We additionally challenged LTβR−/− , TNFR1−/− and LTα−/− mice intranasally with RML prions ( Fig . 7D–H ) . Under these conditions all LTβR−/− , TNFR1−/− and LTα−/− mice developed terminal scrapie ( LTβR−/− mice: n = 6 , 291±52 dpi; TNFR1−/− mice: n = 3 , 213±1 dpi; LTα−/− mice: n = 6 , 251±20 dpi ) ( Fig . 7D–H ) . Terminal scrapie was confirmed by immunohistochemistry , histoblot and WB analysis ( Fig . 7F and H , data not shown ) . Spleens of intranasally inoculated LTβR−/− and TNFR1−/− mice displayed no PK resistant material ( LTβR−/− mice: 0/6; TNFR1−/− mice: 0/3 ) . In LTα−/− mice 1 out of 6 spleens contained PrPSc , while splenic PrPSc deposits of PK-resistant material were abundantly found in terminally scrapie-sick tga20 mice ( tga20 mice: 2/10 ) ( Fig . S6I–L ) .
Although aerial transmission is common for many bacteria and viruses , it has not been thoroughly investigated for prion aerosols [11] , [12] , [29] , [30] , [31] , [32] , [33] , [34] , [62] and prions are not generally considered to be airborne pathogens . Yet olfactory nerves have been discussed as a possible entry site for prions [11] , and indeed contact-mediated prion exposure of nostrils can efficiently infect various species . We therefore set out to investigate the possible hazards of prion infection deriving from exposure to prion aerosols . Our results establish that aerosolized prion-containing brain homogenates that aerosols are efficacious prion vectors . Incubation time and attack rate after exposure to prion aerosols depended primarily on the exposure time , the PrPC expression level of recipients and , to a lesser degree , the prion titer of the materials used to generate prion aerosols in a standardized inhalation chamber . The paramount role of the exposure time suggests that the rate of transepithelial ingress of prion through the airways may be limiting even when prions are offered in relatively low concentrations . Conversely , the total prion uptake capacity by the respiratory system was never rate-limiting , because the incubation time of scrapie decreased progressively with higher concentrations and longer exposure times , and because we were unable to establish a response plateau . The latter phenomenon may be explained by the large alveolar surface potentially available for prion uptake . Since it occurred in wt mice of disparate genetic backgrounds ( C57BL/6; CD1; 129SvxC57BL/6 ) , aerosolic infection may represent a universal phenomenon untied to the genetic peculiarities of any specific mouse strain ( Fig . S7 features a representative panel of histological features in CD1 mice ) . However , in CD1 mice the rapidity of progression to clinical disease did not correlate with the exposure time at a given concentration of IBH used for generating prion aerosols , suggesting the existence of genetic factors modulating the saturation of aerogenic prion intake . The passage of infectivity from the peritoneum to the brain requires a non-hematopoietic conduit that expresses PrPC [63] . We therefore sought to determine whether such a conduit would be required for transfer of infectivity from the aerosols to the brains of recipients . Using NSE-PrP transgenic mice , we found that neuron-selective expression of PrPC sufficed to confer susceptibility of mice to prion infection by aerosols and intranasal application . Hence PrPC expression in non-neural tissues is not required for aerosolic or intranasal neuroinvasion . Following peripheral exposure , many TSE agents accumulate and replicate in host lymphoid tissues , including spleen , lymph nodes , Peyers' patches , and tonsils [59] , [64] , [65] , [66] , [67] , [68] , [69] in B-cell and lymphotoxin-dependent process [70] , [71] . After peripheral replication in the LRS , prions gain access to the CNS primarily via peripheral nerves [23]; the innervation of secondary lymphoid organs and the distance between FDCs and splenic nerve endings is rate-limiting step for neuroinvasion [38] [39] . In contrast to the above , aerosolic and intranasal exposure led to prion infection in the absence of B- , T- , NK-cells and mature FDCs . Although a trend towards a slight delay in incubation time was detected in certain immunodeficient mice ( e . g . LTβR−/− and C3C4−/− ) and after LTβR-Ig treatment , these differences were not statistically significant , and all other immunodeficient ( JH−/− , Rag1−/− and γCRag2−/− ) as well as complement-deficient ( e . g . C3C4 and CD21 ) mice were susceptible to aerosolic and intranasal prion infection similarly to control mice . We conclude that transmission into the CNS upon aerosolic prion inoculation requires neither a functional adaptive immune system nor microanatomically intact germinal centers with mature FDCs . Further , the interference with LT signaling , be it by LTβR-Ig treatment or through ablation of the LTβR , indicates that the anatomical and functional intactness of lymphoid organs is dispensable for prion neuroinvasion , brain prion replication , and clinical scrapie . Since genetic removal of the main cellular components of the LRS ( e . g . by intercrosses with mice lacking T- , B-cells or NK-cells in , JH−/− or γCRag2−/− mice ) as well as genetic ( LTα−/−; LTβR−/− ) or pharmacological ( LTβR-Ig ) depletion of follicular dendritic cells - the main cell responsible for prion replication in secondary lymphoid organs - did not change the course of disease upon infection with prion aerosols , we conclude that the above data demonstrate that the LRS is dispensable for prion infection through the aerogenic route . We therefore propose that airborne prions follow a pathway of direct prion neuroinvasion along olfactory neurons which extend to the surface of the olfactory epithelium . The infectibility of newborn mice supports this hypothesis , since these mice lacked a fully mature immune system at the time of prion exposure . Our results contradict previous studies [12] claiming a role for the immune system in neuroinvasion upon intranasal prion infection , but are consistent with recent work [11] showing that prion neuroinvasion from the tongue and the nasal cavity can occur in the absence of a prion-infected LRS . Transmission of CWD to “cervidized” transgenic mice via aerosols and upon intranasal administration has also been shown [45] . Both LTβR−/− and LTα−/− mice lack Peyer's patches and lymph nodes as well as an intact NALT which may influence prion replication competence [11] , [12] , [29] , [30] , [31] , [32] , [33] , [34] , [63] . Furthermore , these mice display chronic interstitial pneumonia . Consistently with a role for LTβR-signaling in peripheral prion infection , these mice do not replicate intraperitoneally administered prions . On the other hand , TNFR1−/− mice lack Peyer's patches , show an aberrant splenic microarchitecture , an abnormal NALT , but have intact lymph nodes where prion replication can occur efficiently [72] . However , prion replication efficacy in spleen is almost completely abrogated [73] and TNFR1−/− mice die due to scrapie after a prolonged incubation time when peripherally challenged with prions . In the present study , all LTα−/− mice succumbed to scrapie upon intranasal infection , whereas some LTα−/− mice acquired prion infection following nasal cavity exposure in a previous study [11] . The requirement for the LRS in intranasal prion infection may depend on the particular prion strain being tested and on the size of the administered inoculum . When present in sufficiently high titers , prions may be able to directly enter the nervous system via the nasal mucosa and olfactory nerve terminals ( Fig . 8 ) . However , at limiting doses , aerial prion infection may be potentiated by an LRS-dependent preamplification step ( Fig . 8 ) , e . g . in the bronchial lymph nodes ( BLNs ) , the nose , the gut-associated lymphoid tissue ( NALT; GALT ) , or the spleen . In this study , the particle size generated by the nebulizer ensured that the entire respiratory tract was flooded by the aerosol so that the prion-containing aerosolized brain homogenate would reach the alveolar surface of the lung . There , prions may also colonize airway-associated lymphoid tissues and gain access to the CNS ( Fig . 8 ) . Infection through conjunctival or corneal structures was not required , since newborn mice succumbed to scrapie with an incidence of 100% despite having closed eyelids . While newborn tga20 , but not CD-1 , mice experienced slightly prolonged incubation times when compared to adult ( 6–8 week-old ) mice of the same genotype , the anatomical structures of the nasopharynx ( e . g . olfactory epithelium and olfactory nerves ) are not similarly developed at postnatal day one when compared to adulthood , potentially leading to a less efficient prion uptake upon aerosol exposure ( e . g . via olfactory nerves ) . Although unlikely , it can not be excluded that infection through conjunctival or corneal structures might contribute to a more efficient prion infection upon aerosol exposure . Be as it may , all newborn mice of either genotypes succumbed to terminal scrapie upon aerosol prion infection despite their lack of fully developed lymphoid organs , thereby bolstering our conclusion that the immune system is dispensable for prion transmission through aerosols . In summary , our results establish aerosols as a surprisingly efficient modality of prion transmission . This novel pathway of prion transmission is not only conceptually relevant for the field of prion research , but also highlights a hitherto unappreciated risk factor for laboratory personnel and personnel of the meat processing industry . In the light of these findings , it may be appropriate to revise current prion-related biosafety guidelines and health standards in diagnostic and scientific laboratories being potentially confronted with prion infected materials . While we did not investigate whether production of prion aerosols in nature suffices to cause horizontal prion transmission , the finding of prions in biological fluids such as saliva , urine and blood suggests that it may be worth testing this possibility in future studies .
Animals were maintained under specific pathogen-free conditions and experiments were approved and conform to the guidelines of the Swiss Animal Protection Law , Veterinary office , Canton Zurich . Mouse experiments were performed under licenses 40/2002 and 30/2005 according to the regulations of the Veterinary office of the Canton Zurich and in accordance with the regulations of the Veterinary office Tübingen . Exposure of mice to aerosols was performed in inhalation chambers containing a nebulizer device ( Art . No . 73-1963 , Pari GmbH , Munich , Germany ) run with a pressure of 1 . 5 bar generating 100% particles below 10 µm with 60% of the particles below 2 . 5 µm and 52% below 1 . 2 µm . Such particle sizes are considered to be able to reach upper and lower airways [74] . Prion infected material used throughout this study was RML6 strain obtained from the brains of diseased CD1 mice in its 6th passage ( RML6 ) . Mice were exposed to aerosolized prion infected brain homogenates for one , five or ten minutes . tga20 mice serving as indicator mice were inoculated i . c . with brain tissue homogenate using 30 µl volumes ( RML6 0 . 1% , 3×105 LD50 scrapie prions ) . The animals were checked on a daily basis and were sacrificed when showing defined neurological signs such as severe gait disorders . Mice were anesthetized with Ketamine/Xylazin hydrochloride anaesthesia . 10 µl of RML6 ( 0 . 1% ) were intranasally inoculated in each nostril and on the nasal epithelium by using a 10 µl pipette . The mice were held horizontally during inoculation process and for 1 minute following the inoculation . The whole procedure was repeated after a break of 20 minutes , reaching a final volume of 40 µl of RML6 , 0 . 1% ( 4×105 LD50 scrapie prions ) . Mice were anesthetized with Ketamine/Xylazin hydrochloride anaesthesia . 100 µl of RML6 ( 0 . 1% ) 1×106 LD50 scrapie prions were i . p . inoculated into Rag1−/− and γCRag2−/− mice . Tissue homogenates were prepared in sterile 0 . 32 M sucrose using a Fast Prep FP120 ( Savant , Holbrook , NY , USA ) or a Precellys 24 ( Bertin Technologies ) . For detection of PrPSc 15µl of a brain homogenate were digested with Proteinase K ( 30 µg/ml ) and incubated for 30 min at 37°C . For detection of PrPC no digestion was performed . Proteins were separated by SDS-PAGE and transferred to a PVDF ( Immobilon-P , Millipore , Bedford , Mass . , USA ) or nitrocellulose membrane ( Schleicher & Söhne ) . Prion proteins were detected by enhanced chemiluminescence ( Western blotting reagent , Santa Cruz Biotechnology , Heidelberg , Germany ) or ECL ( from PerbioScience , Lausanne , CH ) , using mouse monoclonal anti-PrP antibody POM-1 and horseradish peroxidase ( HRP ) conjugated goat anti-mouse IgG1 antibody ( Zymed ) . Histoblots were performed as described previously [73] . Frozen brains that were cut into 12 µm-thick slices were mounted on nitrocellulose membranes . Total PrP , as well as PrPSc after digestion with 50 or 100 µg/ml proteinase K for 4 hrs at 37°C , were detected with the anti-prion POM1 antibody ( 1∶10000 , NBT/BCIP , Roche Diagnostics ) . Formalin-fixed tissues were treated with concentrated formic acid for 60 min to inactivate prion infectivity . Paraffin sections ( 2µm ) and frozen sections ( 5 or 10µm ) of brains were stained with hematoxylin/eosin . Antibodies GFAP ( 1∶300; DAKO , Carpinteris , CA ) for astrocytes were applied and visualized using standard methods . Iba-1 ( 1∶1000; Wako Chemicals GmbH , Germany ) was used for highlighting activated microglial cells . Postfixation in formalin was performed for ∼8 hrs and tissues were embedded in paraffin . After deparaffinisation , for PrP staining sections were incubated for 6 min in 98% formic acid and washed in distilled water for 30 min . Sections were heated to 100°C in a steamer in citrate buffer ( pH 6 . 0 ) for 3 min , and allowed to cool down to room temperature . Sections were incubated in Ventana buffer and stains were performed on a NEXEX immunohistochemistry robot ( Ventana instruments , Switzerland ) using an IVIEW DAB Detection Kit ( Ventana ) . After incubation with protease 1 ( Ventana ) for 16 min , sections were incubated with anti-PrP SAF-84 ( SPI bio , A03208 , 1∶200 ) for 32 min . Sections were counterstained with hematoxylin . We selected 5 anatomic brain regions from all investigated or at least 3 mice per experimental group . We evaluated spongiosis on a scale of 0–4 ( not detectable , mild , moderate , severe and status spongiosus ) . Gliosis and PrP immunological reactivity was scored on a 0–3 scale ( not detectable , mild , moderate , severe ) . A sum of the three scores resulted in the value obtained for the lesion profile for the individual animal . The ‘radar plots’ depict the scores for spongiform changes , gliosis and PrP deposition . Numbers correspond to the following brain regions: ( 1 ) hippocampus , ( 2 ) cerebellum , ( 3 ) olfactory bulb , ( 4 ) frontal white matter , ( 5 ) temporal white matter . Investigators blinded to animal identification performed histological analyses . Misfolded Protein Assay ( MPA ) was performed as described previously [75] . The assay , which was performed on a 96-well plate is divided into two parts: the PSR1 Capture and an ELISA . For the PSR1 Capture the set up of each reaction was as following: 3µL of PSR1 beads ( buffer removed ) and 100µl of 1× TBSTT were spiked with brain homogenate , incubated at 37°C for 1hr with shaking at 750rpm , the beads were washed on the plate washer ( ELX405 Biotek ) 8 times with residual 50 µl/well TBST . Then 75µl/well of denaturing buffer was added . This was incubated at RT for 10min with shaking at 750rpm . Subsequently 30µl/well of neutralizing buffer were added . An additional incubation at RT for 5min with shaking at 750rpm followed . The beads were pulled down with a magnet . The ELISA was performed as follows: 150µL/well of the sample was transferred to an ELISA plate which was coated with POM19 . An incubation step at 37°C for 1hr with shaking at 300rpm followed . That was washed 6 times with wash buffer . POM2-AP conjugate had to be diluted to 0 . 01µg/mL in conjugate diluent . 150µL/well of diluted conjugate was added . Incubation at 37°C for 1hr without shaking followed . Washing 6 times with wash buffer was followed by preparation of enhanced substrate by adding 910µL of enhancer to 10mL of substrate ( Lumiphos plus , Lumigen ) . 150µL/well of enhanced substrate was added . Incubation at 37°C for 30min was followed by reading by luminometer ( Luminoskan Ascent ) at default PMT , filter scale = 1 . Real-time PCR was performed on purified genomic DNA from mouse tails on a 7900 HT Fast Real-Time PCR System ( AB ) . Data were generated and analyzed using SDS 2 . 3 and RQ manager 1 . 2 software . The following primers were used: Forward primer annealing in mouse Prnp gene intron 1: 5′ - GGT TTG ATG ATT TGC ATA TTA G - 3′ . Reverse primer annealing in mouse Prnp gene exon 2: 5′ - GGA AGG CAG AAT GCT TCA GC - 3′ . The PCR product is approximately 200 bps in length . For control , the mouse Lymphotoxin alpha gene was analyzed . The following primers were used: Forward primer annealing to the Exon 1 of the mouse Lymphotoxin alpha gene: 5′ - CCT GGT GAC CCT GTT GTT GG - 3′ . Reverse primer annealing to the mouse Lymphotoxin alpha gene Intron 1: 5′ - GTG GGC AGA AGC ACA GCC - 3′ . The PCR product is approximately 160 bps in lenght . Real time PCR analysis revealed 2–4 transgene copies per Prnp allele in Prnpo/o/NSE-PrP mice . Results are expressed as the mean+standard error of the mean ( SEM ) or standard deviation ( SD ) as indicated . Statistical significance between experimental groups was assessed using an unpaired two-sample Student's t-Test ( Excel ) and two-sample Welch t-Test for distributions with unequal variance ( R ) . For survival analyses , Kaplan-Meier-survival curves were generated using SPSS or R software , statistical significance was assessed by performing log rank tests ( R ) . Linear regression fits and analyses of variance ( ANOVA ) were conducted in R ( www . r-project . org ) .
|
Prions , which are the cause of fatal neurodegenerative disorders termed transmissible spongiform encephalopathies ( TSEs ) , can be experimentally or naturally transmitted via prion-contaminated food , blood , milk , saliva , feces and urine . Here we demonstrate that prions can be transmitted through aerosols in mice . This also occurs in the absence of immune cells as demonstrated by experiments with mice lacking B- , T- , follicular dendritic cells ( FDCs ) , lymphotoxin signaling or with complement-deficient mice . Therefore , a functionally intact immune system is not strictly needed for aerogenic prion infection . These results suggest that current biosafety guidelines applied in diagnostic and scientific laboratories ought to include prion aerosols as a potential vector for prion infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] |
[
"neuroscience",
"pathology/neuropathology"
] |
2011
|
Aerosols Transmit Prions to Immunocompetent and Immunodeficient Mice
|
Tsetse ( Glossina sensu stricto ) are cyclical vectors of human and animal trypanosomoses , that are presently targeted by the Pan African Tsetse and Trypanosomiasis Eradication Campaign ( PATTEC ) coordinated by the African Union . In order to achieve effective control of tsetse , there is need to produce elaborate plans to guide intervention programmes . A model intended to aid in the planning of intervention programmes and assist a fuller understanding of tsetse distribution was applied , in a pilot study in the Masoka area , Mid-Zambezi valley in Zimbabwe , and targeting two savannah species , Glossina morsitans morsitans and Glossina pallidipes . The field study was conducted between March and December 2015 in 105 sites following a standardized grid sampling frame . Presence data were used to study habitat suitability of both species based on climatic and environmental data derived from MODIS and SPOT 5 satellite images . Factors influencing distribution were studied using an Ecological Niche Factor Analysis ( ENFA ) whilst habitat suitability was predicted using a Maximum Entropy ( MaxEnt ) model at a spatial resolution of 250 m . Area Under the Curve ( AUC ) , an indicator of model performance , was 0 . 89 for G . m . morsitans and 0 . 96 for G . pallidipes . We then used the predicted suitable areas to calculate the probability that flies were really absent from the grid cells where they were not captured during the study based on a probability model using a risk threshold of 0 . 05 . Apart from grid cells where G . m . morsitans and G . pallidipes were captured , there was a high probability of presence in an additional 128 km2 and 144 km2 respectively . The modelling process promised to be useful in optimizing the outputs of presence/absence surveys , allowing the definition of tsetse infested areas with improved accuracy . The methodology proposed here can be extended to all the tsetse infested parts of Zimbabwe and may also be useful for other PATTEC national initiatives in other African countries .
Trypanosomosis is one of the major constraints to rural development in sub-Saharan Africa [1] . Tsetse ( Glossina spp . ) , the primary vectors of animal and human trypanosomosis , are found in the semi-arid , sub-humid and humid lowlands of 37 countries across the continent with a potential distribution range of some 8 . 7 million km2[2] . This disease places approximately 50 million cattle at risk with losses amounting to US$4 . 75 billion annually [3] . In Zimbabwe , an area of approximately 180 , 000 km2 of the total 390 , 757 km2 was deemed to be ecologically suitable for tsetse before the rinderpest epizootic of 1896 [4] . Sustained interventions resulted in the clearance of tsetse flies from most of this area , with 50 , 000 km2 being cleared since 1980 . Tsetse are now confined to approximately 28 , 000 km2 in North-Western and Northern Zimbabwe . However , tsetse transmitted trypanosomosis remains a challenge in areas close to tsetse infested areas with a total of 240 African Animal Trypanosomosis ( AAT ) cases being reported to the OIE between 2009 and 2015[5] . A Human African trypanosomosis ( HAT ) focus also exist in the Hurungwe District and Mana Pools areas in the Northern parts of the country [6]where 25 cases of the acute form of HAT caused by Trypanosoma rhodesiense were detected through passive surveillance between 2009 and 2015 [7] . The country has committed to eradicate tsetse and trypanosomiasis in the framework of the African Union coordinated Pan-African Tsetse and Trypanosomiasis Eradication Campaign ( AU-PATTEC ) , a decision ( AHG/156 ( XXXVI ) ) by African Heads of State and Government during the 36th Ordinary Summit of the OAU , Lome , Togo held in July 2000 . The distribution of tsetse and their abundance play an important role in the epidemiology of trypanosomosis and often forms the basis for intervention programmes . Insect intervention and pre-intervention programmes require accurate and up–to–date information on the spatial and temporal distribution of target insects[8] . Strategies to control or eventually eliminate the problem posed by trypanosomosis must rely on tsetse ecology and suitable fly distribution data [9] . However , it has been decades since the latest tsetse distribution maps at the continental level were produced [10] . A number of studies have been carried out in order to understand tsetse population dynamics and these have resulted in an increased understanding of the link between the environment and tsetse presence and abundance [11 , 12] . It has also been established that tsetse are highly dependent on particular habitats for their survival , therefore ecological and land use change has a major impact on fly populations and the associated disease risks [13] . The distribution , prevalence and impact of vector-borne diseases are often affected by anthropogenic environmental changes that alter interactions between the host , the parasite and the vector [14] . Recent advances in geospatial technology have enabled the development of models in the study of diseases and parasites . Georeferenced datasets and spatial analysis techniques have great potential to support the planning and implementation of interventions against human and animal diseases including African trypanosomosis [15] . Geographic Information Systems ( GIS ) based distribution mapping can help identify areas of occurrence at the micro-level , where species-specific , environmentally friendly control measures can be strengthened[16] . In recent years , tsetse and trypanosomosis distribution models have been developed at different scales . Distribution models have been produced at a continental scale from low spatial resolution data , using the Advanced Very High Resolution Radiometer ( AVHRR ) data from the NOAA ( www . noaa . gov ) satellite that present a spatial resolution of 28 km [17] . This level of resolution does not allow an accurate identification of suitable habitats for tsetse flies that are found diluted in the surrounding pixels[11] . On the contrary , studies conducted at a higher resolution in Senegal recently and based on Maximum Entropy ( MaxEnt ) models assisted in the identification of pockets of infestation that had been missed by surveys [18] . A study conducted in the North-Western parts of Zimbabwe has also shown great potential in modelling the distribution of suitable tsetse habitats , information that can be used in the planning of intervention programmes [19] . Here we propose to combine this approach ( Maxent models ) to probability models that has been used previously to delimit tsetse control areas and that stipulate that tsetse can still be present despite a series of zero catches [20 , 21] . The goal is to prepare for control operations in the study area , but also to produce a standardized method allowing optimizing the definition of tsetse infested areas within the framework of PATTEC .
The field work was authorised by the Tsetse Control and Division , Department of Livestock and Veterinary Services . The study was conducted in Masoka area , Mbire District ( 16 . 00° to 16 . 28°S and 30 . 1° to 30 . 28°E ) between March 2015 and December 2015 ( Fig 1 ) . This area belongs to the Natural Farming Region IV of Zimbabwe which receives between 650 and 800 mm of rainfall annually and is suitable for livestock and drought resistant crop production . During the dry season , most of the vegetation sheds its leaves and annual grasses and shrubs dry out . A concentration of leafy vegetation is left along water courses , although most of these are temporary . The area is part of the Community Areas Management Programme for Indigenous Resources ( CAMPFIRE ) scheme , which advocates for the conservation of natural resources , including wildlife . The area thus has a variety of wild animals , the most common ones being buffalo ( Syncerus caffer ) , elephants ( Loxodonta africana ) , warthog ( Phacochoerus africanus ) , among other important tsetse hosts . The distribution of these wild hosts in the dry season is mainly influenced by water availability , as more animals were sighted towards Chewore Safari Area , a protected Parks and Wildlife Authority of Zimbabwe Estate . According to a census conducted by Zimstat the community has an estimated population of 1 , 632 inhabitants distributed among 300 households [22] . Agriculture is the major activity , with production centred on cattle and goat rearing , cotton and small grains production . Cattle form an important source of blood meal for tsetse , especially in areas with low wild host densities [23] . The Masoka community has a herd of 180 cattle ( Division of Veterinary Services Nov 2015 Census ) . The tsetse population occupying this area has not been affected by intervention programmes instituted by the Division of Tsetse Control over the past 19 years [24] with control activities concentrated along Manyame River some 70 km away . Tsetse data were obtained using a grid based sampling method outlined in the Food and Agriculture Organisation ( FAO ) /International Atomic Energy Agency ( IAEA ) entomological baseline data collection manual of 2008 [25] . The study area was divided into a grid of 110 identical cells measuring 2 km × 2 km and a minimum of one and a maximum of three epsilon traps baited with sachets containing a mixture of 3-n-propyl-phenol , o-cten-3-ol and 4-methyl-phenol in the ratio of 1;4;8 [26] were placed in cells perceived to have suitable habitat . In each sampled cell , sites perceived to be suitable tsetse habitat were chosen based on a supervised classification of a SPOT 5 image . A survey team led by an experienced Tsetse Field Assistant , chose the actual site on the ground based on recommendations by Vale [12] in order to maximise on catches . Each sampling site was geo-referenced using a hand-held GPS receiver and monitoring was done after seven days . Samples were collected from 105 sites between March 2015 and December 2015 . Captured flies were identified morphologically using identification keys developed by Buxton [27] and Mulligan [28] and specimens were preserved in 90% alcohol . The temperature is a parameter that plays an important role in the tsetse life cycle and Land Surface Temperature is among the commonly used temperature indicators . Land Surface Temperature ( LST ) is calculated from the measurement of radiation emitted by the earth surface and it is highly correlated withthe air temperature [30] . 8 days daytime ( DLST ) and night-time LST ( NLST ) were extracted at 1km spatial resolution from MODIS MOD11A2/MYD11A2 temperature and emissivity products . The data werefiltered and temporally aggregated into statistics that can be used to describe the thermal profile of the study area . LST is used in many studies of species distribution and spatial epidemiology . In this study , they were used as proxies for both air and soil temperature which play an important role in tsetse habitat selection . Among the indices commonly used in epidemiological studies are vegetation indices , a measurement of chlorophyll activity . These indices allow the differentiation of bare ground from the vegetation and also of various vegetation types . The most commonly used is the NDVI ( Normalized Difference Vegetation Index ) , in addition to the NDVI , other vegetation indices such as EVI ( Enhanced Vegetation Index ) can be used but according to Hay [31] , EVI is particularly useful since it performs better than NDVI over high biomass areas . The vegetation continuous field ( VCF ) is also an important vegetation index that can be used to capture the density of tree cover . Regarding our study area , we used both EVI and the VCF ( Treecover ) to capture the effect of woody vegetation on tsetse habitat . It is also important to note that EVI and other vegetation indices have already been used several times to predict tsetse flies’ density in West Africa [11 , 18 , 32 , 33] . The reflectance , in the mid-infrared is used to measure the radiation of bare soils . This index is correlated with the Land surface temperature . Luxuriant vegetation is characterized by a low MIR . With the EVI , this index allows to characterize the vegetation well as the soil temperature . Various topographic indices such as slope , topographic wetness index ( TWI ) and aspect ( slope direction ) can be extracted from Digital Elevation Model ( DEM ) . DEM , slope and aspect can be used to describe the elevation and exposure to the sun whereas topographic wetness index measure soil humidity . These indices were also used to model habitat suitability for the two species . A high-resolution remotely-sensed satellite image acquired on the 9th of November 2014 by the Satellite Pour l’Observation de la Terre 5 ( SPOT 5 with a spatial resolution of 2 . 5m ) was used to identify suitable areas for tsetse . A supervised classification of land cover was realized with Envi 5 . 1software ( www . exelisvis . co . uk ) , based on a maximum likelihood classifier ( Fig 2 ) . Eighty three polygons were digitized manually and 106 GPS filed observations were used to validate the classification . The supervised classification was validated from the calculation of a confusion matrix and the Kappa Index of Agreement coefficient ( 0 . 95 ) . The pair comparisons of the landcover classes gave a separability coefficient between 1 . 97 and 2 , corresponding to an absence of confusion of the pixels allocated within each class [34] . Seven classes of land cover were identified from which 4 ( mopane , riverine forest , crop field and bush land ) , were used as predictors in the habitat suitability models . For each of these classes , the patch density ( number of patches ) and the surface of patches inside the prediction pixels at a resolution of 250m were calculated . The list of the remotely-sensed data and their spatial and temporal resolution used in the present study are presented in Table 1 . In the second step , we used a statistical model to predict suitable habitats for both species . We used the Maximum Entropy ( MaxEnt ) ( www . cs . princeton . edu/~schapire/maxent ) , a species distribution model . MaxEnt is a machine learning algorithm that applies the principle of maximum entropy to predict the potential distribution of species from presence-only data and environmental variables . We resampled climatic and environmental data to a spatial resolution of 250 m and used them as the known features in determining the suitability index of each tsetse species within the study area . Each tsetse species , G . morsitans and G . pallidipes was modelled separately and for each we used presence data and a set of randomly generated pseudo-absence . We used leave one out cross validation ( LOOCV ) to compute all the model quality metrics . The model was trained n times ( n = sample size ) and each time we removed one observation for validation and at the end we aggregated the n metrics calculated on the validation point . The absence data were used only to assess the accuracy of each model and set a threshold for the model . We used the Receiver Operator Characteristic ( ROC ) curve and the associated , Area Under the Curve ( AUC ) as a metric for assessing the quality and performance of our prediction [37] . An AUC with values closer to 1 indicating excellent prediction . The MaxEnt software was used through its R interface in the dismo package [38] .
A total of 73 cells ( 292 km2 ) of the 110 cells were sampled with 105 traps . Survey results demonstrated a mean density of 0 . 27 ( sd = 0 . 54 ) flies/trap/day for G . m . morsitans , with a presence in 40 sites distributed in 31 cells ( 124 km2 ) The mean density of Glossina pallidipes was 0 . 05 ( sd = 0 . 16 ) flies/trap/day , with a presence in 15 trapping sites within 13 cells ( 52 km2 ) ( Fig 3 ) . The first plan of the ENFA showed that G . m . morsitans occurrence was positively correlated with vegetation indices ( EVI , Riverine Forest , Average tree cover and MIR ) . However , most of the temperature indices exhibited a negative correlation to the species ( Fig 4 ) . Mopane woodland patch density and aspect exhibited an important influence on the habitat for the species as they were strongly correlated with the specificity axis . Average EVI accounted for most of the variance and fell outside the cloud of average conditions available in the study area . The occurrence of G . pallidipes also showed a positive correlation to vegetation related indices with most of the temperature indices exhibiting a negative correlation . The topographic wetness index ( TWI ) was positively related to G . pallidipes but negatively with G . m . morsitans whilst night land surface temperature was the only temperature related covariate which showed a positive correlation with the occurrence of both species . The habitat suitability models for G . m . morsitans and G . pallidipes had an Area Under the Curve ( AUC ) of 0 . 89 and 0 . 94 respectively ( Fig 5 ) . Both figures were close to one although the G . pallidipes model had a better prediction ability . However there were differences in covariates contributing to the models . The most contributive variable was “aspect” in the G . m . morsitans model and “riverine forest patch density” in the G . pallidipes model . The resultant maps depicting habitat suitability for the species ( Fig 6 ) show a wider area suitable for G . m . morsitans than G . pallidipes . There was a concentration of suitable habitat to the west of the study area which is a protected wildlife area . We applied the probability model to 42 grid cells where no G . m . morsitans were caught . The analysis indicated a probability of G . m . morsitans presence below 0 . 05 ( the level of risk accepted ) in 10 grid cells where no tsetse were captured whilst 32 grid cells had a probability greater than 0 . 05 that G . m . morsitans was still present despite a sequence of zero catches . We observed the area infested with G . m . morsitans to be 124 km2 ( 28% ) and a further 124 km2 ( 28% ) had a high probability of being infested . An area of 40 km2 ( 9% ) had a low probability of tsetse presence whilst the remaining 148 km2 ( 34% ) were not sampled ( Fig 7 ) . We also applied the probability model to 60 grid cells where no G . pallidipes were captured . The analysis indicated a probability of tsetse presence below 0 . 05 ( the level of risk accepted ) in 24 grid cells where no tsetse were captured whilst 36 grid cells had a probability greater than 0 . 05 that G . pallidipes was still present despite a sequence of zero catches . Area infested with G . pallidipes was therefore observed to be 52 km2 ( 10% ) , area with a low probability of tsetse presence was 96 km2 ( 22% ) whilst the remaining 148 km2 ( 34% ) were not sampled ( Fig 7 ) .
Habitats suitable for G . m . morsitans and G . pallidipes can be modelled using presence data and environmental variables [19] . This study produced habitat suitability models at a high resolution ( 250 m ) , a level which can be translated into operational plans . The habitat suitability models produced in this study had relatively large AUCs , an indication of a good predictive power . This showed that the habitat of both species , to a large extent , can be explained by the covariates used . The two species under consideration , G . m . morsitans and G . pallidipes were positively correlated with vegetation indices on the first plan of the ENFA , indicating that the requirement for these covariates for these species was different than the mean conditions in the study area . The link between vegetation and tsetse has been well established through various studies [11 , 12] . Whilst both species were found in habitats along watercourses during the dry season , G . m . morsitans was also captured in deciduous woodlands of predominantly mopane trees . Studies by Vale at Rekomichi showed that there was variability in G . m . morsitans catches across vegetation types with seasonal effects evident whilst catches of G . pallidipes were distinctly higher in thickets than in mopane woodlands [12] . According to Cecchi et . al . , deciduous woodlands and deciduous shrub-lands with sparse trees account for over 50% of the total distribution of the morsitans group [9] . In their model in North Western Zimbabwe , Matawa et al , observed that higher altitude was not associated with suitable tsetse habitat for both G . m . morsitans and G . pallidipes [19] . They attributed this to the effect of altitude on other climatic factors such as temperature . In our study however , the effect of altitude could not be fully explored as the elevation in the study area was more homogenous than the in North-Western Zimbabwe hence there was little variability to examine . Aspect , however , seemed to play a role in the determination of suitable habitat probably due to its association with the amount of sunlight received and subsequently temperature . This study also demonstrated a negative correlation between suitable tsetse habitat and day land surface temperature which is a measure of air temperature . This negative correlation means G . m . morsitans and G . pallidipes require lower temperatures than the average values in the study area . Whilst studies on artificial refuges by Vale could not pin-point the exact temperature at which all tsetse occupy refuges , they clearly demonstrated that temperatures beyond 30°C affect tsetse [39] . Further work by Hargrove and Muzari revealed an increase in catches of male and pre-full term pregnant female tsetse in refugia at around 32°C [40] . Although no temperature measurements were made in this study , online weather sources reported episodes of maximum temperatures in excess of 40°C in the study area between October and December , values which compare well with 42 . 5°C observed by Hargrove and Muzari during 1998 [40] . Maximum temperature has also shown to have significant effect on tsetse survival [41] with laboratory studies showing an increase in daily mortality due to temperature [42] The aim of modelling is to improve the quality of intervention plans leading to a reduction in costs hence accuracy of the models is of paramount importance if they are to be the basis of intervention . The distribution of tsetse hosts is a critical determinant of tsetse distribution , particularly in the morsitans group [13 , 23] . We however obtained very good predictions in the present study , probably because the density of these wild hosts is correlated with the vegetation habitats that were integrated in the prediction models . This study confirmed that the absence of tsetse catches in traps does not imply absence in a locality [12 , 21] . Unlike other probability models build using vegetation only [21] this study used the habitat suitability model , a factor which captures major characteristics of the habitat thus increasing the robustness of the model . Suitable tsetse habitats are influenced by more factors other than vegetation alone thus the probability model produced in this study has got a greater chance of detecting grids with higher chances of infestation . The probability model showed greater chances of both G . m . morsitans and G . pallidipes presence in wider parts of the study area than observed through surveys . A number of factors can be attributed to this result . Firstly , the absence of tsetse in traps , especially G . m . morsitans , could have resulted from a lower efficiency of traps in capturing the species ( 0 . 001 ) [20] , a parameter which is linked to the behaviour of the species . Resting G . m . morsitans respond more to moving objects than G . pallidipes [43] . Whilst great care was taken to place traps in optimal sites , siting in itself is a factor which can influence the efficiency of traps [44] . The model however , still showed a high probability of G . pallidipes presence , contrary to catches recorded in traps which were low despite a better trapping efficiency ( 0 . 01 ) [20] . This new methodology is presented here for the first time and will allow a great enhancement of future tsetse sampling efforts . It has the potential to generate surface information ( raster data ) from point data ( trap catches ) thus providing operational information to guide planning and decision making . The model can be applied in planning the placement of insecticide treated targets as it is grid based and can also be applied to direct the focus of further surveys . Most remote sensing products are now freely available making the processing of data much cheaper thus helping national entities working on tsetse control programmes to make informed decisions in the judicious allocation of scarce resources ( Prioritization of target areas based on assessed risk ) . Previous applications of probability modelling on riverine species in West Africa allowed the detection of isolated pockets of tsetse in areas which had been missed by surveys [18 , 45] . This study also demonstrated that information on areas not surveyed within target areas can be generated to guide the planning process . This is of importance as some areas can be difficult to access whilst at times resources may be limiting to obtain data from every location of the target area . However , the presence of predicted suitable habitats in these not sampled areas will be the basis to consider them as infested or not but this will need to be confirmed by additional sampling efforts . Whilst local conditions may differ from place to place , we believe adoption of the methodology presented here would assist the country in the drafting of elaborate tsetse control and survey plans for implementation under the AU-PATTEC initiative . The methodology can also serve as a template for other PATTEC national initiatives and can be extended to assess the success of vector control programmes .
|
Tse-tse flies are vectors of human and animal trypanosomoses , that are presently targeted by the Pan African Tsetse and Trypanosomiasis Eradication Campaign ( PATTEC ) coordinated by the African Union . In Zimbabwe , the government has devoted a full section of the veterinary services to tsetse and trypanosomosis control but the delimitation of tsetse infested areas , which is a pre-requisite to achieve effective control still requires improvement . Here we present a methodology that could help delimit target areas throughout the country , in a pilot study area located in the Masoka area , Mid-Zambezi valley in Zimbabwe , and targeting two savannah species , Glossina morsitans morsitans and Glossina pallidipes . The study , which was carried out in preparation for a vector control campaign , allowed to discriminate areas where tsetse presence was certain , likely or unlikely Habitat degradation due to agricultural activities seemed to play a pivotal role in determining the infestation by tsetse since settled areas had low probabilities for both species which was expected in this group . Application of this model will help reduce the cost of delineating tsetse infested areas in other parts of Zimbabwe and may also be useful for other PATTEC national initiatives in other African countries at a time when funding for tsetse control programmes is reduced .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"livestock",
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences",
"ecological",
"niches",
"geographical",
"locations",
"trees",
"parasitic",
"diseases",
"animals",
"glossina",
"materials",
"science",
"habitats",
"surface",
"properties",
"plants",
"africa",
"surface",
"temperature",
"infectious",
"diseases",
"zoonoses",
"protozoan",
"infections",
"zimbabwe",
"trypanosomiasis",
"insects",
"agriculture",
"arthropoda",
"people",
"and",
"places",
"ecology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"material",
"properties",
"organisms"
] |
2017
|
A pilot study to delimit tsetse target populations in Zimbabwe
|
Although it has been known for 50 years that adenoviruses ( Ads ) interact with erythrocytes ex vivo , the molecular and structural basis for this interaction , which has been serendipitously exploited for diagnostic tests , is unknown . In this study , we characterized the interaction between erythrocytes and unrelated Ad serotypes , human 5 ( HAd5 ) and 37 ( HAd37 ) , and canine 2 ( CAV-2 ) . While these serotypes agglutinate human erythrocytes , they use different receptors , have different tropisms and/or infect different species . Using molecular , biochemical , structural and transgenic animal-based analyses , we found that the primary erythrocyte interaction domain for HAd37 is its sialic acid binding site , while CAV-2 binding depends on at least three factors: electrostatic interactions , sialic acid binding and , unexpectedly , binding to the coxsackievirus and adenovirus receptor ( CAR ) on human erythrocytes . We show that the presence of CAR on erythrocytes leads to prolonged in vivo blood half-life and significantly reduced liver infection when a CAR-tropic Ad is injected intravenously . This study provides i ) a molecular and structural rationale for Ad–erythrocyte interactions , ii ) a basis to improve vector-mediated gene transfer and iii ) a mechanism that may explain the biodistribution and pathogenic inconsistencies found between human and animal models .
Adenoviruses ( Ads ) are nonenveloped double-stranded DNA pathogens that infect all vertebrate classes . To date , all Ads display a characteristic , icosahedral symmetry in which 240 subunits of the trimeric hexon protein form the facets and 12 copies of the penton , comprising the pentameric penton base protein and the externally projecting trimeric fiber , form the vertices [1] . While the stoichiometry of the penton base and hexon is apparently conserved , the fiber can exist as a single or double copy at each vertex [2] . At least in vitro and for most cell types , the fiber mediates the initial attachment to primary receptors , such as the D1 domain of the coxsackievirus and adenovirus receptor ( CAR ) , sialic acids , CD46 , and others ( for review see Zhang & Bergelson [3] ) . Interaction with auxiliary receptor ( s ) , in particular some of the dimeric integrins via the Arg-Gly-Asp sequence ( an integrin-interacting motif ) on the penton base , may induce internalization of some serotypes . However , other auxiliary receptors or mechanism of internalization may exist for human serotypes 40 and 41 ( HAd40/41 ) , and canine serotype 2 ( CAV-2 ) , which have no identifiable integrin-interacting motif in the penton base [4] , [5] . Greater than 150 Ad serotypes have been isolated . Approximately 50 of these are currently classed as human pathogens that , in most cases , generate subclinical ocular , respiratory and gastrointestinal tract infections . In the immunocompromised host however , lethal HAd infections can spread , via unknown mechanisms , to the kidney , liver and brain [6] , [7] . The human Ads ( HAds ) are divided into subgroups ( or species or subgenera ) A - F . The triage of the human serotypes into subgroups is based in part on serotype-specific ex vivo erythrocyte cross-linking ( or hemagglutination ) [8] . The clinical hemagglutination assays use lysates from virus-infected cells to crosslink erythrocytes from a handful of species . This highly heterogeneous lysate contains whole virus particles , empty capsids , penton monomers , penton dodecahedrons , fiber monomers , hexon etc . By using fractionated infected cell lysate , a handful of laboratories found that the multivalent complexes containing fiber were responsible for hemagglutination [9]–[12] . Erythrocyte membranes contain highly sialated glycoproteins and glycolipids . One of the most abundant glycoproteins on erythrocytes is glycophorin A ( ∼105 copies/cell ) . With its high sialic acid content , glycophorin A is the main contributor to the net negative cell-surface charge and is critical for minimizing cell–cell interactions and preventing erythrocyte aggregation [13] . Sialic acid is a collective term for a family of 9-carbon monosaccharides , which are often found as terminal sugar residues on glycans of glycoproteins and glycolipids ( usually α2-3 , -6 or -8 linked ) . In addition to some HAds , a number of viruses , including orthomyxoviruses , paramyxoviruses , picornaviruses , papovaviruses , coronaviruses , reoviruses and parvoviruses bind to sialic acids [14]–[18] . Possibly because erythrocytes from different species vary in their sialic acid content , the hemagglutination properties of sialic acid-binding viruses may also diverge [19] . HAd subgroup D ( serotypes 9 , 15 , 19p and 37 ) and B:2 ( serotypes 11a , 11p and 34a ) erythrocyte binding depends on the fiber head and several attempts have been made to define the region ( s ) responsible [20]–[23] . Among the HAds , serotype 37 is unusual: its fiber head can bind CAR , CD46 and sialic acid , but the virus appears to use only the latter two as functional receptors [24]–[26] . Burmeister et al . found that the sialic acid moiety bound to a basic patch close to the center of the trimeric fiber heads of HAd37 and HAd19p [27] . Interestingly , the putative hemagglutination domain in subgroup D heads partially aligns with the sialyl-lactose binding site , which on the basis of sequence alignments is likely to be conserved in other members of this subgroup . These observations raise the question as to whether Ad-erythrocyte interaction is a consequence of fiber head-sialic acid interaction . In this study we initially characterize the erythrocyte binding of unrelated serotypes , HAd5 , HAd37 and CAV-2 . Although both HAd37 and CAV-2 agglutinate human erythrocytes at low particle-to-cell ratios , they use different receptors ( sialic acid and CD46 vs . CAR ) [5] , [25] , [26] , have different clinical tropisms ( ocular vs . respiratory tract ) and infect different species . We found that HAd37-erythrocyte interaction is primarily due to sialic acid binding . We show by structural analysis that the CAV-2 fiber head also contains a sialic acid binding site , but in contrast to HAd37 this site is modestly involved in hemagglutination . Unexpectedly , our biochemical and competition analyses suggested that CAV-2-erythrocyte interactions also depend on binding to CAR on human and rat erythrocytes . Using a transgenic mouse that expresses CAR on erythrocytes [28] we demonstrate that CAR-binding Ads can be sequestered by CAR-expressing erythrocytes , and prevent liver infection . Our study provides a molecular and structural rationale for the 50-year-old enigma of ex vivo Ad-erythrocyte interactions . In addition to the relevance for Ad pathogenesis and vector biology , the expression of CAR by human erythrocytes may shed light on the role of cell adhesion molecules during erythropoiesis .
Consistent with previous reports , we found that at a modest physical particle ( pp ) -to-erythrocyte ratio ( ∼130 ) HAd37 efficiently agglutinates human erythrocytes ( Figure 1A ) . When compared to HAd37 , CAV-2 hemagglutinates at ∼20-fold lower ratio ( i . e . more efficiently , Figure 1B ) . In our hands , and consistent with others [29] , HAd5 also agglutinated human erythrocytes , but at a higher ratio ( >800 pp/erythrocyte ) ( Figure 1C ) . To assay the roles of HAd37 and CAV-2 capsid proteins , we incubated erythrocytes with chimeric HAd5 vectors harboring the fiber from HAd37 [30] or the fiber head from CAV-2 [31] . We found that the HAd5-HAd37F and HAd5-CAV-2H hybrid capsids induced agglutination at lower ratios than HAd5 but not quite to those of HAd37 and CAV-2 ( Figure 1D and E ) . Pre-incubating erythrocytes with neuraminidase , which removes sialic acid from their membranes , eliminated agglutination by HAd37 and a HAd5-HAd37F hybrid capsid ( Figure 1 , right hand column ) . In contrast , removing sialic acid from erythrocyte membranes only modestly reduced CAV-2 and the hybrid HAd5-CAV-2H agglutination ( data summarized in Table 1 ) . Together , these data suggest that the CAV-2 fiber head , like in subgroup D and B:2 HAds , is responsible for hemagglutination and that sialic acid binding plays a more significant role in HAd37 binding of human erythrocytes than it does for CAV-2 . To mimic the hemagglutination caused by HAd37 fiber head in a virion-like context , we generated a multivalent protein complex similar to the “complete hemagglutinin” described for HAd9 by Norrby et al . [9] . For this purpose HAd3 penton dodecahedra were incubated with chimeric “mini fibers” consisting of HAd3 fiber tail and shaft motifs 1+2 and HAd37 fiber head . In addition , we generated HAd37 fiber heads containing mutations in the sialic acid binding site [27] ( see Table 2 for a list of fiber head mutants ) . The HAd3 penton dodecahedra without a fiber did not cause hemagglutination ( data not shown ) . The dodecahedra containing a wild type HAd37 fiber head ( HAd37Hwt ) ( Figure S1 ) agglutinated erythrocytes at low protein concentrations ( about 104-fold less mass than HAd37 virions ) ( Table 2 ) . Similar to HAd37 , this “complete hemagglutinin” poorly agglutinates neuraminidase-treated erythrocytes ( Table 2 ) . The dodecahedra containing a HAd37 fiber with a Lys to Glu mutation at amino acid 345 in the sialic acid binding site ( HAd37HSA−1 ) , had at least a 105-fold reduced hemagglutination activity compared to the dodecahedra containing HAd37Hwt . The chimeric wild type CAV-2 fiber head ( CAV-2Hwt ) did not bind to HAd3 penton dodecahedra , which precluded equivalent experiments . These data suggest that the HAd37 fiber shaft , hexon and pIX do not play key roles during binding and that the sialic acid binding site is involved in agglutination of human erythrocytes . To assay the binding of recombinant fiber heads , we incubated HAd37Hwt and mutant fiber heads with erythrocytes and then added anti-fiber head antibodies or antiserum . We then quantified attachment by flow cytometry . HAd37Hwt ( Figure 2A ) and mutants carrying point mutations in the CAR binding site ( HAd37HCAR−1 and HAd37HCAR−2 , mutation in Glu351 and Ser299 , data not shown ) [24] bound to erythrocytes in a dose-dependent manner . However , HAd37HSA−1 and HAd37HSA−2 ( a Tyr to Ala mutation at amino acid 312 ) , which harbor mutations in the sialic acid binding site , poorly bound erythrocytes ( Figure 2A ) . In addition , all of the HAd37 fiber heads poorly bound neuraminidase-treated erythrocytes ( Figure 2A , bottom row and data not shown ) . Together , these data suggest that the HAd37 fiber head sialic acid binding site is a key determinant in the attachment to human erythrocytes . Using the above approach , we detected minimal binding of CAV-2Hwt to erythrocytes ( data not shown ) , possibly because the affinity of single CAV-2 fiber heads is low , the high affinity polyclonal antibody binding out-competed the weaker erythrocyte binding , and/or the number of CAV-2 fiber head receptors on erythrocytes is low . To address the potential competition , we incubated fluorescently labeled CAV-2 fiber heads with erythrocytes to circumvent the use of antibodies . Here we detected low , but reproducible , binding of CAV-2Hwt to mock- , as well as neuraminidase-treated , erythrocytes ( Figure 2B ) . Together , these data suggest that CAV-2 erythrocyte binding is notably less dependent on sialic acid than HAd37 , and the lower binding may be due to reduced affinity or fewer receptors . To address sialic acid-fiber head interaction using a cell-free system , glycophorin or asialoglycophorin ( asialoGP ) were immobilized on BIAcore sensor chips and increasing concentrations of HAd37 fiber heads were assayed . All binding curves had a square shape , suggesting high on and off rates ( Figure S2 ) . HAd37Hwt bound to glycophorin , but less efficiently to asialoGP . Compared to HAd37Hwt , HAd37HSA−2 bound equally to asialoGP , but less to glycophorin . Interestingly , HAd37HSA−1 ( Lys345Glu ) bound less efficiently to glycophorin and asialoGP compared to HAd37Hwt and HAd37HSA−2 , which suggests that HAd37 agglutination may be charge-dependent . Consistent with the erythrocyte binding data , we found very little binding between CAV-2Hwt and glycophorin or asialoGP ( data not shown ) . Overall , the SPR results reflect the results obtained using flow cytometry with mock or neuraminidase-treated erythrocytes . Our results suggested that HAd37 hemagglutination was due to sialic acid binding via the sialic acid binding site in the fiber head . Although we have no evidence suggesting that CAV-2 can use sialic acid as a functional receptor , it is possible that the efficient CAV-2 agglutination of human erythrocytes is due to multiple and coordinated binding of sialic acid . For example , each of the 12 homotrimeric fiber heads on the end of the flexible CAV-2 shaft [32] could bind three sialic acid moieties ( theoretically up to 36/capsid ) . Because the predicted subgroup D fiber head hemagglutination domains ( in the CD and GH loops ) appear to be well conserved , we tried to identify the corresponding domain in the CAV-2 fiber head using a mutagenesis strategy based on sequence homology ( Figure S3 ) . This approach was unsuccessful ( data not shown ) , suggesting that Ad hemagglutination sites are not strictly conserved . To address possible sialic acid binding via an alternative approach , we soaked CAV-2 fiber head crystals in solutions containing 2-3 sialyl-D-lactose . The crystal diffracted to 1 . 9 Å and the structure was solved by molecular replacement using our model of CAV-2 fiber head [24] ( crystallographic details are summarized in Table S1 ) . The electron density maps showed clear density for six N-acetyl neuraminic acid ( Neu5AC ) moieties in the asymmetric unit , three per fiber head trimer . Sialyl-lactose is composed of three sugar rings , the sialic acid moiety being linked to lactose ( galactose-glucose ) . The density for the lactose moieties was weak , implying that these are flexible within the crystal . Only one of the six galactose rings in the asymmetric unit could be modeled . This molecule is located in between the two fiber head trimers in the asymmetric unit and forms a distant hydrogen bond ( distance 3 . 25 Å ) to Lys503 via the galactose oxygen 6 . It is unlikely that this interaction is enough to immobilize the galactose ring because the other five galactose moieties in the asymmetric unit are not equally well visible in the electron density . More likely , the stabilization is due to the spatial restrictions imposed by the proximity of the other fiber head trimer . We found that the sialic acid binding site on CAV-2 head is distinct , both in sequence and location , from that found on the HAd37 fiber head . On the CAV-2 fibers head , sialic acid binds further away from the three-fold symmetry axis at the periphery of the trimer ( Figure 3A–D ) . The residues involved in binding ( Asn435 , Ser419 , Ser416 , Gln417 and Arg515 ) do not align with those involved in sialic acid-binding by HAd37 fiber head ( Figure S2 ) . The HAd37 sialic acid binding site consists of three residues forming hydrogen bonds ( Tyr213 , Pro317 and Lys345 ) and two residues contacting sialic acid in hydrophobic interactions ( Tyr308 and Val322 ) ( Figure 3E ) . All seven interactions between CAV-2 fiber head and sialic acid are hydrogen bonds or salt bridges ( Figure 3F ) , suggesting a relatively strong interaction compared to HAd37 . Unlike HAd37 fiber head [33] , no hydrophobic contacts contribute to sialic acid binding . Finally , sialic acid binds within a basic patch in each head ( Figure 3C and D ) . Our results showing that the CAV-2 fiber head contains a sialic acid binding site creates a paradox: both HAd37 and CAV-2 fiber heads contain CAR and sialic acid binding sites - but HAd37 uses sialic acid and CD46 , while CAV-2 uses CAR ( CAV-2 does not use CD46 to infect cells , unpublished data ) to infect cells . To determine if sialic acid and CAR binding were mutually exclusive , we soaked CAV-2 or HAd37 fiber head crystals in complex with CAR in solutions containing sialyl-D-lactose . Crystals containing the complex with CAV-2 fiber head diffracted to 2 . 9 Å and contained 12 chains of fiber head bound to 12 chains of CAR D1 and 12 sialyl-D-lactose molecules in the asymmetric unit ( space group I422 ) . Crystals containing the complex with HAd37 fiber head diffracted to 1 . 55 Å and contained one chain of fiber head in complex with one CAR D1 molecule and one sialyl-D-lactose molecule in the asymmetric unit ( space group I23 ) . For both structures the electron density maps showed clear density for sialic acid , but not lactose , at the expected positions on the fiber heads ( Figure 3G and H ) . Together , our data demonstrate that CAV-2 and HAd37 sialic acid binding sites do not overlap with the CAR binding sites , and both fiber heads can bind CAR and sialyl-D-lactose simultaneously . Arnberg and colleagues previously showed that HAd37 interaction with sialic acid was inhibited at high salt concentration [33] . To determine if a CAV-2-erythrocyte interaction was at least partially charge dependent , we incubated a CAV-2 vector expressing GFP ( CAVGFP ) or a HAd5 vector expressing GFP ( AdGFP ) with mock- or neuraminidase-treated erythrocytes in PBS containing increasing concentrations of ions . We then pelleted the erythrocytes by centrifugation , removed an aliquot of the supernatant , added it to cells , and assayed the cells for GFP expression by flow cytometry 24 hr post-infection . Using mock-treated erythrocytes ( Figure 4 ) , we found that at physiological salt concentrations ∼80% of CAVGFP was removed from the supernatant , while at higher salt concentration ( 300 mM NaCl ) ∼60% of CAVGFP was removed . When using neuraminidase-treated erythrocytes , we again found that CAVGFP was efficiently removed from the supernatant . All the test samples were significantly ( P<0 . 01 ) different from the control , as well as from the mock-treated 150 mM NaCl ( P<0 . 05 ) . Consistent with other studies [29] , [34] AdGFP was also efficiently ( 90% ) removed from the supernatant after the incubation with mock- ( or neuraminidase- ) treated erythrocytes . These data suggest that , like HAd37 , a modest degree of CAV-2-erythrocyte interactions may depend on electrostatic interactions . In addition , both the CAR-tropic adenoviruses ( HAd5 and CAV-2 ) bind human erythrocytes at physiological salt concentrations . While our results show that HAd37 hemagglutination is due primarily to sialic acid binding , CAV-2 hemagglutination appeared more complex . We therefore developed additional tests to address the molecular and structural basis . Due its sensitive and semi-quantitative potential , we returned to hemagglutination assays to understand the erythrocyte interactions . Freshly purified CAV-2Hwt , which predominantly consists of individual trimeric fiber heads , did not cause hemagglutination , presumably because it is not sufficiently multivalent to crosslink erythrocytes . However , several His-tagged Ad fiber heads , including those of HAd41 short fiber [35] and HAd37 , form multimers that dissociate into single trimeric fiber heads upon removal of the histidine tag . Similarly , His-tagged CAV-2 fiber head forms multimers of a defined size that are stable on a size-exclusion column and can be visualized with an electron microscope ( not shown ) . Removal of the His-tag yields single fiber heads . We found that His-tagged CAV-2Hwt agglutinated erythrocytes , while His-tagged CAV-2HSA−1 and CAV-2HSA−2 ( two CAV-2 fiber heads with one or two mutations in the sialic acid binding site , see Table 2 ) showed reduced hemagglutination titers . However , we could not exclude the possibility that the reduced hemagglutination in these latter CAV-2H constructs was due to electrostatic interactions ( the mutations modified the charge of the fiber heads ) . We next assayed CAV-2 hemagglutination using a competition assay ( see Figure S4 for schema ) . In these competition assays , we pre-incubated erythrocytes with fiber heads , antibodies , salt and/or neuraminidase . Then the erythrocytes were incubated with CAV-2 and compared to mock-treated erythrocytes . We found that hemagglutination could be >256-fold reduced by pre-incubating CAV-2 with anti-fiber head antibodies , or pre-incubating erythrocytes with CAV-2Hwt ( see Figure 5A for specific example and 5B for cumulative data ) . Pre-incubating the erythrocytes with the head from HAd5 or performing the assay in 225 mM NaCl led to modest 2 to 4-fold reductions . Unexpectedly , we found that like CAV-2Hwt , CAV-2HSA−1 notably reduced agglutination ( ∼16-fold ) , while a CAV-2 fiber head with a mutation in the CAR binding site ( CAV-2HCAR− ) had a modest ∼2-fold reduction . In most cases , pre-treating erythrocytes with neuraminidase had a fairly small additive effect of CAV-2 hemagglutination . Together , these data suggest that the CAR binding site , which is present in CAV-2Hwt and CAV-2HSA−1 but not CAV-2HCAR− , is involved in CAV-2 agglutination of human erythrocytes . That the CAV-2 CAR binding site is involved in erythrocyte binding is consistent with the study by Nicol et al . [36] , which showed that a CAR-ablated HAd5 vector no longer agglutinated human and rat erythrocytes . The presence of CAR on erythrocytes would be inconsistent with other reports [34] . Among other functions , CAR acts as a homodimeric cell adhesion molecule at tight gap junctions [37] . However , the expression of cell adhesion molecules during erythropoiesis is not unprecedented [38] . Erythrocytes from some species express cell adhesion molecules during the early stages of differentiation that are thought to be involved in interaction with macrophages . We therefore incubated erythrocytes with anti-CAR antibodies that recognize the extracellular domain of CAR and assayed expression using flow cytometry . We found low , but reproducible , CAR expression on the cell surface of human erythrocytes ( Figure 6A ) . To assay CAR expression using another approach , we used western blot analysis to screen erythrocytes from several species . Using an anti-CAR Ab that recognizes the cytoplasmic domain of CAR we found CAR expression on human and rat , but not on mouse , dog , rabbit and nonhuman primate erythrocytes ( Figure 6B and data not shown ) . Again , our results showed a relatively low level of CAR on human erythrocytes , which is consistent with the low level of CAV-2Hwt binding to erythrocytes ( Figure 2B ) . These data suggest that CAR binding is a significant factor in CAV-2 , and likely other CAR-tropic Ads , interaction with human erythrocytes . This also is consistent with the fact that CAV-2 agglutinates rat erythrocytes , but not erythrocytes from mice , dogs , rabbits or some nonhuman primates ( as well as other species ) ( Figure S5 ) . If CAR binding plays a role in CAV-2 agglutination , then knocking down/blocking or artificially expressing CAR on erythrocytes should prevent/induce hemagglutination . Eliminating CAR on mature enucleated erythrocytes is technically challenging . Furthermore , to the best of our knowledge there are no known anti-CAR antibodies that completely block Ad attachment . On the other hand , CAR has been expressed on erythrocytes using a transgenic mouse line ( GATA1-CAR ) with the CAR cDNA downstream of the globin transcription factor 1 promoter [28] . Using flow cytometry and western blot analysis ( Figure 6A & B ) , we compared the levels of human CAR expressed by GATA1-CAR and human erythrocytes . We then repeated the hemagglutination assays using erythrocytes from control ( C57BL/6 ) and GATA1-CAR mice . We also found that CAV-2 agglutinated GATA1-CAR erythrocytes at a low particle-to-cell ratio ( Figure 6C ) , while there was no agglutination of C57BL/6 erythrocytes . Equally important , competition assays with recombinant CAV-2 fiber heads gave profiles that were similar to when we used human erythrocytes ( Figure 6D , and data not shown ) . Together , our data demonstrate that CAR expression by erythrocytes can lead to agglutination by CAR-tropic adenoviruses . In spite of recent notable advances [39] , [40] , in vivo Ad biodistribution , tropism and pathogenesis for the CAR-tropic HAds are still poorly understood . Group C HAd serotypes 2 and 5 are the prototype Ads in terms of structure , tropism and pathogenesis . However , tropism has been primarily studied in vitro , ex vivo or in animal models . Our results showing that human and rat erythrocytes harbor CAR on their external membranes while mice and nonhuman primates do not , suggests that these latter animals poorly mimic the in vivo environment that Ads encounter . To better address in vivo biodistribution of CAR-tropic Ads , we injected GATA1-CAR and control C57BL/6 mice with a HAd5 vector and quantified viral genome blood half-life and tissue distribution . We found that the viral load in the blood was 1000-fold higher ( P<0 . 01 ) in GATA1-CAR versus isogenic control mice ( CAR-negative erythrocytes ) during the first 72 hr post-injection ( Figure 7A ) . These data are reminiscent of the studies where human blood cells are routinely positive by qPCR for wild type HAd sequences [6] , [7] . In addition , notably absent from the AdGFP-injected GATA1-CAR mice was transgene expression in the liver . Lack of transgene expression was also consistent with the significant ( P<0 . 01 ) ∼25-fold difference in the mean viral load as quantified by qPCR ( Figure 7B ) . The lack of efficient liver infection is also consistent with the generally unexpected and poor infection of rat liver compared to mice and nonhuman primates following injection of CAR-tropic Ad vectors . Together , these data suggest that CAR expression by rat and human erythrocytes plays a significant role in HAd in vivo distribution and , in turn , determining which tissues are susceptible to infection .
We initiated this study to examine a 50-year-old enigma of adenovirus biology . Our results demonstrate that HAd37 interaction with human erythrocytes is primarily due to sialic acid binding via a conserved sialic acid binding site on this subgroup D HAd fiber head . CAV-2 , a serotype that like the prototype HAd5 is “CAR-tropic” and by most criteria is unrelated to HAd37 , interacts with human erythrocytes via a mechanism depending on several factors , including most notably binding to CAR . Most subgroup D HAd sialic acid binding site residues are conserved [27] , and structural analysis suggest that they bind sialic acid in an equivalent fashion . Unexpectedly , we found that the CAV-2 fiber head harbors a sialic acid binding site . But , in contrast to the well-conserved location of the CAR binding domains on some Ad fiber heads , location of the sialic acid binding site is not conserved . Following the structure-based identification of the amino acids in the CAV-2 and HAd37 fiber heads that interact with sialic acid , we introduced mutations to assay the role of these sites in sialic acid binding and erythrocyte interaction . In a number of conditions and approaches we showed that hemagglutination by HAd37 depends primarily on sialic acid binding . First , removing sialic acids from erythrocytes with neuraminidase eliminated erythrocyte cross-linking by HAd37 and a chimeric capsid harboring the HAd37 fiber head . Likewise , pre-incubation with neuraminidase significantly reduced hemagglutination caused by protein complexes containing multiple copies HAd37 fiber heads . Second , mutating single residues in the sialic acid binding sites reduced the hemagglutination activity of these protein complexes . Third , wild type HAd37 fiber heads bound more efficiently to erythrocytes or glycophorin than those carrying a point mutation in the sialic acid binding site . Our binding assays using wild type and mutated HAd37 heads also suggested that the sialic acid affinity is partially charge-driven . In addition , although HAd37 head also binds CAR [24] , in the context of the virus it does not appear to use CAR as a receptor , possibly due to its relatively short fiber . Therefore , CAR on erythrocytes is likely to be less important for HAd37 biodistribution . In spite of our structural data showing a well-defined sialic acid binding site , we could not detect notable binding between CAV-2 fiber head and glycophorin , a highly sialated protein . It was possible that the affinity of CAV-2 fiber head to erythrocytes or glycophorin was lower than that of HAd37 fiber head , and that the efficient erythrocyte binding of CAV-2 depended on the avidity of multiple fibers . As the CAV-2 fiber head is less positive than HAd37 fiber head ( pI's of 8 . 4 and 9 . 2 respectively ) , this could have explained the lower affinity . The affinity between virus proteins and sialic acid is usually in the millimolar range; therefore it was conceivable that the affinity of single CAV-2 fiber heads to erythrocytes or glycophorin was low . Consistent with this assumption , we previously found that CAV-2 is more neutrally charged than other Ads [32] . Paradoxically though , CAV-2 agglutinates at a lower particle-to-cell ratio than HAd37 and our crystallography data suggested that sialic acid binding should be at least as strong as HAd37 . However , one cannot reliably predict affinity from structural data . The identification of CAR expression by erythrocytes from species that are agglutinated by CAR-tropic Ads is a crucial observation . Using i ) competition assays with recombinant fiber heads harboring point mutations in the sialic acid and CAR binding sites and ii ) transgenic mice expressing CAR on erythrocytes , we characterized the unexpected and significant role of CAR in Ad binding . As mentioned previously , Ad serotypes differ in their ability to bind erythrocytes from various species . Our study suggests that this may be due to i ) an interaction with different chemical variants , linkages and ratios of sialic acid , ii ) the presence and affinity to CAR , and/or iii ) the pI's of the head [33] . For example , some non-CAR-tropic Ads that agglutinate human erythrocytes may preferentially bind Neu5AC , the most abundant sialic acid on human erythrocytes . With respect to charge , both HAd37 and HAd19p bind sialic acid with equal affinity and the only two residues that differ between their heads are not close to the sialic acid-binding site [33] . With the limited amount of data available , subgroup D HAds tend to have heads with higher pI's and interact with sialic acid [33] . Interestingly , the residues that make up the CAV-2 sialic acid binding site are conserved in the CAV-1 fiber head ( data not shown ) , suggesting a conserved sialic acid binding function . With respect to CAR affinity , the CAV-2 fiber head binds human CAR with the highest affinity of any fiber head known [24] , which may favor its efficient hemagglutination . In contrast to other Ads , higher temperatures poorly inhibit CAV-2 hemagglutination [12] , again suggesting a role for the high-affinity attachment to CAR . This is also consistent with a rather large difference between HAd5 and CAV-2 hemagglutination: to the best of our knowledge the HAd5 head does not harbor a sialic acid binding site and its pI ( 6 . 25 ) is less basic than HAd37 or CAV-2 . Finally , we cannot exclude a role of the inter- and intra-species differences in the quantity of CAR expressed by erythrocytes . Phylogenetically , CAR expression by erythrocytes appears to be random; we tested several other species ( dogs , mice , rabbits , lemurs and monkeys ) and did not find CAR expression . However , CAR levels on rat erythrocytes were relatively high ( Figure 6 ) and consistent with our hypothesis that CAR-tropic Ads agglutinates human and rat erythrocytes via CAR binding . It is possible , but unlikely , that our lack of detection of CAR on erythrocytes on some species was due to the limit of sensitivity of our bank of anti-CAR antibodies , which nonetheless detected CAR expression on other cell types from these species ( not shown ) . In the context of the significant amount of HAd5-mediated gene transfer data , our data highlights an important parallel . While a 30-gram mouse can be injected intravenously with up to 1012 pp of an HAd5 vector with minimal side effects , injection of the same dose in a 250-gram rat is normally lethal [41] . Whether our data have a bearing on the death following portal vein injection of a HAd5 vector during a phase I trial [42] is unknown , but deserves consideration . Similar to the CAR binding domain in HAd37 , we can only speculate about the importance of sialic acid binding in the biology of CAV-2 . Although CAV-2 does not agglutinate dog erythrocytes ( Figure S4 ) , Canis lupus familiaris may not be the original host of CAV-2: seroprevalence against CAV-2 can be found in coyotes , bears , pandas , skunks , mongooses , raccoons and foxes [43]–[45] . It is paradoxical that HAd37 binds CD46 , sialic acid and CAR - yet does not use CAR [25] - while CAV-2 binds both sialic acid and CAR and to the best of our knowledge uses only CAR as a receptor [5] . We cannot exclude the possibility that in some cell types sialic acid binding provides a first low-affinity attachment to the cell surface , while CAR-binding is followed in a second step , providing a high-affinity binding . Similar two-step mechanisms have been proposed for other viruses [46] . It is tempting to speculate that sialic acid binding may play a role in the preferential transduction of neurons by CAV-2 vectors [47]–[50] . Our crystal structure of CAV-2 fiber head in complex with CAR and sialyl-D-lactose demonstrates that the ternary complex of the three molecules is stable . There is no indication that this should not be the case also in vivo . Our data and numerous reports describing the ex vivo interaction of CAR-binding HAd5 with human/rat erythrocytes [12] , [21]–[23] , [29] , [36] , [51]–[53] strongly suggest this interaction probably occurs after the intravascular injection of CAR-tropic vectors . Notably , Lyons et al . showed that that >90% HAd5 vector DNA was associated with blood cells following intratumoral injection during a clinical trial [34] . It is likely that erythrocyte binding occurs during wild type infection of some HAds ( as well as coxsackie B viruses ) that use CAR , which will certainly lead to altered biodistribution . HAd DNA is routinely found in human blood cells by PCR . It is also a common misconception that Ads are rapidly cleared following a classical immune response . Numerous clinical cases strongly suggest that latent HAds can readily resurface if the host is immunosuppressed . The fate of particles that stick to erythrocytes under natural or artificial ( i . e . vector injections ) conditions is probably complex . For example , HAd5-induced liver disease in immunocompromised humans is relatively common . In contrast , immunocompromised nonhuman primates are rarely diagnosed with simian Ad ( SAV ) -induced liver disease [54] . Does CAR expression on erythrocytes lead to an advantage for host or virus ? Has CAR expression on erythrocytes put a selective pressure on Ads ( and coxsackie B viruses that also bind CAR ) to avoid an erythrocyte virus trap [28] ? Or has CAR expression by erythrocytes allowed HAds to thrive because it has allowed open access to so many more tissues and cell types ? In summary , our results resolve a longstanding enigma of Ad-erythrocyte interaction in vitro and in vivo . In addition , we provide new insights into virus-erythrocyte interactions that will allow us to better understand HAd pathogenesis and facilitate the engineering of safer , more efficient gene transfer vectors . Although in most in vivo scenarios hemagglutination per se is unlikely to occur due to the turbulence that erythrocytes encounter in the circulation , sequestering of vector particles by erythrocytes must diminish gene transfer efficacy . Interest in Ad biology is continually growing due to the increasing incidence of HAd-induced morbidity and mortality during immunosuppression [55] , [56] , and because Ad-derived vectors are the most commonly used vectors in gene therapy clinical trials . The identification of fiber head mutants that do not bind human erythrocytes may be of interest . Equally important may be the need to screen fiber heads from other serotypes for sialic acid and CAR binding domains . Combined with HAd interactions with vitamin K-dependent coagulation factors [39] , our study adds another critical element in the interaction with blood components , biodistribution and pathogenesis .
CAV-2 ( CAVGFP ) , HAd5 ( AdGFP ) , HAd37 , and the hybrid HAd5-CAV-2H ( Ad5Luc1-CK ) and HAd5-HAd37F ( Ad37f ) vectors were prepared as previously described [30] , [31] , [57] , [58] . Briefly , CAVGFP and AdGFP are E1-deleted vectors expressing GFP . The capsids contain no modifications . The hybrid HAd5-CAV-2H vector contains the fiber head of CAV-2 on the HAd5 capsid . HAd5-HAd37F vector contains the HAd37 fiber ( shaft and head ) on the HAd5 capsid . All vectors were purified by double banding on CsCl gradients , CsCl was removed using PD-10 columns ( Pharmacia ) . The vectors were stored in phosphate-buffered saline ( PBS ) containing 10% glycerol at −80°C . Stock titers were >1×1012 physical particles/ml with >1 infectious particle/5 physical particles . Anti-CAR antibodies tested in this study included E1 . 1 a monoclonal mouse ( S . Hemmi , University of Zurich ) , CAR1605 polyclonal rabbit ( J . Zabner ) , MoAbE ( mh ) 1 monoclonal mouse ( S . Carson , University of Nebraska ) , and AF2654 ( anti-mouse ) and AF3336 ( anti-human ) polyclonal goat ( R & D Systems ) . The recombinant fiber heads and dodecahedral sample were loaded between the mica-carbon interface as described [59] . The samples were stained using 2% sodium silico tungstate pH 7 . 5 and air-dried . Images were taken under low-dose conditions in an EX1200-II JEOL electron microscope working at 100 kV and with a nominal magnification of 40 , 000 . The images were scanned on a Z/I Imaging scanner ( Photoscan TD ) with a pixel size of 14 mm ( 3 . 5 Å per pixel at the sample level ) . Fiber head constructs were cloned into pPROEX HTb ( Life Technologies ) and expressed with a cleavable His-tag as described previously [24] . HAd37 fiber head constructs contain residues 177–365 and CAV-2 fiber head constructs contain residues 358–542 . Point mutations were introduced using the QuikChange Site-Directed Mutagenesis Kit ( Stratagene ) and polymerase chain reaction ( PCR ) . Protein purification was performed as described [24] . Briefly , cells were incubated in lysis buffer ( 20 mM Tris-HCl pH 7 . 5 , 300 mM NaCl , 20 mM imidazole , ( Boehringer Complete EDTA-free protease inhibitor cocktail ) , centrifuged , and fiber head bound to a Ni-NTA column ( Qiagen ) . Eluted protein was either directly loaded onto a Superdex200 column for hemagglutination experiments with tagged fiber head , or incubated overnight with 1/100 His-tagged tobacco etch virus ( TEV ) protease at 10°C . Proteins were then dialyzed against lysis buffer and uncleaved protein and TEV protease bound to a Ni-NTA resin . Untagged fiber head was loaded onto a Superdex200 column using the same buffer as for tagged protein ( 20 mM Tris pH7 . 5 and 300 mM NaCl ) . CAV-2HCAR− was labeled using Alexa488 Microscale Protein Labeling Kit ( Molecular Probes ) . CAV-2Hwt was dialyzed in PBS 0 . 1 M NaCO3 pH 9 . 3 and labeled using mono-reactive dyes ( Cy3 , Cy5 or Alex488 , Amersham Bioscience ) for 45 min at room temperature . The elution of labeled protein was performed with 2 ml of PBS using NAP5 column ( GE Healthcare ) pre-equilibrated with 10 ml PBS . The final dye/protein ratios ( ∼2 . 4 for each ) were determined using NanoDrop ND-100 spectrophotometer . Untagged CAV-2 fiber head was concentrated to 17 mg/ml in crystallization buffer ( 150 mM NaCl , 20 mM Tris pH 7 . 5 , 8 mM sialyl-D-lactose ) and crystallized in hanging drops containing 1 µl protein solution and 1 µl well solution ( 5% PEG 4000 , 5% isopropanol , 0 . 1 M HEPES pH 7 . 4 ) . Single crystals were transferred and frozen in a cryoprotectant solution ( 65% well solution , 25% glycerol , approximately 80 mM sialyl-D-lactose ) . The sialyl-D-lactose ( Sigma ) corresponds to α2-3 N-acetylneuraminosyl-D-lactose ( according to the supplier ) . The cloning , expression and crystallization of HAd37 and CAV-2 fiber head CAR D1 complexes ( space groups I23 and I422 ) was performed as described previously [24] . After crystal growth , sialyl-D-lactose ( Sigma ) was added to the drop to a final concentration of 10 to 50 mM . Crystals were frozen the next day without adding cryoprotectant solution to crystals with CAV-2 fiber head complex , and with 20% glycerol in the crystallization condition for crystals with HAd37 fiber head complex . Details of the data collection and refinement of the three structures ( CAV-2+sialyl-lactose , CAV-2+CAR D1+sialyl-lactose and HAd37+CAR D1+sialyl-lactose ) are given in Table S1 . All crystallographic data were collected at the European Synchrotron Radiation Facility ( ESRF ) and processed with XDS [60] . All structures were solved by molecular replacement using PHASER [61] and refined with REFMAC [62] . COOT [63] was used for the visualization of all models and electron density maps , and for the superposition of different models . PROCHECK [64] and MolProbity [65] were used for the validation of the obtained models . Representations of the protein models and the electron density were made with PYMOL [66] and GRASP [67] . Binding interfaces were visualized with DIMPLOT [68] . The structure-based sequence alignment was made with SARF , modified manually in SEAVIEW [69] and visualized with ESPript [70] . The CAV-2+sialyl-lactose structure contains two fiber head trimers in the asymmetric unit . The CAV-2+CAR D1+sialyl-lactose contains four trimeric fiber head - CAR D1 complexes in the asymmetric unit . Due to the modest resolution of this structure , NCS restraints and TLS refinement was used . The HAd37+CAR D1+sialyl-lactose structure has one fiber head monomer with bound CAR D1 in the asymmetric unit . Chimeric mini-fiber constructs were cloned by PCR in two steps . First , DNA fragments were produced using as template HAd3 genomic DNA ( for fragment 1 ) or pPROEX HTb vectors coding for wild type or mutant HAd37 fiber head ( for fragments 2 and 3 ) . Fragment 1 coded for HAd3 tail and shaft motives 1+2 ( primers were AAT AAT CCA TGG CCA AGC GAG CTC GG and GTT CCA TGC TAC CAA GGA TCC ATC AGT AG ) . Fragments 2 and 3 coded for wild type and mutant ( K345E ) HAd37 fiber head ( primers used were CTA CTG ATG GAT CCT TGG TAG CAT GGA AC and AAT AAT GAA TTC TCA TTC TTG GGC AAT ATA GG ) . In a second PCR reaction , fragment 1 was annealed to fragment 2 or 3 using primers AAT AAT GAA TTC TCA TTC TTG GGC AAT ATA GG and AAT AAT CCA TGG CCA AGC GAG CTC GG . The resulting longer fragments coded for complete mini-fiber containing HAd3 tail+shaft and either wild type or mutant ( K345E ) HAd37 fiber head . These were cloned into pPROEX HTb , and expressed at 25°C in E . coli strain BL21 Star ( DE3 ) ( Life Technologies ) together with an N-terminal cleavable 6×His-tag . Cells were re-suspended in lysis buffer and sonicated . The cell lysate was centrifuged for 30 min at 25000g and the supernatant loaded on a Ni-NTA column ( Qiagen ) . Protein bound to the resin was washed with 20 mM Tris-HCl pH 7 . 5 , 300 mM NaCl , 50 mM imidazole and eluted in 20 mM Tris-HCl pH 7 . 5 , 150 mM NaCl , 500 mM imidazole . To remove the His-tag , mini-fibers were incubated overnight with 1/100 His-tagged TEV protease at 10°C . Imidazole was removed by dialysis against lysis buffer . Uncleaved protein and TEV protease were removed by binding to a Ni-NTA resin . HAd3 penton forms multimers ( dodecahedra ) consisting of twelve penton bases each [9] , [71] . P . Fender ( Institut de Biologie Structurale , Grenoble ) generously supplied purified HAd3 dodecahedra . Mini-fibers and dodecahedra were mixed at an approximate molecular ratio of 1:10 and incubated at 4°C overnight . The mixture was loaded onto a Superdex200 column ( Amersham ) to remove excess fiber . The buffer contained 20 mM Tris pH7 . 5 and 300 mM NaCl . Dodecahedra in complex with mini-fiber were eluted with the void volume and were visualized with an electron microscope . All mini-fiber constructs carrying HAd3 tail+shaft and a CAV-2 fiber head were unstable , possibly because they were misfolded , or they did not bind to HAd3 dodecahedra . Non-treated , mock-treated or neuraminidase-treated erythrocytes and non-tagged fiber heads were incubated for 20 min at 4°C . Cell samples serving as negative control were incubated with PBS instead of fiber head solution . The cells were pelleted at 800g , re-suspended in PBS containing 1:100 rabbit anti-HAd37 fiber head serum ( gift from N . Arnberg , University of Umeå ) or 1:100 purified rabbit antibodies against CAV-2 fiber head , and incubated for 20 min at 4°C . The cells were pelleted as before , re-suspended in PBS containing 1:100 anti-rabbit FITC labeled antibody ( Sigma ) and incubated in the dark at 4°C for 20 min . The cells were pelleted again , re-suspended in PBS , and injected into a FACSCalibur apparatus ( BD Biosciences ) . The results were analyzed using CellQuest and FlowJo software ( BD Biosciences ) . Human , rat ( Wistar ) , dog ( beagle ) , vervet ( Chlorocebus pygerythrus ) , cynomologous monkey , rhesus macaque , guinea pig , rabbit ( New Zealand White ) , mouse ( C57BL/6 ) or GATA1-CAR [28] mouse blood was collected in EDTA , heparin or Alsever's solution . Erythrocytes were purified using Ficol gradients , washed twice in PBS EDTA ( 5 min at 2000 rpm ) and stored less than 48 hrs in PBS containing 5 mM EDTA . For the human and GATA1-CAR mouse erythrocytes two fractions of packed erythrocytes were re-suspended separately in neuraminidase buffer each . Neuraminidase α ( 2-3 , -6 , -8 and-9 ) from Arthrobacter ureafaciens ( Sigma ) or α ( 2-3 , -6 and -8 ) ( Ozyme ) was activated as recommended , and added to one of the two erythrocytes fractions . Both mock- ( without neuraminidase ) and neuraminidase-containing fractions were incubated for 1 hr at 37°C . The cells were then washed 5 times with PBS to remove neuraminidase and buffer , and re-suspended in PBS containing 5 mM EDTA . Virus , dodecahedra ± chimeric mini-fibers and CAV-2 fiber head multimers were diluted with PBS in 10-fold ( in the first column ) and then 2-fold ( horizontal ) steps and added to 96-well plates with cone-shaped bottoms . An equal number of purified erythrocytes ( either non-treated , neuraminidase-treated , or mock-treated ) was added to each well and incubated at room temperature or 37°C for at least 3 hrs . The blocking protein ( ∼1 µg/well ) was incubated with erythrocytes for 1 h at 4°C with slow rotation . All assays were repeated at least 3 times . In a volume of 100 µl , CAVGFP or AdGFP ( 1 particle/cell ) was incubated with 2 . 5×107 mock-treated or with neuraminidase-treated erythrocytes with PBS , or PBS supplemented with NaCl to bring its final concentration to 225 or 400 mM . The samples were incubated at room temperature for 15 min , and then centrifuged for 5 min at 5000 rpm in a microfuge . Aliquots of the supernatant were removed and incubated with 1×105 cells in 12-well plates . The cells were trypsinized and assayed for GFP expression by flow cytometry 24 hrs post-incubation . Data were analyzed using CellQuest . The assays were performed twice and in quadruplicate . Control cells and tissues were lysed with 100 µl SDS buffer ( 106 cells ) and benzonase for 1 h at 37°C . The liver , peripheral blood mononuclear cells , NIH 3T3 cells ( mouse fibroblasts ) and 293 cell ( human embryonic kidney ) extracts ( 40 mg/ml ) were resuspended in 100 µl of SDS-sample buffer . Erythrocytes ( ∼200 µl ) were lysed in ddH2O , the membranes were pellet for 10 min at 14 , 000 RPM in a microfuge and resuspended in 100 µl of SDS-sample buffer . SDS-PAGE was performed using a 5% acrylamide/bis-acrylamide stacking gel and a 12% acrylamide/bis-acrylamide running gel . Membranes were blocked with TBS-Tween , 10% milk at room temperature . The rabbit anti-CAR antibody CAR1605 was diluted 2000-fold for 1 h at RT in TBS-Tween 10% milk . The secondary anti-rabbit antibody was used at a dilution of 1/5000 for 30 min at room temperature in TBS-Tween 10% milk . Four adult GATA1-CAR and four control C57BL/6 mice were injected with 1 . 2×1011 pp of AdGFP via the tail vein . Blood ( ∼100 µl ) was taken by tail vein bleeds at 0 . 25 , 6 , 24 and 72 hr . The mice were sacrificed at 72 hr by lethal injection , and the organs were perfused with PBS via cardiac puncture . The liver , lung and spleen were recovered , divided into parts for qPCR or histology . The organs used for histology were fixed in 4% PFA for 24 hr then placed in 20% sucrose for 24 hr , and embedded in OCT matrix ( CellPath , Powys , UK ) . Sections ( 10-µm-thick ) were stained with 0 . 2 µg/ml bisBenzimide Hoechst ( Sigma-Aldrich ) and 1 ng/ml phalloidin-TRITC ( Sigma-Aldrich ) before being mounted . Images were acquired using a Zeiss microscope and processes using the MetaMorph ( Molecular Devices , Wokingham , UK ) . The experimental protocols involving animals were approved by the University of Massachusetts Medical School Institutional Animal Care and Use Committee . Total DNA from blood and liver were extracted by using the High Pure DNA Isolation kit ( Roche Diagnostics ) . qPCR was performed with a Light Cycler ( Roche Diagnostics ) using the Platinum Taq DNA polymerase ( Invitrogen ) and SYBR Green qPCR master mix [72] . The primer pairs used for GAPDH were: GAPDH forward , 5′ ACA GTC CAT GCC ATC ACT GCC 3′; GAPDH reverse , 5′ GCC TGC TTC ACC ACC TTC TTG 3′; and the EGFP: forward , 5′ CAG AAG AAC GGC ATC AAG GT 3′; eGFP reverse , 5′ CTG GGT GCT CAG GTA GTG G 3′ . Data are expressed as a ratio of GAPDH to EGFP . Data were analyzed using a one-way ANOVA and post-hoc comparisons were made using an unpaired Student's t-test .
|
In most cases , adenoviruses are thought to initially enter the host via contact with epithelial cells and spread within the host via an unknown mechanism . Most adenovirus serotypes use a cell adhesion molecule dubbed “CAR” to attach to cells . To assess , predict and understand adenovirus biology and vectorology , many in vivo studies use mice and monkeys . These animal models have been considered reliable models in the realm of viral pathogenesis and gene transfer . One of the implications of our study suggests that the rat may be a more appropriate model during intravenous adenovirus delivery because like humans , and unlike mice and monkeys , they also express CAR on their erythrocytes . The identification of CAR on human erythrocytes explains a 50-year-old enigma of adenovirus hemagglutination , helps us better understand adenovirus in vivo biology and may open new avenues to understand the role of cell adhesion molecules during erythropoiesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry/molecular",
"evolution",
"virology/persistence",
"and",
"latency",
"virology/virus",
"evolution",
"and",
"symbiosis",
"virology/virion",
"structure,",
"assembly,",
"and",
"egress",
"infectious",
"diseases/viral",
"infections",
"biochemistry/biomacromolecule-ligand",
"interactions",
"virology/immune",
"evasion",
"hematology",
"virology/host",
"invasion",
"and",
"cell",
"entry"
] |
2009
|
The Cell Adhesion Molecule “CAR” and Sialic Acid on Human Erythrocytes Influence Adenovirus In Vivo Biodistribution
|
Recently , the heterogeneity that arises from stochastic fate decisions has been reported for several types of cancer-derived cell lines and several types of clonal cells grown under constant environmental conditions . However , the relation between this stochasticity and the responsiveness to extracellular stimuli remains largely unknown . Here we focused on the fate decisions of the PC12 cell line , which was derived from rat pheochromocytoma , and is a model system to study differentiation into sympathetic neurons . Whereas epidermal growth factor ( EGF ) stimulates the proliferation of populations of PC12 cells , nerve growth factor ( NGF ) promotes the differentiation of neurites to neuron-like cells . We found that phenotypic heterogeneity increased with time at several surrounding serum concentrations , suggesting stochastic cell-fate decisions in single cells . We made a simple mathematical model assuming Markovian transitions of the cell fates , and estimated the transition rates based on Bayes' theorem . The model suggests that depending on the serum concentration , EGF ( NGF ) even directs differentiation ( proliferation ) at the single-cell level . The maximum effects of the growth factors were ensured when the transition rates were appropriately controlled by the serum concentration to produce a nonextremal , moderate amount of cell-fate heterogeneity . Our model was validated by the experimental finding that the means and variances of the local cell densities obey a power-law relationship . These results suggest that even when efficient responses to growth factors are observed at the population level , the growth factors stochastically direct the cell-fate decisions in different directions at the single-cell level .
Phenotypic heterogeneity , which has been thoroughly discussed for tumor cells , is not a unique property of cancerous cells but has also been observed in normal clonal cells in culture [1] . Differences in tumor cells have been attributed to differences in cell lineages that arise from genetic or epigenetic processes . However , even clonal cells show phenotypes that are certainly not identical because they are subject to various sources of stochasticity other than genetic heterogeneity [2] . Recent insights have suggested that molecular ‘noise’ caused by fluctuations in gene expression , signal transduction , and other processes affects phenotypic heterogeneity in organisms ranging from microbes to mammals [3] . In the presence of such noise , even cells with the same overall phenotypic profile fluctuate randomly , causing them to have subtly different phenotypes at any particular time . This is a key mechanism that generates the cellular diversity that is sometimes used as bet-hedging of bacterial persistence in Escherichia coli or competence in Bacillus subtilis [4] , [5] . In mammals , several types of cells use stochastic decision-making to regulate development [6]–[10] . Although heterogeneity was observed in several types of cells , the effects of extracellular stimuli on those heterogeneous populations have not been fully clarified . Here , we used PC12 cells to attempt to address this issue . The rat pheochromocytoma clone PC12 , which was developed from an adrenal medullary tumor derived from the adrenergic neural crest [11] , has been used as a model of neural differentiation . Healthy PC12 cells can be grown under appropriate content percentage of serum in a medium for culture , and they have several properties that resemble those of adrenal medullary chromaffin cells [12] . In the presence of nerve growth factor ( NGF ) , PC12 cells stop dividing , display electrical excitability , produce neurite-like outgrowths , and differentiate into cells with a sympathetic neuron-like phenotype . Although the removal of NGF for sympathetic neurons leads to cell death , the differentiation of PC12 cells appears to be reversible insofar as removal of NGF causes them to lose the properties acquired after differentiation . It has been reported that the effects of NGF on differentiation are efficient under serum-starved conditions [13] . The epidermal growth factor ( EGF ) receptor is also expressed in PC12 cells [12] . As in other cell types [14] , EGF acts as a mitogen in PC12 cells [15] . However , the effect of EGF can be masked by culture conditions , especially the content percentage of serum [15] . The response of PC12 cells to growth factors is heterogeneous on the level of individual cells and is affected by the surrounding serum concentration , but the relationship between cell heterogeneity and cell responsiveness to growth factors has not been measured quantitatively . Here , we report the effects of the growth factors EGF and NGF under three different serum conditions , in which PC12 cells show different degrees of heterogeneity in their fate decisions . We measured the time courses of the numbers of cells in three states ( proliferating , differentiated , and dead ) and constructed a mathematical model . By definition , we regard concentrations of surrounding serum as an environmental condition , and regard cell responses as meaning changes in cell fate triggered by stimuli , such as EGF and NGF , at that particular serum concentration . In this study , we first used entropy values to define the heterogeneity of a population of cells containing a large fraction of proliferative cells , and found that the entropy values increased as a function of time , suggesting that stochastic cell-fate decisions in single PC12 cells increased the heterogeneity of the population , regardless of the surrounding serum . This heterogeneity decreased as the concentration of the surrounding serum increased . We made a simple mathematical model assuming that cells determine their fates with constant probabilities , and estimated the probabilities for which the model can explain the experimental results , based on Bayes' theorem . This method enabled us to use experimental data collected using populations of cells to estimate the rates with which single cells made cell-fate decisions . Based on the model and the parameter values , we redefined the effects of the growth factors: EGF increased the proliferation rate at the single-cell level , although the effect could be covered and was affected by the surrounding serum at the population level , as shown in previous experiments . At some serum concentrations , EGF ( NGF ) even directed differentiation ( proliferation ) at the single-cell level . Even when efficient responses to the growth factors were observed at the population level , the growth factors only stochastically directed single cells to different cell fates . We evaluated the strength of the responses to the growth factors at the population level on a phase portrait using the Malthus coefficient and the fluxes of the three phenotypes , and examined the relationship between cell heterogeneity and cell responsiveness . The strength of the responses to EGF and NGF as a function of entropy ( cell-fate heterogeneity ) peaked at a moderate entropy value . Finally , we showed that the relationship between the means and variances of the local cell densities obeyed a power-law relationship , which could be explained by a stochastic simulation that supported the idea that the cells have a single set of transition rates with constant values under each condition .
In this section , we first report experimental results that describe dynamic changes in the number of cells at the population level . Next , we propose the use of a mathematical model to calculate transition rates at the single cell level from experimental results . The parameter values estimated in the previous section denote single-cell-level transition rates . However , at the population level , the effects of these parameters are very complex ( see Models section ) , and it is difficult to evaluate the effects of the growth factors on the population dynamics from the estimated parameter values . In this section , we introduce a method to use a phase portrait to capture dynamic changes in the number of cells at the population level , and to compare the responses to growth factors under three serum conditions . We use this portrait to quantitatively characterize the responses to growth factors . Finally , we evaluate the relationships between the heterogeneity of a population and the response indexes . Here , we focused on the variances of the cell densities in a culture dish . First , we show the relationships between means and variances in the experimental results , which exhibits power-law relation . Second , we extend our model in order to characterize the stochastic properties of the number of cells , and validate the assumption made in the previous sections that parameter values are constant . We also exclude the possibility that a variety of cells with distinct rate constants explain the observed variances in cell densities . We analyzed the time courses of the cell fate decision processes of PC12 cells after the addition of either of the growth factors EGF or NGF under three different environmental serum conditions . The results are summarized in Figure 8A–C . The population of cells became heterogeneous under all of the conditions tested , and high concentrations of serum suppressed the level of heterogeneity . The effects of growth factors depended on the environmental serum conditions , with each of the two growth factors affecting several transition pathways . Using a mathematical model , we could derive the effects of growth factors at the single-cell level . Stochastic single-cell responses to growth factors induced differentiation following exposure to EGF and proliferation following exposure to NGF . The use of phase portraits to capture dynamic changes in cell-fate decisions at the population level enabled us to evaluate the effects of serum concentrations and growth factors , and to discover conditions that promote efficient responses to growth factors . Moreover , we found that responses to growth factors were efficient when an appropriate concentration of serum induced a population with a moderate degree of heterogeneity . Finally , we have demonstrated a power-law relation between the means and variances of the local cell density . Stochastic simulation of our model could explain these results and support the validity of our model .
Rat PC12 pheochromocytoma cells from the Riken Cell Bank ( Tsukuba , Japan ) were cultured and maintained at , in Dulbecco's Modified Eagle Medium ( DMEM , containing glucose ) supplemented with horse serum ( HS ) and fetal bovine serum ( FBS ) . Cells were transferred to a 60-mm culture dish with a density of . One day after the transfer , the medium was exchanged for DMEM without phenol red and supplemented with three different concentrations of serum: ( i ) HS and FBS ( high-serum condition ) , ( ii ) HS and FBS ( low-serum condition ) , and ( iii ) no serum but supplemented with bovine serum albumin ( BSA ) ( serum-free condition ) . Two days after subculture , cells were treated with of mouse 2 . 5S nerve growth factor ( NGF ) ( final concentration; Alomone Labs . , Jerusalem , Israel ) , of recombinant murine epidermal growth factor ( EGF ) ( final concentration; Peprotech , London , UK ) , or of Hank's balanced salt solution for controls . The medium and the reagents were exchanged every second day . To count the densities of proliferating , differentiated , and dead cells , thirty images of living PC12 cells within thirty areas of or in size were taken for each dish every day after subculture using a phase-contrast microscope . The states of cells were determined from the morphologies in the captured images , with proliferating cells identified by their rounded shapes and their failure to extend neurites . The differentiated cells were defined by extension of at least one neurite with a fiber length longer than the diameter of the cell body . Dead cells were identified as shrunken or fragmented cell bodies . Examples of the micrograph of the three states of cells are shown in Figure 1A . Four independent experiments were done and the results of a typical experiment from the four are shown in Figures 2 and 7 . We used the parameter set estimated from the typical experiment ( Table S1 ) to prepare the data shown in Figure 3 , and Figure S5 . The average values of estimated parameters or fluxes calculated from the four independent experiments were used to prepare Figures 4 , and 5 . In addition to using the -values determined using a -test , we used effect size to compare two groups of data: ( 2 ) where denotes a mean for treatment group and denotes a mean for control group , and and denote standard deviations for them . This value compares distance of mean values of two groups with mean values of standard deviations of them . When the value of is large , the difference between two groups is large compared with the standard deviations . Significant values of effect size are arbitrarily defined depending on research fields , although the values of for the large effect size , for the middle effect size , and for small effect size are usually used [25] . Analysis of variance was used when more than three population averages were available for comparison . We assumed that there were levels or conditions of experiments ( ) and that experiments were repeated times in each level . We defined the value as th data in a level . For example , a level denoted each serum condition , and we repeated times in each level . The effect size was defined as followswhere denoted the between-groups sum of squares , and denoted the total sum of squares:where denoted the total number of experiments . The significant effect size was also arbitrarily defined , but the values of for the large effect size , for the middle effect size , and for small effect size were the typically used criteria , as in the effect size [25] .
|
Elucidation of the mechanisms that regulate cell fate has become one of the primary goals of research in cell biology and regenerative medicine . Growth factors are often used to regulate cell fate . However , stochastic cellular responses to growth regulators have prevented precise control of cell fate . We report our investigation of the relationship between heterogeneity and responsiveness in cell fate decisions by both single cells and populations of cells . Our study involved PC12 , a cultured cell line for which cell-fates are affected by exposure to growth factors and culture conditions . Computational methods using a mathematical model enabled us to determine the cell-fate decisions rate in single PC12 cells and analyze the population responses to growth factors from experimental data . Our findings reveal that growth factors control cell-fate decisions rate in single PC12 cells , and suggest distinct differences in the mechanisms of actions of growth factors under different culture conditions . In addition , we observed maximum effects of growth factors when a nonextremal , moderate amount of cell-fate heterogeneity exists . Our results give several insights into stochastic cell responses , including the effects of anticancer agents on cancer cells and the optimization of methods to induce the differentiation of stem cells .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Optimality Conditions for Cell-Fate Heterogeneity That Maximize the Effects of Growth Factors in PC12 Cells
|
Spermatogenesis is a key developmental process allowing for a formation of a mature male gamete . During its final phase , spermiogenesis , haploid round spermatids undergo cellular differentiation into spermatozoa , which involves extensive restructuring of cell morphology , DNA , and epigenome . Using mouse models with abrogated Y chromosome gene complements and Y-derived transgene we identified Y chromosome encoded Zfy2 as the gene responsible for sperm formation and function . In the presence of a Zfy2 transgene , mice lacking the Y chromosome and transgenic for two other Y-derived genes , Sry driving sex determination and Eif2s3y initiating spermatogenesis , are capable of producing sperm which when injected into the oocytes yield live offspring . Therefore , only three Y chromosome genes , Sry , Eif2s3y and Zfy2 , constitute the minimum Y chromosome complement compatible with successful intracytoplasmic sperm injection in the mouse .
Y chromosome has always been considered a symbol of maleness as it encodes testis determining gene Sry which acts in the developing gonads and induces the development of testes rather than ovaries [1–3] . Mammalian Y chromosomes encode a number of other genes most of which are thought to be involved in various aspects of male reproduction , and other playing roles of broadly expressed regulators of transcription , translation and protein stability [4] . In spite of these clearly important functions , the knowledge linking the roles of specific Y chromosome genes to specific reproductive processes remains limited . We recently investigated spermatogenesis progression and germ cell function in male mice with significantly abrogated Y chromosome complements [5] . We have shown that males with the Y chromosome contribution provided by two transgenes , the testis determinant Sry and the spermatogonial proliferation factor Eif2s3y ( Fig 1B , XEOSry and XEY*XSry ) have meiotic and postmeiotic arrest , the rare spermatids present in the testes do not elongate , and sperm are not formed . When round spermatids from these males were injected into the oocytes , live mouse progeny were obtained . The success of round spermatid injection ( ROSI ) was low , with less than 10% of transplanted embryos developing to live offspring . Interestingly , when the Sry transgene was replaced with the Y chromosome derived sex reversal factor Sxrb , encoding for Sry , H2al2y , Prssly , Teyorf1 , Rbmy gene cluster , and Zfy2/1 fusion gene ( Fig 1A , Sxrb ) the resulting males ( Fig 1B , XESxrbO and XESxrbY*X ) had more advanced spermatid development with clear elongation of these cells , occasional appearance of sperm , and increased ROSI efficiency . These findings indicated that a gene/s encoded within Sxrb plays a role in spermiogenesis progression and germ cell function . Here , we identify Zfy2 as the gene responsible . We present the evidence that the Y chromosome gene Zfy2 promotes sperm morphogenesis , improves ROSI success , and is necessary for a formation of sperm capable of yielding live offspring after intracytoplasmic injection into the oocytes .
The presence of Sxrb enables spermatid elongation in XESxrbO and XESxrbY*X males , with occasional development of mature testicular sperm [5 , 7] ( S1 Fig ) . To test for the ability of these testicular sperm to participate in successful assisted fertilization and embryo development , we performed intracytoplasmic sperm injection ( ICSI ) . No live offspring were obtained from XESxrbO males ( n = 4 males , 0/94 fetuses from embryos transferred ) , while ICSI with sperm from XESxrbY*X males yielded a single fetus ( n = 5 males , 1/84 fetuses from embryos transferred ) . Thus , sperm from XESxrbO and XESxrbY*X males are virtually not successful in assisted fertilization . A possible reason for the lack of live offspring from these sperm could be sperm diploidy . We have previously shown that the great majority ( 86% ) of round spermatids from XESxrbO males were diploid while the opposite was true for XESxrbY*X males , in which haploid round spermatids predominated ( 71% ) [5] . To test whether testicular sperm from these males carried doubled DNA content we performed zygotic chromosome analysis after sperm injection ( Table 1 , S2 Fig ) . This analysis demonstrated that only about one-fourth of the embryos obtained after ICSI with sperm from XESxrbO males were diploid ( 26% , 12/46 , Table 1 ) ; the remaining zygotes were triploid and thus presumably derived from diploid sperm . Zygotes obtained after ICSI with sperm from XESxrbY*X males were predominantly diploid ( 64% , 9/14 , Table 1 ) . These data support that while sperm diploidy might be responsible for the lack of ICSI success with XESxrbO , it is not likely the case with XESxrbY*X males . We have shown earlier that testicular sperm from males carrying the Y chromosome derived sex reversal factor Sxra ( Fig 1A & 1B , Sxra , XY*XSxra ) are haploid [8] . Thus , functional deficiency of sperm from XESxrbY*X males is likely due to lack of one or more Y chromosome genes that are present in XY*XSxra and not in XESxrbY*X . Overall , the data demonstrate that testicular sperm from XESxrbO and XESxrbY*X males are not functional in assisted fertilization and that this may reflect lack of certain Y gene/s . We next investigated which of the Sxrb genes is responsible for spermatid elongation . The gene content of Sxrb is represented by few copies of Rbmy , two copies of H2al2y , one copy of Sry , Prssly and Teyorf1 , and a Zfy2/1 fusion gene spanning the Sxrb deletion breakpoint ( Fig 1A ) . Rbmy appears early during spermatogenesis and is not expressed , and certainly not translated , after the zygotene stage [9] . H2al2y has been shown to be expressed late during spermiogenesis [10 , 11] and Sry transcripts in adult gonads are thought to be aberrant and not translatable [12 , 13] . Prssly and Teyorf1 are newly discovered genes [6] whose expression has not been characterized; we were not aware of these genes existence when the study was initiated . Based on the known expression pattern , we excluded Rbmy , H2al2y , and Sry as the candidates for ensuring sperm head and tail morphogenesis in males with Sxrb , and focused our attention on the Zfy2/1 fusion gene . Postnatal expression of Zfy1 and Zfy2 is restricted to spermatogenic cells [14–16] . Both genes are first expressed in the testis around the time when cells enter meiosis . They then undergo transcriptional silencing as the cells enter pachynema . The expression is reactivated in secondary spermatocytes and continues postmeiotically [11 , 17] . In round spermatids there is a clear predominance of Zfy2 transcripts in Y-bearing round spermatids; this strong expression appears because of the activity of 'acquired' strong Cypt-derived spermatid-specific promoter driving Zfy2 expression [17] . The CYPT exon of Zfy2 , encoding the Cypt1 promoter is thought to be derived from the Cypt1 gene [14 , 17] , which belongs to the CYPT spermatid-specific gene family [18] . The expression of Zfy1 , which lacks the Cypt promoter , is limited at the round spermatid stage . The Zfy2/1 fusion gene is driven by the Cypt promoter and is strongly expressed in spermatids . Our expectation therefore was that Zfy2 , and not Zfy1 , would mimic the effect of Sxrb . To assess Zfy2 role in spermatogenesis progression we investigated testis histology in XEY*XSry males transgenic for Zfy2 ( Fig 2 and S3 Fig ) . These males are subsequently called XE , Z2Y*XSry ( Fig 1B , S1 Table ) . While in XEY*XSry males spermatid development did not progress beyond the round spermatid stage , in XE , Z2Y*XSry males spermatids elongated ( Fig 2A ) . The elongated , condensed spermatids were more frequently observed in XE , Z2Y*XSry than in XESxrbY*X males; in the latter genotype elongation often ceased earlier ( at step 12–13 ) and the spermatid nuclei were less compacted , with lighter staining pattern . Quantitative analysis of spermatogenesis progression ( Fig 2B ) demonstrated that XE , Z2Y*XSry had more round spermatids than XEY*XSry ( ~2 . 8-fold increase ) , reaching a level similar to that observed with XESxrbY*X males , but less than in wild-type controls . The number of elongating/elongated spermatids in XE , Z2Y*XSry and XESxrbY*X was not significantly different , and ~10-fold lower than in wild-type controls . In the quantitative analysis of testis sections we did not distinguish between the elongating and elongated spermatids because the abnormal morphology of developing spermatids , which ultimately resulted in severely morphologically abnormal headshape of testicular and epididymal sperm , made such distinction impossible . Sperm from XE , Z2Y*XSry males were also observed in live epididymal and testicular cell suspension , with and on silver stained testicular cell spreads ( S4 Fig ) . The epididymal sperm were extremely rare; only few immotile sperm were noted in 4 out of 5 males . The headshape of both testicular and epididymal sperm was abnormal , as expected from males lacking Y chromosome long arm [19] . To characterize structural sperm defects in more detail we performed the analysis of sperm headshape on silver stained testicular cell spreads ( Fig 3 ) . Only sperm with fully developed tails were included in this analysis . In XY males , the great majority of testicular sperm were normal ( 84% , 31/37 ) , with remaining having slight headshape defects , comparable to those noted earlier in epididymal sperm [19] . In XY*XSxra , XESxrbY*X and XE , Z2Y*XSry males all sperm were morphologically abnormal . We divided the observable headshape defects into 7 categories ( A-G ) ( Fig 3B ) and quantified their incidence ( Fig 3A ) . In XE , Z2Y*XSry males sperm heads were either oval or rounded in shape , with no hint of a hooked tip , and frequently highly condensed ( categories D and E , 61% ) , or were elongated with no curvature reminiscent of crescent shape typical for mouse sperm , and occasional hint of a hooked tip ( category B , 39% ) . Sperm in XY*XSxra males , which similarly as XE , Z2Y*XSry lack Y chromosome long arm , had better developed heads than in XE , Z2Y*XSry males , with predominance of sperm with clear head elongation with or without curvature , and with and without a marked hooked tip ( category A and B , 66% ) , suggesting that presence of additional Y genes within Sxra facilitates head restructuring . In XESxrbY*X males the great majority of sperm were scored as elongated but poorly condensed ( category G , 74%; this category was specific for this genotype ) . The tail development in all examined genotypes was normal ( see S4 Fig for images of whole testicular sperm from XE , Z2Y*XSry males ) . Altogether , our data support that presence of Zfy2 enables round spermatids to initiate and undergo head morphogenesis and complete tail development . The Zfy2 in XE , Z2Y*XSry makes this transition much more effectively than Zfy2/1 in XESxrbY*X but does not reach the level observed in males with Y gene contribution provided by Sxra . In all genotypes this restructuring does not proceed normally and yields sperm with severely amorphous heads . In our recent study we have shown that ROSI success with XEOSry and XEY*XSry males was below 10% ( 9% and 6% , respectively ) while with XESxrbO and XEY*XSxrb males ROSI efficiency increased by ~2–2 . 5 fold ( 20% and 16% , respectively ) , suggesting that a gene/s encoded within Sxrb provides some benefit for assisted reproduction success [5] . Considering the Zfy2 role in meiotic progression [11] and spermatid elongation ( Figs 2 and 3 , S3 & S4 Figs ) we decided to test whether Zfy2 is beneficial for germ cell function . When ROSI was performed with round spermatids from XE , Z2OSry and XE , Z2Y*XSry males ( Fig 1B , S1 Table ) live offspring rate increased , reaching 27% and 43% , respectively ( Table 2 ) . Sperm from XESxrbO and XESxrbY*X males were not functional in assisted fertilization , and for XESxrbY*X males it could be attributed to Y gene deficiency . To test whether Zfy2 influences ICSI outcome , we performed injections with sperm from an XE , Z2Y*XSxrb male , which had both Sxrb ( encoding the Zfy2/1 fusion gene ) and the Zfy2 transgene ( Fig 1B ) . We had only one male of this genotype available as sperm donor for ICSI , and only 7 embryos were transferred but those yielded 2 live offspring ( 29% , 2/7 ) . Encouraged by this result we moved on to test sperm from XE , Z2Y*XSry males . Out of 5 males examined , 4 had testicular sperm and yielded live ICSI offspring ( Table 2 ) . The efficiency of ICSI with sperm from XE , Z2Y*XSry males was lower than with sperm wild-type XY controls ( 23% vs . 57% , P<0 . 001 ) . Because each XE , Z2Y*XSry and XY male provided both round spermatids ( ROSI ) and sperm ( ICSI ) for injections , we were able to perform a direct comparison of these two types of germ cells for their ability to participate in successful assisted fertilization . In XY males , as expected , the efficiency of ROSI was lower than ICSI ( Table 2 , 30% vs . 57% , P<0 . 01 ) . Interestingly , this pattern was reversed in XE , Z2Y*XSry males , in which ROSI success was significantly better ( Table 2 , 43% vs . 23% , P<0 . 01 ) . XE , Z2Y*XSry males generate several types of gametes , which consequently lead to several possible progeny genotypes . Genotyping of all progeny obtained after ROSI and ICSI revealed that anticipated offspring types were produced and their frequency met the expectancy , with 4 predominating genotypes accounting for 98 . 5% of all genotypes and distributed within 16%-31% range , and 1 rare genotype , which originated from atypical segregation of sex chromosomes ( Fig 4 and S2 Table ) . To correlate spermiogenic phenotype with Zfy expression we performed Zfy transcript quantification on whole testes from XEY*XSry ( no Zfy ) , XESxrbY*X ( Zfy2/1 fusion gene ) , XE , Z2Y*XSry ( Zfy2 transgene ) , with XY*XSxra ( endogenous Zfy2 and Zfy1 and no NPYq ) and XY ( intact Y chromosome ) serving as controls . Because Zfy2 and Zfy1 are very similar ( 97% and 94% for transcript and amino acid identity , respectively ) we failed to design real-time PCR primers that were specific to Zfy2 . We therefore quantified the expression of Zfy1 , and Zfy1 and Zfy2 combined ( global ) ( Fig 5 and S5 Fig ) . As expected , no Zfy1 transcripts were detected in XEY*XSry and XE , Z2Y*XSry males . XY*XSxra males had ~2-fold higher Zfy1 levels than XY males but the difference did not reach significance ( P = 0 . 08 ) . This minor increase is likely due to the lack of NPYq genes known to result in the upregulation of sex chromosome genes , including Zfy [20] . In XESxrbY*X males Zfy1 levels were ~10-fold and ~5-fold higher than in XY and XY*XSxra , respectively , representing a combined effect of the NPYq absence and activity of a strong Cypt-derived spermatid-specific promoter driving expression of the Zfy2/1 fusion gene . The global Zfy expression was again higher in XESxrbY*X and XY*XSxra than in XY but the difference was lower in magnitude . Zfy global levels in XE , Z2Y*XSry males were similar to those of XY*XSxra , and higher than in XY . When compared to XESxrbY*X , Zfy global levels in XE , Z2Y*XSry males were ~2-fold lower but the difference did not reach significance ( P = 0 . 1 ) . In XE , Z2Y*XSry males the global levels were reflective exclusively of Zfy2 , while in XESxrbY*X males primarily of Zfy1 since the Zfy2/1 fusion gene encodes Zfy1 coding region under the control of Zfy2 promoter [17] . When the spermiogenic phenotype is viewed in the context of Zfy expression our data support that it is Zfy2 , and not Zfy1 ( even if present in abundance ) , that enables the formation of sperm functional in ICSI .
Y chromosome encoded zinc finger protein genes , Zfy , have once been in the center of attention as potential candidates for the testis-determining factors [21–23] . When the fame went to another Y gene , Sry [1–3] , Zfy genes were quickly forgotten and it has taken more than two decades for these genes to re-emerge with newly ascribed spermiogenic roles . Zfy1 and Zfy2 were shown to play spermatogenic quality functions during the pachytene stage of meiosis and during MI by triggering the apoptotic elimination of spermatocytes [24 , 25] and to facilitate the second meiotic division [11] . It has also been shown that a gene/s from Sxra , partially retained in Sxrb , is necessary for the initiation of sperm morphogenesis [7] and increases the efficiency of round spermatid injection [5]; Zfy genes were proposed as the most likely candidates . Here we tested this assumption by investigating the effects of transgenic Zfy2 addition into Y chromosome deficient males , which have a postmeiotic arrest at the round spermatid stage . We demonstrated that Zfy2 is responsible for formation of sperm functional in assisted fertilization . In our previous study we reported that only two Y chromosome genes , the testis determinant factor Sry and the spermatogonial proliferation factor Eif2s3y , are sufficient to make a male mouse whose germ cells are functional in assisted fertilization ( ROSI ) and yield live progeny [5] . When Y chromosome contribution was expanded by substituting Sry for Sxrb , ROSI efficiency improved , and we speculated that this was due to the presence of the Zfy2/1 fusion gene , which facilitated the second meiotic division in the testis in the presence of Y*X , or in the oocytes after fertilization when the meiotic pairing partner was missing [5] . Spermatid elongation and occasional formation of mature testicular sperm were previously observed in males with Sxrb [5 , 7] but their function in fertilization has not been tested . Here we have shown that these sperm are not successful in assisted fertilization ( ICSI ) . In XESxrbO males the great majority of elongating spermatids are diploid [7] and so are the testicular sperm as shown in this study . The fact that live offspring were obtained with ROSI , and not with ICSI , could therefore be due to the highly condensed nature of the sperm chromatin . In diploid round spermatids from XESxrbO males the chromatin is still histone-bound and the homologous chromosomes are presumably still paired as in meiosis II . Secondary spermatocytes , with the same chromosomal state prematurely condense upon injection into oocytes and complete meiosis II along with the maternal chromatin , expelling a polar body-like structure with the haploid complement of paternal DNA [26] . We have proposed that a similar process occurs with the round spermatids from XESxrbO males , resulting in normal , diploid zygotes [5] . The diploid spermatozoa , however , cannot have the normal histone component because in order to complete spermiogenesis most of the histones would have had to be replaced by protamines . This sperm chromatin must be completely reorganized , which normally takes one to two hours after ICSI [27] . By this time , the maternal chromatin has already completed meiosis II , and the zygote can no longer support the completion of meiosis II for the paternal DNA . Congruent with this explanation , ICSI should be successful with sperm from XESxrbY*X males , which yielded predominantly diploid zygotes . However , only one offspring was obtained , suggesting that sperm ability to support embryonic and fetal development was highly impaired . We have previously shown that XY*XSxra males can be reproduced by ICSI [19] . Thus , one or more Y genes that are present and active in Sxra , and not in Sxrb , are likely to be responsible for rendering sperm functional . We now show that this gene is Zfy2 , and that sperm dysfunction in XESxrbY*X males can be overcome with the transgenic addition of Zfy2 . Why is it that Zfy2 renders sperm functional in assisted fertilization while the Zfy2/1 fusion gene does not ? The Zfy2/1 fusion gene , present within Sxrb , encodes a protein that is almost identical to that encoded by Zfy1 but the transcription is driven by a Cypt-derived Zfy2 specific promoter [17] . Both Zfy2 and Zfy2/1 are therefore strongly expressed postmeiotically because both have the Cypt-derived promoter , which drives strong expression in spermatids [14] . However , in the case of the Zfy2/1 fusion gene , the transcript produced is almost identical to that of Zfy1 . Alternative splicing results in the majority of Zfy1 transcripts lacking exon 6 , which encodes the ZFY protein transactivating domain ( TA ) , while most of the Zfy2 transcripts retain exon 6 [17] . The protein encoded by Zfy1 lacking exon 6 is expected to bind but not transactivate target genes and consequently can serve as a competitive inhibitor of full length ZFY proteins . Moreover , the TA domain in ZFY1 protein , when present , is ~5 . 5-fold less active than that of ZFY2 protein [11] . The Zfy2/1 fusion gene produces transcripts that are spliced like those of Zfy1 so that a substantial proportion of them lack the exon 6 encoding the TA domain transcription factor function [17] and when TA domain is present , it is equivalent to that encoded by ZFY1 and therefore less potent . These transcript and protein specific differences explain why Zfy2/1 in XESxrbY*X males is not sufficient for promoting sperm function , and why addition of Zfy2 to this genotype rescues this defect . The Zfy2 levels in XE , Z2Y*XSry males were ~1 . 9 fold higher than in XY . It therefore cannot be disregarded that the spermiogenic phenotype results from Zfy2 overexpression . The Zfy2 transgene was provided as a single copy Zfy2 BAC inserted by cassette mediated exchange ( CME ) into the Hprt locus on the X chromosome [24 , 25] . This transgene , because of its localization on the X chromosome , behaves in the same way as the endogenous gene , i . e . undergoes meiotic sex chromosome silencing ( MSCI ) [24] , and its level of expression should be equivalent to that of the endogenous gene . The overexpression observed in XE , Z2Y*XSry males is likely due to the lack of Y chromosome long arm genes , known to result in global upregulation of several sex chromosome genes , including Zfy2 [20] . The fact that we see Zfy overexpression in XY*XSxra males , which have endogenous Zfy1 and Zfy2 but lack the Y chromosome long arm , supports this case . To bring the Zfy2 expression to the physiological level XE , Z2Y*XSry males , we would need to provide the genes from the Y long arm but those would likely influence spermiogenic phenotype . Thus , with the current tools , we cannot test whether Zfy2 expression at physiological levels would also induce spermatid elongation and promote sperm function . The resolution can come from the analysis of Zfy2 knockout mice , interpreted in the context of the expression data presented here . The fact that we obtained ICSI offspring from XE , Z2Y*XSry males represents a significant advancement in establishing a minimum Y complement compatible with successful assisted fertilization . Although we have shown earlier that only two Y genes are sufficient to generate progeny [5] , this was achieved with round spermatid injection ( ROSI ) , a method which is considered experimental in human ART due to concerns regarding the safety of injecting immature germ cells and technical difficulties [28] . Intracytoplasmic sperm injection ( ICSI ) , however , is a common procedure in human ART , rendering our mouse data more directly translational . With the reemergence of Zfy genes from the backstage and their recently acknowledged roles during spermatogenesis [5 , 7 , 11 , 17 , 24 , 25] ( and this study ) , it will now be important to characterize the mechanism and identify the target genes that these transcription factors regulate . In humans , there is a single ZFY gene on the Y chromosome , which is ubiquitously expressed . No mutations of ZFY have been described and there is therefore no information concerning its possible contribution to human germ cell development or male fertility . The newly acquired mouse data regarding the role of Zfy gene in spermatogenesis may therefore trigger re-evaluation of ZFY function in humans .
The mice were maintained in accordance with the guidelines of the Laboratory Animal Services at the University of Hawaii and guidelines presented in National Research Council’s ( NCR ) “Guide for Care and Use of Laboratory Animals” published by Institute for Laboratory Animal Research ( ILAR ) of the National Academy of Science , Bethesda , MD , 2011 . Anesthesia , necessary for performing embryo transfers and semi-castration , was achieved by intraperitoneal injection of Avertin . Euthanasia was achieved by cervival dislocation . MAW has an active protocol for animal handling and treatment procedures ( protocol number 06–010 ) , reviewed and approved by local Animal Care and Use Committees annually . Pregnant mares’ serum gonadotrophin ( eCG ) and human chorionic gonadotrophin ( hCG ) were purchased from Calbiochem ( San Diego , CA ) . All other chemicals were obtained from Sigma Chemical Co . ( St Louis , MO ) unless otherwise stated . Sperm and oocyte collection and subsequent manipulation , including microinjections were done in HEPES-buffered CZB medium ( HEPES-CZB ) [29] . Culture of injected oocytes and embryos was done in CZB medium [30] . The mice of interest in this study were mice with limited Y gene complement ( Fig 1 , S1 Table ) : The XEOSry and XEY*XSry males were produced ‘in house’ by breeding XPafO or XPafY*X females [37] carrying the X-linked coat marker Patchy-fur [38] and XEif2s3yYTdym1Sry males that have the X chromosome carrying an Eif2s3y transgene [32] and a Y chromosome with an 11-kb deletion removing the testis determinant Sry ( dl1Rlb ) [1 , 39] , complemented by an autosomally-located Sry transgene [Tg ( Sry ) 2Ei] [31] . The XESxrbO and XESxrbY*X males were produced ‘in house’ by breeding XPafO or XPafY*X females described above and XEif2s3yYSxrb males that have the X chromosome carrying an Eif2s3y transgene [32] and a Y-chromosome that has Tp ( Y ) 1CtSxr-b sex reversal factor [40 , 41] attached distal to its PAR region . The XY*XSxra males were produced by ICSI with sperm from males of the same genotype and oocytes from wild-type females . Males transgenic for Zfy2 ( XE , Z2OSry , XE , Z2Y*XSry and XE , Z2SxrbY*X ) were produced as described above but with the father carrying XE , Z2 rather than XE . The crosses utilized in production of mice with limited Y gene complement yield a variety of progeny genotypes . The males of interest were identified among the progeny by genotyping for X and Y chromosome markers , scoring fur appearance , and evaluation of testes size . All mice were on MF1 genetic background , except for XY*XSxra which was C57BL/6 . XYRIII MF1 were used as wild-type controls; YRIII chromosome is the strain of Y chromosome from which Sxra and Sxrb derive . For assisted reproduction , six-to-twelve week-old B6D2F1 ( C57BL/6 x DBA/2 ) females ( NCI , Raleigh , NC ) were used as oocyte donors and CD-1 ( Charles River , Wilmington , MA ) or Swiss Webster ( NCI , Raleigh , NC ) mice were used as vasectomized males and surrogate/foster females for embryo transfer . The mice were fed ad libitum with a standard diet and maintained in a temperature and light-controlled room ( 22°C , 14h light/10h dark ) , in accordance with the guidelines of the Laboratory Animal Services at the University of Hawaii and guidelines presented in National Research Council’s ( NCR ) “Guide for Care and Use of Laboratory Animals” published by Institute for Laboratory Animal Research ( ILAR ) of the National Academy of Science , Bethesda , MD , 2011 . The protocol for animal handling and treatment procedures was reviewed and approved by local Animal Care and Use Committees . Anesthesia , necessary for performing embryo transfers and semi-castration , was achieved by intraperitoneal injection of Avertin . Euthanasia was achieved by cervival dislocation . MAW has an active protocol for animal handling and treatment procedures ( protocol number 06–010 ) , reviewed and approved by local Animal Care and Use Committees annually . For histology analysis , part of the testes were fixed in Bouin overnight and then stored in 70% ethanol prior to embedding in paraffin wax , sectioning at 5 μm , and staining with hematoxylin-eosin ( H&E ) and Periodic acid Schiff and hematoxylin ( PAS-H ) . The stages of seminiferous tubules were identified based on the composition of cells near the basal membrane according to the method described by Ahmed [42] , and as described by us before [5] . This was necessary because of meiotic and post-meiotic arrests present in males with limited Y gene complement , which prevented staging based on the changes of acrosome and nuclear morphology of spermatids . Sperm morphology was examined on surface spreads of spermatogenic cells prepared from frozen testis tissue as described earlier [25] . The images of sperm were captured at 1000x magnification . Injections with testicular cells were performed as described before [5 , 19] . Testes were collected twice from each male following initial semi-castration , and used for preparation of testicular cell suspensions for injections . The metaphase II ( MII ) oocytes for ROSI were collected from superovulated ( 5 iu eCG and 5 iu hCG given 48 hrs apart ) female mice and incubated at 37°C , 5% CO2 until injection . Testicular sperm suspension was diluted with HEPES-CZB containing 1% ( w/v ) polyvinyl pyrrolidone ( PVP ) on the injection dish . Spermatids were injected individually into the oocytes . The total duration of ROSI was no longer than 1 hour . The oocytes were activated shortly after injection by incubation in Ca2+-free CZB medium supplemented with 2 . 5 mM SrCl2 at 37°C , 5% CO2 for 4 hrs , after which time they were transferred into standard CZB medium for subsequent culture . At ~6–8 hrs after injection the oocytes were assessed for polar body extrusion and pronuclei development . Normally fertilized oocytes exhibiting two pronuclei ( PN ) and extruded second polar body ( PBII ) were allowed to cleave and were subjected to embryo transfer . Surrogate mothers were subjected to caesarian section on day 18 of pregnancy to allow for scoring of the numbers of fetuses and implantation sites . Chromosome preparation and analysis were performed as previously described [8] . In wild-type males the spermatid used for injection is considered chromosomally normal when resulting zygote contains 40 normal metaphase chromosomes , 20 maternal and 20 paternal . Males with limited Y gene complement either lack the Y chromosome or carry a minute Y*X chromosome variant . Thus , for these genotypes lack of one chromosome in the paternal chromosome complement and the presence of one small variant were considered normal . Offspring produced with ICSI and ROSI were genotyped by PCR to identify presence of Eif2s3y , Zfy2 , and Sry transgenes . Presence of Y*X was recognized by copy number assessment . Genomic DNA was isolated from mouse tails using phenol chloroform extraction and ethanol precipitation . DNA was used to amplify single copies of X-linked Prdx4 ( absent in Y*X ) and Amelx ( present in Y*X ) , and Atr ( on chromosome 9 ) for normalization using Power SYBR Green PCR Master Mix on a Quant Studio 12K Flex machine ( Applied Biosystems ) . The following conditions were used: 95°C for 10 min , followed by 40 cycles of 95°C for 10 sec and 60°C for 60 sec . Two PCR reactions were used to detect the presence of Y*X and the number of X-chromosomes . An 82-bp Prdx4 fragment were amplified using primers Prdx4-F and Prdx4-R and a 162-bp Amelx fragment with primers Amelx-F and Amelx-R . All samples were tested in quadruplicate per assay using XO samples as a reference control . Copy number estimation for each gene was calculated with the ΔΔCt method . Briefly , ΔCt values were calculated as difference between tested gene and Atr . ΔΔCt values were calculated by subtracting ΔCt of tested genes from the reference samples . The copy numbers were calculated by raising 2 to the power of ΔΔCt ( 2ΔΔCt ) . The genotypes were inferred from the copies of each target gene: XO , 1 Prdx4 + 1 Amelx; XY*X , 1 Prdx4 + 2 Amelx; XX , 2 Prdx4 + 2 Amelx; XXY*X , 2 Prdx4 + 3 Amelx . Primer sequences are shown in S3 Table . For real-time reverse transcriptase polymerase chain reaction ( RT-PCR ) , total testis RNA was extracted using Trizol and DNaseI treatment ( Ambion , Austin , TX , USA ) , and purified using an RNeasy kit ( Qiagen , Valencia , CA , USA ) . Reverse transcription of polyadenylated RNA was performed with Superscript Reverse Transcriptase III , according to the manufacturer’s guidelines ( Invitrogen , Carlsbad , CA , USA ) . Real-time PCR was performed using SYBR Green PCR Master mix on an ABI QuantStudio 12K Flex machine ( Applied Biosystems , Carlsbad , CA , USA ) . PCR reactions were incubated at 95°C for 10 min followed by 40 PCR cycles ( 10 s at 95°C and 60 s at 60°C ) . For analysis of Zfy expression , two types of PCR reactions were performed: ( 1 ) ‘Zfy1’ amplifying only Zfy1 transcripts and ( 2 ) ‘Zfy Global‴ amplifying both Zfy1 and Zfy2 . Three mice per genotype were analyzed , all reactions were carried out in quadruplicates per assay , and four different loading controls , two ubiquitously expressed genes ( actin and Sdha ) and two spermatid-specific genes ( Act and Acrv ) were used . DCt value for each individual sample was calculated by subtracting either the average Ct or geometric mean of loading control ( s ) from the average Ct of a tested gene . DDCt value was calculated by subtracting the DCt of each tested male from the average DCt of wild-type XY males , which served as references . The data were expressed as a fold value of expression level . Fisher's Exact Test was used to assess the differences between the genotypes for ROSI and ICSI and zygotic chromosome analysis data . Student t-test was used for gene expression and histology analyses .
|
The mammalian Y chromosome was once thought to be a genetic wasteland with testis determinant Sry being the only gene of importance . We now know that there are many genes on this chromosome crucial for male reproduction but their specific roles are often undefined . Here , we investigated the function of the Y chromosome gene Zfy2 during a final step of male gamete formation . We demonstrated that Zfy2 is responsible for allowing sperm precursor cells , haploid round spermatids , to undergo transformation into spermatozoa , and that these sperm are capable of yielding live offspring when injected into the oocytes . Thus , we identified a novel role of the Zfy2 gene during spermatogenesis and fertilization . Considering that in human sperm formation is a prerequisite for male infertility treatment using assisted reproduction technologies , our finding bear translational significance .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Mouse Y-Encoded Transcription Factor Zfy2 Is Essential for Sperm Formation and Function in Assisted Fertilization
|
Hepatitis C virus ( HCV ) assembly and envelopment are coordinated by a complex protein interaction network that includes most of the viral structural and nonstructural proteins . While the nonstructural protein 4A ( NS4A ) is known to be important for viral particle production , the specific function of NS4A in this process is not well understood . We performed mutagenesis of the C-terminal acidic domain of NS4A and found that mutation of several of these amino acids prevented the formation of the viral envelope , and therefore the production of infectious virions , without affecting viral RNA replication . In an overexpression system , we found that NS4A interacted with several viral proteins known to coordinate envelopment , including the viral E1 glycoprotein . One of the NS4A C-terminal mutations , Y45F , disrupted the interaction of NS4A with E1 . Specifically , NS4A interacted with the first hydrophobic region of E1 , a region previously described as regulating viral particle production . Indeed , we found that an E1 mutation in this region , D72A , also disrupted the interaction of NS4A with E1 . Supernatants from HCV NS4A Y45F transfected cells had significantly reduced levels of HCV RNA , however they contained equivalent levels of Core protein . Interestingly , the Core protein secreted from these cells formed high order oligomers with a density matching the infectious virus secreted from wild-type cells . These results suggest that this Y45F mutation in NS4A causes secretion of low-density Core particles lacking genomic HCV RNA . These results corroborate previous findings showing that the E1 D72A mutation also causes secretion of Core complexes lacking genomic HCV RNA , and therefore suggest that the interaction between NS4A and E1 is involved in the incorporation of viral RNA into infectious HCV particles . Our findings define a new role for NS4A in the HCV lifecycle and help elucidate the protein interactions necessary for production of infectious virus .
Hepatitis C virus ( HCV ) is a positive-sense RNA virus of the genus Hepacivirus in the Flaviviridae family . Over 70 million people worldwide are chronically infected with HCV and this chronic infection can lead to liver cirrhosis and hepatocellular cancer [1] . In the years spanning 2003–2013 , HCV-related deaths numbered more than any other CDC-reported infectious disease [2] . Despite the availability of newly designed , highly effective direct-acting antivirals , disease prevalence remains high , and no vaccine exists for the virus [3–5] . HCV encodes a single stranded , positive-sense RNA genome of approximately 9 . 6 kilobases in length . Upon virus entry into hepatocytes , the viral genome is translated to form a single polyprotein . The polyprotein is co- and post-translationally cleaved by both host and viral proteases , including the NS3-NS4A viral protein complex , to form ten individual proteins . These ten proteins include both structural proteins , which eventually make up the virion , and nonstructural proteins , which coordinate RNA replication and the other steps in the viral lifecycle , including virion assembly and envelopment ( reviewed in [6] ) . The late stages of the viral lifecycle , including assembly and envelopment , are just beginning to be dissected . While many details of these processes are not understood , recent work has uncovered several key steps that lead to production of infectious virus . Following RNA replication , HCV RNA is shuttled to the lipid droplet where Core protein accumulates , oligomerizes , and also recruits the NS3 and NS5A proteins [7–10] . NS5A is thought to play a role in RNA recruitment to the lipid droplet , whereas NS3 likely aids in movement of Core bound to RNA from the lipid droplet to nearby sites on the endoplasmic reticulum ( ER ) [11–16] . This process is coordinated by the NS2 protein which acts as a bridge between the nonstructural protein NS3 and the structural protein E2 , to link virion assembly at the lipid droplet to envelopment at the ER [17–22] . The role of NS2 in these steps is supported by the actions of the p7 protein [23 , 24] . Lastly , Core oligomers bound to viral RNA bud into the ER lumen , acquiring an ER-derived lipid bilayer envelope that contains the viral E1 and E2 transmembrane glycoproteins [25 , 26] . It is unclear what signals are necessary for the membrane curvature that results in budding , but it is clear that E1 and E2 are necessary for successful envelopment , as deletion of E1 and E2 prevents formation of the viral envelope and production of infectious virions [24] . Following virion budding into the ER lumen , the virion is transported through the very-low-density lipoprotein ( VLDL ) secretory pathway acquiring apolipoproteins and other lipids , and is ultimately released from the cell in a noncytolytic manner as a lipoviroparticle [27–30] . In addition to the viral proteins mentioned here , a number of host proteins also facilitate HCV morphogenesis [31] ( reviewed in [32] ) . In addition to its roles in viral assembly and envelopment , the NS3-NS4A protein complex has several other well-established functions in the HCV lifecycle . It is essential for viral polyprotein processing , viral RNA replication , and negative regulation of antiviral innate immunity ( reviewed in [33] ) . NS3 functions as both a serine protease and an RNA helicase and requires its cofactor NS4A to enhance these activities and to target the protease complex to intracellular membranes [34 , 35] . NS4A is 54 amino acids long and contains three domains , an N-terminal transmembrane domain that anchors NS3 to intracellular membranes , a central NS3-interaction domain required for proper folding of NS3 , and a C-terminal domain that contains a kink region followed by an acidic region with a high number of acidic amino acids [36–39] . While the specific roles of NS4A in the HCV lifecycle are largely thought to occur indirectly through its function as a cofactor for NS3 , mutation-based studies of the NS4A acidic domain suggest some independent roles for NS4A in regulating the HCV lifecycle , including during assembly and envelopment , as described below [36 , 40] . There is strong evidence that NS3 and NS4A are each involved in the steps of virion assembly and envelopment . Specifically , NS3 has been shown to be involved in viral particle production through interactions with both Core and NS2 [15 , 21 , 22 , 41] . In addition , culture adaptive amino acid mutations in the α0 helix of NS3 have been shown to promote viral assembly [16] . Separately , HCV particle production is also regulated by specific amino acids in the NS4A acidic domain such that when mutated , infectious virus formation is inhibited without affecting RNA replication . Some of these mutations can be partially rescued by compensatory amino acid substitutions in NS3 , suggesting that NS3 and NS4A together can cooperate to regulate HCV particle production [40] . However , both the extent of the role of NS4A in assembly and envelopment and the specific function of NS4A in regulating the production of infectious HCV remain unclear . Here , we define a new role for the NS4A protein in regulating HCV envelopment . We have identified amino acids in the acidic domain of NS4A that are required for the formation of the viral envelope . Further , we have found that NS4A alone can interact with a number of viral proteins that coordinate viral envelopment , including Core , E1 , E2 , and NS5A . Interestingly , disruption of the NS4A-E1 interaction , by mutation of either NS4A or E1 , prevents envelopment of the HCV particle and results in secretion of Core particles that are not associated with viral RNA . Taken together , our findings reveal a new role for NS4A in coordinating the HCV lifecycle and define new viral interactions that lead to successful HCV particle envelopment .
The acidic domain of NS4A has significant sequence homology between all HCV genotypes , with amino acids 40–54 in the acidic domain differing by at most 3 amino acids ( Fig 1A ) . In particular , the tyrosine residue at position 45 is conserved across 99 . 5% of the 865 HCV genome sequences in the Los Alamos HCV sequence database , which includes sequences from seven genotypes [42] . While previous studies have implicated the acidic domain of NS4A in regulating HCV RNA replication and particle production , the mechanism of this regulation was not explored [40] . We sought to investigate how the NS4A acidic domain contributes to HCV particle production . We engineered a structurally conservative amino acid substitution , changing the tyrosine residue ( Y , TAT ) at position 45 to a phenylalanine ( F , TTT ) in a genotype 2A strain of HCV , Japanese fulminant hepatitis-1 ( JFH1 ) [43] . We then generated wild-type ( WT ) or NS4A Y45F in vitro transcribed RNA and transfected it into Huh7 . 5 cells . At 3-days post-transfection , while the WT RNA produced more than 3 logs of infectious virus , no infectious virus was detected from cells transfected with RNA containing the Y45F mutation , as measured by focus forming assay ( Fig 1B ) . After several cell passages , Y45F RNA began to produce infectious virus , and after 14 days it produced equivalent titers to that of the WT virus ( Fig 1B ) . Sequencing of the NS4A region of HCV RNA extracted from cells at 1 , 3 , and 14 days post-transfection revealed that the Y45F mutation had reverted back to WT by day 14 , with some reversion detected as early as day 3 ( Fig 1C ) . The entire HCV genome was also sequenced following passage of the Y45F-transfected cells over 6 weeks . In this time , we only observed reversion of the Y45F mutation back to WT and did not detect the emergence of any second-site mutations . These results reveal that substitution of the tyrosine at position 45 with phenylalanine in NS4A prevents production of infectious HCV , indicating that Tyr-45 is required for the production of infectious virus . To determine if the loss of infectious HCV production by the NS4A Y45F amino acid change was due to altered HCV RNA replication , we engineered the Y45F mutation into an HCV subgenomic replicon construct containing a luciferase reporter and measured luciferase production over time following transfection of Huh7 . 5 cells with in vitro transcribed HCV RNA . We found that HCV replicon RNA with the Y45F mutation in NS4A replicated as efficiently as WT , while the HCV RNA with a lethal mutation in the NS5B RNA dependent RNA polymerase ( GND ) did not replicate ( Fig 2A ) . Additionally , the HCV proteins NS3 , NS4A and NS5A were expressed in lysates harvested at 48 hours post-transfection of either WT or Y45F RNA , indicating that the Y45F mutation did not affect the production of these viral proteins ( Fig 2B ) . Of note , the epitope of the NS4A antibody is in the C-terminal domain of NS4A that contains Tyr-45 . Therefore , the reduced detection of the NS4A band by immunoblotting in the mutant condition suggests that the Y45F mutation reduces NS4A recognition by this antibody ( Fig 2B ) . Further , the fact that HCV RNA replication is not altered by the Y45F mutation indicates that the NS4A protein must be stably expressed , as NS4A is required for HCV RNA replication [36 , 40 , 44] . We also confirmed that the Y45F mutation did not impact RNA replication in another genotype of HCV by measuring long-term HCV RNA replication in cells transduced with a genotype 1B subgenomic replicon RNA encoding a G418 selectable marker [45] . Indeed , there was no difference in the number of G418-resistant colonies that arose between WT and Y45F transduced cells , revealing that the Y45F mutation also did not impact replication of a genotype 1B subgenomic replicon ( S1A and S1B Fig ) . Since the interaction of NS3 with NS4A is essential for viral replication , we tested if the Y45F mutation affected this interaction using a co-immunoprecipitation experiment with overexpressed proteins . The results show that the Y45F mutation does not alter NS3-NS4A complex formation ( Fig 2C ) . Together , these results indicate that the Y45F mutation in NS4A does not alter HCV RNA replication , HCV protein expression , or NS3-NS4A complex formation . Therefore , the NS4A Y45F mutation in HCV must cause a defect at a later stage of the viral lifecycle . As the NS4A Y45F mutation did not alter HCV RNA replication but did prevent infectious virus production , we next tested if this mutation affected viral assembly and envelopment or viral release . We first examined if the Y45F mutation caused a viral release defect by measuring both intracellular and extracellular titer . We transfected Huh7 . 5 cells with WT , Y45F , or GND HCV RNA and measured the viral titer from the supernatant ( extracellular titer ) or from lysates generated by freeze-thaw cycles ( intracellular titer ) by using a focus forming assay . As before , HCV NS4A Y45F RNA did not produce extracellular titer ( Fig 3A ) , and here we found that it also did not produce intracellular titer ( Fig 3B ) . Taken together , these results indicate that the Y45F mutation impairs viral particle production prior to the formation of fully infectious virions . An infectious HCV virion contains viral RNA , encapsidated by the viral Core protein , surrounded by an outer lipid envelope containing the viral glycoproteins , E1 and E2 , and cellular lipids and lipoproteins [32 , 46 , 47] . Taking advantage of these structural properties of the HCV virion , we next tested if the Y45F mutation prevented viral envelopment by using a proteinase K protection assay . The HCV Core protein in enveloped virions is protected from degradation following proteinase K treatment by the outer lipid envelope [24] . Because the HCV glycoproteins are required for acquisition of the lipid bilayer membrane , a viral RNA with a deletion in the E1 and E2 coding region ( ΔE1/E2 ) can be used a negative control for envelopment [24] . Lysates were harvested from HCV RNA ( WT , Y45F , or ΔE1/E2 ) transfected Huh7 . 5 cells , incubated with proteinase K , and analyzed by immunoblot for Core . We found that while Core was protected from proteinase K digestion in WT , it was not protected in lysates containing the Y45F mutation , similar to ΔE1/E2 ( Fig 3C ) . These data indicate that the Y45F mutation prevents envelopment of the virion , resulting in a lack of both intracellular and extracellular viral titer , suggesting that Y45 may be an important residue for HCV envelopment . Based on our findings that HCV RNA with the NS4A Y45F mutation has a defect in viral envelopment , we hypothesized that other amino acids in the NS4A C-terminal acidic domain may also play a role in envelopment . To test this , we introduced several mutations into the acidic domain of NS4A that were previously found to be important for production of infectious HCV particles ( K41A , L44A , and E52A ) and tested their effects on viral envelopment [40] . We performed a proteinase K protection assay , as in Fig 3 , and found that the K41A , L44A and E52A mutants all resulted in a quantifiable decrease in protease-resistant Core as compared to WT , suggesting that these mutations also caused a defect in envelope formation ( Fig 4A and 4B ) . We additionally tested the impact of these amino acids on RNA replication , HCV protein expression , and production of both intracellular and extracellular titer; and also tested two additional mutations with known replication defects , Y45A and D49A , as controls [40] . HCV RNA containing the NS4A K41A , L44A , and E52A mutations all replicated and expressed HCV proteins to a similar extent as WT , while NS4A D49A and Y45A showed mild to severe replication defects ( Fig 4C and 4D ) . All of these mutations prevented intracellular and extracellular infectious virus from being produced , as seen previously by others ( Fig 4E and 4F ) [40] . Taken together , these data show that multiple amino acids within the acidic domain of NS4A are important for formation of the viral envelope and production of infectious virus . Because a complex network of HCV proteins regulates HCV assembly and envelopment , we hypothesized that NS4A may facilitate an interaction between either structural ( Core , E1 , or E2 ) or nonstructural ( p7 , NS2 or NS5A ) proteins to regulate these processes . Therefore , we first tested if overexpressed NS4A WT or Y45F interacted with Core , E1 , or E2 using co-immunoprecipitation in Huh7 . 5 cells . We found that overexpressed NS4A WT interacts with Core , E1 and E2 ( Figs 5A , S2A and S2B ) . While Core and E2 interactions with NS4A were equivalent for WT and Y45F ( S2A and S2B Fig ) , the NS4A Y45F mutation greatly decreased NS4A and E1 interaction ( Fig 5A and 5B ) . We also performed this immunoprecipitation with NS4A and E1 overexpression constructs from two additional genotypes , 1B and 3 . Indeed , we found that in both of these genotypes , NS4A interacts with E1 during overexpression and that the Y45F mutation reduces this interaction ( S1C and S1D Fig ) . To determine if NS4A WT interacts with E1 in the context of HCV infection we transfected Huh7 . 5 cells with an infectious clone of HCV containing an N-terminal HA tag on E1 [41] . We then immunoprecipitated E1 by using an anti-HA antibody and found that , indeed , NS4A and E1 can interact during HCV infection ( Fig 5C ) . We also tested the interactions of NS4A with p7 , NS2 , and NS5A , nonstructural proteins that all have roles in HCV envelopment [10–12 , 17–21 , 23 , 24] . NS4A did interact with NS5A , but this interaction was not altered by the Y45F mutation in NS4A ( S2D Fig ) . We found no interaction between overexpressed NS4A WT and either NS2 or p7 ( S2C Fig ) . To determine if a tyrosine at position 45 was absolutely essential for NS4A-E1 interaction , we created several additional mutations at position 45 ( Y45T , Y45R , and Y45D ) and tested their ability to interact with E1 . Interestingly , both NS4A Y45T and Y45D , but not Y45R , interacted with E1 ( S3 Fig ) . Together , these data show that NS4A can bind to Core , E1 , E2 , and NS5A , and that specific mutation of NS4A at Y45 can disrupt its binding to the E1 protein . To investigate the mechanism of how the NS4A-E1 interaction might facilitate viral envelopment , we mapped the binding site of NS4A on E1 . The E1 and E2 glycoproteins are translated in the ER membrane and are cleaved from the viral polyprotein by a host protease . After cleavage , E1 and E2 form a stable heterodimer and are retained in the ER . E1 has an N-terminal ectodomain , two hydrophobic regions , and a transmembrane region near the C-terminus ( Fig 6A ) ( reviewed in [48] ) . We therefore created a series of N-terminally Flag-tagged E1 truncation mutants based on these known domains of E1 ( Fig 6A ) . We overexpressed the truncation mutants with NS4A containing a C-terminal HA-tag in Huh7 . 5 cells and then performed Flag immunoprecipitations followed by immunoblotting for NS4A-HA . We found that NS4A co-immunoprecipitated with E1 aa 1–106 , aa 1–138 , and aa 67–192 , all of which contain the first hydrophobic region of E1 , but did not interact with aa 1–66 or aa 107–192 , which lack this region ( Fig 6B ) . These data suggest that NS4A binds to E1 via the first hydrophobic region of the protein . The first hydrophobic region of E1 has previously been implicated in HCV particle production , and several amino acids within this region are important for viral infectivity [49 , 50] . In particular , one specific E1 mutation , D72A ( HCV polyprotein aa 263 ) , attenuated viral infectivity even though Core protein was still secreted into cellular supernatants [50] . Because of these data and our finding that NS4A binds to the first hydrophobic region of E1 , which contains D72 , we hypothesized that the D72A mutation in E1 would specifically disrupt the NS4A-E1 interaction , similar to the Y45F mutation in NS4A ( Fig 5A ) . Indeed , the E1 D72A mutation diminished the interaction between NS4A and E1 during overexpression , supporting the conclusion that NS4A binds to the first hydrophobic region of E1 ( Fig 6C ) . Together , these data suggest that NS4A binds to the first hydrophobic region of E1 and this interaction specifically involves NS4A Y45 and E1 D72 . Others have shown that the D72A mutation in E1 in full-length HCV RNA results in release of non-infectious , partially formed virions containing Core protein but lacking HCV RNA [50] . Taken together with our findings that NS4A binds to the first hydrophobic region of E1 , which contains D72 , we hypothesized that Y45F may have a similar phenotype . Therefore , we measured the secretion of HCV RNA and Core protein into supernatants from cells replicating HCV NS4A WT or Y45F RNA . As a control , we also measured secretion of RNA and Core protein from cells transfected with HCV ΔE1/E2 RNA . We found that the Y45F mutation resulted in lower levels of extracellular HCV RNA as compared to WT , similar to ΔE1/E2 , as measured by RT-qPCR . However , secretion of Core into the supernatant was only slightly reduced , as measured by ELISA ( Fig 7A and 7B ) . In contrast , Core levels in the supernatant from HCV ΔE1/E2 were about 65% of what was seen for WT . These results for Y45F were similar to the previously published profile of Core and RNA secreted from cells transfected with the D72A mutant [50] . Because of these data , we sought to profile the viral components in the supernatants from Y45F-transfected cells . We collected and concentrated cellular supernatants from HCV NS4A WT or Y45F-transfected cells and then centrifuged these samples over iodixanol gradients . We collected 10 equal fractions from the top , with fraction 1 having the lowest density and fraction 10 having the highest density . Each fraction was analyzed by RT-qPCR for HCV RNA , by focus forming assay for viral titer , and by immunoblot for Core protein . In the WT samples , fractions 2 and 3 had the highest levels of both HCV RNA and viral titer ( Fig 7C ) . These fractions also contained high molecular weight complexes of Core protein ( Fig 7D , lanes 3–5 ) . These Core complexes could be reduced to the molecular weight of monomeric Core protein ( 21 kDa ) by treating the fractions with reducing agents and boiling them before SDS-PAGE analysis ( Fig 7D , bottom panels ) . We observed a second peak of HCV RNA in fractions 7 , 8 , and 9 , however these fractions had significantly less viral titer ( Fig 7D , lanes 7–9 ) . Therefore , the HCV RNA in these higher density fractions is likely non-infectious and may represent secreted membrane-associated RNA from replication complexes [51] . On the other hand , we saw little to no infectious viral titer from any fraction in the Y45F samples and found the majority of HCV RNA present in fractions 7–9 , while the expected infectious fractions ( 2–4 ) had little RNA ( Fig 7C ) . Interestingly , high molecular weight complexes of Core protein were still observed in fractions 3–5 , similar to the distribution of Core in WT samples ( Fig 7D ) . The fact that Core formed oligomers that were in a different fraction than the peak of HCV RNA suggests that the Y45F mutation results in release of partially formed virions containing Core protein oligomers but lacking HCV RNA . Furthermore , the fraction containing the majority of Core protein in both WT and Y45F also contained the E1 and E2 glycoproteins , suggesting that the primary component missing from the partially formed virions in Y45F is HCV RNA ( S4B Fig ) . Aside from Y45 , we identified several amino acids within the acidic domain of NS4A that have roles in HCV envelopment ( Fig 4A and 4B ) . To determine if one of these amino acids regulates the same function during envelopment as Y45 , we also performed the fractionation on supernatant from cells transfected with HCV RNA containing the NS4A K41A mutation . Similar to what we observed earlier , neither Y45F or K41A RNA produced infectious virions ( Figs 4E and S4A ) . We also included the E1 D72A mutant in these experiments to confirm the work of others showing that E1 D72A causes a defect in viral envelopment ( S4A Fig ) [50] . While the profile of RNA and Core protein from the supernatants of cells transfected with E1 D72A RNA was similar to that of cells transfected with the NS4A Y45F RNA , we found that the distribution of the K41A RNA in the supernatant was different from that of both NS4A Y45F and E1 D72A . The RNA secreted from the K41A transfected cells was present in fractions 2–4 as in WT , but not in fractions 7–9 , where the peak of RNA for Y45F and D72A was found . However , the supernatant of cells transfected with K41A RNA did contain high molecular weight oligomers of Core protein in the infectious fraction , similar to that seen with WT and Y45F ( S4A and S4B Fig ) . These results suggest that while K41A does have a defect in envelopment , K41 likely contributes to envelopment through a different mechanism than Y45 and E1 D72 . Together our data suggest that the NS4A Y45F mutation results in secretion of low-density , partially formed virions lacking HCV RNA .
Our results define a new role for NS4A in the late stages of the HCV lifecycle . Specifically , we have found that the acidic domain of NS4A is important for regulating assembly and that mutation of specific amino acids within this domain prevents formation of the viral envelope . Further , we have identified new interactions between NS4A and HCV structural and nonstructural viral proteins . This suggests that NS4A may act to bridge two stages of the HCV lifecycle , linking virion assembly at the lipid droplet to virion envelopment at the ER , similar to the actions of the NS2 protein . Importantly , we found that NS4A binds to E1 and that antagonizing this interaction with one amino acid change in NS4A prevents viral envelopment . We mapped the binding site of NS4A on E1 and found that it interacts with the first hydrophobic region , specifically at D72 , an amino acid that is known to be important for viral particle production [49 , 50] . Finally , we found that the Y45F mutation in NS4A , which prevents envelopment , also results in secretion of noninfectious , incompletely formed virions that are composed of low-density Core protein oligomers that lack HCV RNA . Together our results reveal a new role for NS4A in coordinating the proper assembly and envelopment of HCV particles to make infectious virus . The NS4A protein contains only 54 amino acids and yet has three distinct domains with specific functions in the HCV lifecycle . While the functions of the NS4A transmembrane domain and the NS3 interaction domain are largely defined [33] , much less is known about the function of the C-terminal region , which contains a high number of acidic amino acids ( Fig 1A ) . We found that mutation of several amino acids in the acidic domain , including Y45F , did not affect viral RNA replication , but did disrupt the formation of the viral envelope and therefore prevented production of infectious virus ( Figs 2A , 3A , 3C and 4 ) . Indeed , the presence of a Tyr at amino acid 45 of NS4A was so essential for the viral lifecycle that a viral RNA containing the Y45F mutation reverted back to the WT sequence after only a few days of passage in cell culture ( Fig 1B and 1C ) . Given that a Tyr to Phe change is a structurally conservative mutation removing only the hydroxyl group , it is formally possible that NS4A could be phosphorylated at this tyrosine . Indeed , a Y45D substitution , which mimics the charge of phosphorylation , retained NS4A binding to E1 ( S3A Fig ) . However , it is unlikely that lack of phosphorylation of this tyrosine would be the sole contributor to the envelopment defects , as several other amino acids within this region also prevented envelopment ( Fig 4 ) [40] . Therefore , the acidic domain likely regulates envelopment through the concerted actions of the amino acids in this acidic region of NS4A . Changing the amino acids in NS4A at K41 , L44 , Y45 , and E52 to alanine all resulted in loss of viral titer due to defects in envelopment ( Fig 4 ) . The acidic domain of NS4A , which has been proposed to have an alpha helical structure , is important for replication , and indeed the Y45A change results in less replication [36 , 40] . However , because mutation of the other amino acids within this C-terminal domain did not alter viral RNA replication , it is unlikely that they disrupt the conformation of the alpha helix . Structural predictions of this alpha helix suggest that K41 , Y45 , and E52 all are on one face of the helix while L44 and D49 would be on the opposite face [36] . Therefore , it is possible that the amino acids we studied in this region could facilitate different protein interactions on opposite faces of the protein , each contributing to HCV envelopment . In support of this hypothesis , previous work has shown that an adaptive mutation in NS3 partially rescues an assembly defect resulting from the K41A mutation , suggesting that NS4A can cooperate with NS3 via K41 for viral particle production [40] . We found that NS4A WT and Y45F bound NS3 equivalently ( Fig 2C ) but that the Y45F mutation prevented NS4A interaction with E1 ( Fig 5A ) . Additionally , analysis of cellular supernatants from cells transfected with HCV NS4A K41A RNA showed a profile of RNA secretion in supernatant fractions that was different than the RNA secreted from both WT and Y45F transfected cells . These data reveal that while the K41A and Y45F mutations both prevent HCV envelopment , these amino acids may function to facilitate different NS4A-protein interactions that regulate envelopment . Supporting the hypothesis that NS4A interacts with several HCV proteins to coordinate virion envelopment , we did identify several previously unknown interactions of NS4A with overexpressed structural and nonstructural proteins , such as Core , E1 , E2 , and NS5A . However , while NS2 is thought to be a main organizer of envelopment , we found that NS4A did not interact with NS2 during overexpression . Therefore , these data suggest that during infection NS2 and NS4A likely work together through a multi-protein complex or perform similar roles in the lifecycle [19] . Indeed , NS2 is considered to be the main organizer of envelopment , binding both structural and nonstructural proteins to link viral assembly steps at the lipid droplet to viral envelopment steps at the ER [17–21 , 23 , 24] . NS4A also binds to proteins involved in both early ( Core , NS3 , and NS5A ) and later ( E1 and E2 ) steps of assembly and envelopment , which could suggest that NS4A may also serve as a link between virion production steps at the lipid droplet and the ER , similar to NS2 . Overall , these results suggest that NS2 and NS4A could play similar roles in organizing and facilitating viral envelopment . We found that NS4A binds to E1 and that this interaction is disrupted by the Y45F mutation in NS4A and also by the D72A mutation in E1 ( Fig 5 , Fig 6C ) , suggesting that the NS4A-E1 interaction is important for envelopment of the virion . The E1 protein has an N-terminal ectodomain , two internal hydrophobic domains , a transmembrane domain , and a very short , 2 amino acid cytoplasmic , C-terminal tail ( Fig 6A ) [48] . The first hydrophobic domain of E1 has previously been shown to bind to Core , and mutations within this region diminish viral particle production [49 , 50 , 52–54] . Surprisingly , we found that NS4A binds to this region of E1 ( Fig 6B ) . Supporting this finding , we also found that a point mutation in the first hydrophobic domain of E1 , D72A , disrupted the NS4A-E1 interaction . Curiously , this E1 domain is hydrophobic and is proposed to associate with the lipid membrane bilayer , while the NS4A acidic domain is cytoplasmic and not known to be membrane-associated . It is possible that the acidic domain of NS4A could associate with the ER membrane to interact with this region of E1 . However , it is equally likely that NS4A and E1 can interact through a membrane-associated host protein or with cellular lipids . Future studies designed to determine how NS4A interacts with E1 would yield further insights into the HCV envelopment process . The finding that NS4A binds to the first hydrophobic region of E1 is particularly interesting , as this region in E1 has recently been shown to regulate viral particle production [49 , 50] . In fact , the D72A mutation in E1 ( polyprotein aa 263 ) , resulted in decreased viral titer and secretion of Core particles that were devoid of HCV genomic RNA . Further , others have shown that this mutation disrupts the localization of E1 with HCV RNA in fluorescence in situ hybridization experiments [50] . In our studies , fractionation of supernatant from cells replicating HCV NS4A Y45F RNA revealed that low-density fractions contained little to no HCV RNA , similar to E1 D72A ( Fig 7 , S4A Fig ) . However , these low-density fractions contained secreted Core oligomers , suggesting that these oligomers were associated with cellular lipids or apolipoproteins . Indeed , transfected Core protein has been shown to self-assemble into higher order complexes , and non-enveloped particles have been found in the serum of HCV infected patients [55 , 56] . Transfection of Core alone can also alter VLDL secretion , and therefore it is possible that secreted Core may be associated with cellular lipids and lipoproteins [57] . Taken together , these data suggest that the NS4A acidic domain and the E1 first hydrophobic domain cooperate during envelopment to aid in the incorporation of viral RNA into the virion . While NS4A itself does not have RNA binding capability , it does form a complex with NS3 , which contains an RNA binding helicase domain , and thus , NS3-NS4A together could cooperate with E1 to incorporate HCV RNA into the developing virion . Our study contributes new insights into the steps required for HCV to form infectious viral particles . As the viral particle lifecycle stages that occur in association with lipid droplets and the ER are tightly linked and likely occur nearly simultaneously , it’s unclear if nucleocapsid intermediates ( Core protein assembled around HCV RNA ) exist separate from fully enveloped nucleocapsids [32] . Our data show that Core protein assembles into oligomers prior to envelopment and suggest that the function of NS4A in viral assembly and envelopment is after this Core oligomerization step . Further , the fact that we identified Core protein oligomers that did not contain a protective envelope or HCV RNA suggests that RNA incorporation into the virion at or near envelopment sites could be a necessary signal for virion budding events to occur . Our data therefore support a model by which NS4A interacts with E1 to link viral RNA to Core oligomers in the forming virion and signal the envelopment of the Core-RNA complex .
Huh7 . 5 cells ( gift of Dr . Michael Gale ) , which have been previously described [45] , were maintained in Dulbecco’s modification of eagle’s medium ( DMEM; Mediatech ) supplemented with 10% fetal bovine serum ( FBS; HyClone ) and 25 mM HEPES ( Thermo-Fisher ) at 37°C with 5% CO2 . The identity of the Huh7 . 5 cells used in this study was verified by using the Promega GenePrint STR kit ( DNA Analysis Facility , Duke University ) , and cells were verified as mycoplasma free by the LookOut Mycoplasma PCR detection kit ( Sigma ) . These plasmids have been described previously: psJFH1-p7+NS [58] , pHCV-HP WT [45] , HJ3-E1/HA-NS2/YFP ( [41] , gift of Dr . MinKyung Yi ) , and pFL-J6/JFH-1-FlagE2 ( [59] , gift of Dr . Matthew Evans ) . psJFH1-p7+NS is a culture adapted strain of JFH1 containing 7 mutations within p7 and the nonstructural proteins that we have described previously [58] . pJFH1-SGR-luc contains a bicistronic replicon as follows: [JFH1-derived untranslated region ( UTR; nt 1–397 ) ]-[in frame Renilla luciferase reporter]-[EMCV IRES-nonstructural genes ( NS3-NS5B ) ] . To make this plasmid , a DNA fragment encoding Renilla luciferase was fused between the T7 promoter sequence-5’ UTR of JFH1 and the EMCV IRES-nonstructural genes from pSGR-JFHI [60] following PCR ( for oligonucleotide sequence see Table 1 ) , digestion ( with inserted BglII site between 5’UTR and 5’end of Renilla , a PmeI site between the 3’end of Renilla and 5’end of the ECMV IRES , and an existing AgeI site in the 5’UTR of pSGR-JFH1 ) , and a 3-piece ligation . Mutagenesis of constructs was performed using the QuikChange lightning site-directed mutagenesis kit ( Stratagene ) on pJFH1-SGR-luc , pHCV-HP , psJFH1-p7+NS , pFL-J6/JFH-1-FlagE2 , pEF1 NS4A-HA , or pEF-Tak Flag-E1 using the indicated oligonucleotides ( Table 1 ) . psJFH1-p7+NS ΔE1/E2 was constructed by removing amino acids 192–720 from the psJFH1 p7+NS background [24] . HCV over-expression constructs ( noted below ) were constructed by PCR amplification of the gene of interest from psJFH1-p7+NS and insertion of the PmeI-NotI digested fragment into pEF-Tak-Flag [61] or the EcoRI-XbaI digested fragment into pEF1 . pEF-Tak Flag-NS2 was created using InFusion ( Clontech ) after PCR . pEF-Tak Flag-E1 ( genotypes 1B and 3 ) and pEF1 NS4A-HA WT and Y45F ( genotypes 1B and 3 ) were made using gBlocks ( IDT ) of the entire coding region with vector overlap , followed by InFusion cloning ( Clontech ) . Table 1 provides the sequence of all oligonucleotides used . Bold letters in the oligonucleotide sequences indicate overlap with vector sequence , and the sequence of the HA tag within the oligonucleotides is underlined . All nucleotide and amino acid positions refer to the JFH1 genome ( GenBank accession number: AB047639 ) . The sequences of all plasmids were verified by DNA sequencing and are available upon request . Plasmid DNA encoding the described HCV constructs was linearized using the XbaI restriction enzyme for full-length or SGR-luc HCV RNA or ScaI for HP subgenomic replicon RNA . Purified linearized DNA was used as a template for in vitro transcription with a MEGAscript T7 transcription kit ( Thermo-Fisher ) . The in vitro transcribed RNA was treated with DNase ( Thermo-Fisher ) and then purified by phenol-chloroform extraction . The quality of the RNA was verified on a denaturing gel . For electroporation , 1 μg ( HCV replicon RNA ) or 5 μg ( HCV full-length RNA ) was electroporated into 4x106 Huh7 . 5 cells in Cytomix electroporation buffer ( 120 mM KCl , 10 mM Potassium Phosphate Buffer , 5 mM MgCl2 , 25 mM HEPES , 0 . 15 mM CaCl2 , 2 mM EGTA , pH 7 . 6 ) at 250 V and 950 μF in a 4 mm cuvette with a Gene Pulser Xcell system ( Bio-Rad ) . Four hours post electroporation , cells were washed extensively with Phosphate Buffered Saline ( PBS ) and cDMEM . Extracellular titer: Supernatants were harvested from Huh7 . 5 cells electroporated with HCV RNA at indicated time points , serially diluted , and used to infect naïve Huh7 . 5 cells in triplicate wells of a 48-well plate for 2 hours . Plates were harvested at 48 hours post infection and fixed with 4% paraformaldehyde . Cells were permeabilized ( 0 . 2% Triton-X-100 in PBS ) , blocked ( 10% FBS in PBS ) and immunostained with mouse anti-HCV NS5A antibody ( 9e10 , 1:500 , gift of Dr . Charles Rice ) . Infected cells were visualized following incubation with horseradish peroxidase ( HRP ) -conjugated secondary antibody ( 1:500; Jackson ImmunoResearch ) and VIP Peroxidase Substrate Kit ( Vector Laboratories ) . Foci were counted at 40X magnification . Titer ( FFU/mL ) was determined as previously described [62] . Intracellular titer: Cell pellets were washed with PBS and resuspended in serum-free DMEM . Cells were then lysed using a series of freeze/thaw cycles in a dry ice/Ethanol bath . Post-nuclear supernatants were used to infect naïve Huh7 . 5 cells , and a focus forming assay was performed as described above . JFH1 SGR-luc in vitro transcribed RNA ( 1 μg ) was electroporated into Huh7 . 5 cells . Cells were suspended in 20 mL cDMEM and plated in 12-well plates . Cells were harvested after a PBS wash by incubation in Renilla lysis buffer ( Promega ) . Renilla luciferase values were measured according to manufacturer’s instructions ( Renilla Luciferase Assay System , Promega ) using a BioTek Synergy 2 microplate reader . RNA was extracted from cells by using the Qiagen RNeasy kit according to manufacturer’s instructions and then used as a template for cDNA synthesis with the Superscript cDNA synthesis kit ( Thermo Fisher ) . The NS4A region of the HCV genome was amplified by nested PCR with the following oligonucleotides: Round 1: 5’–CAGTCCGATGGAGAAGAAGG—3’ , 5’—GCATGGGATGGGGCAGTC—3’ , Round 2: 5’—ACACATAGACGCCCACTTCC—3’ , 5’—GTATGTCCTGGGCCTGCTTA—3’ , and then the 542 bp PCR product was purified and sequenced by Sanger sequencing . RNA from cells was isolated using the RNeasy kit ( Qiagen ) , and RNA from infected supernatants was isolated using the QIAamp viral RNA kit ( Qiagen ) , both according to manufacturer’s instructions . The RNA copy number of harvested RNA was measured in triplicate by RT-qPCR using the TaqMan Fast Virus 1-Step Mix ( Qiagen ) with an HCV-specific probe targeting the 5’ untranslated region of HCV ( Assay ID: Pa03453408_s1 ) . The copy number was calculated by comparison to a standard curve of a full-length in vitro transcribed HCV RNA , as described [58] . Cells were lysed in a modified radio immunoprecipitation assay ( RIPA ) buffer ( 10 mM Tris pH 7 . 5 , 150 mM NaCl , 0 . 5% sodium deoxycholate , 1% Triton X-100 ) supplemented with protease inhibitor cocktail ( Sigma ) and phosphatase inhibitor cocktail ( Millipore ) , and post-nuclear supernatants were harvested by centrifugation . Quantified protein was resolved by SDS/PAGE , transferred to PVDF membranes using the Turbo-transfer system ( BioRad ) and blocked with StartingBlock ( Thermo-Fisher ) or 3% bovine serum albumin ( Sigma ) in PBS with 0 . 1% Tween ( PBS-T ) . Membranes were probed with specific antibodies , washed with PBS-T and incubated with species-specific HRP conjugated antibodies ( Jackson ImmunoResearch , 1:5000 ) , washed again with PBS-T , and treated with Pico PLUS enhanced chemiluminescent ( ECL ) reagent ( Thermo-Fisher ) . The signal was then captured on X-ray film or by using a LICOR Odyssey FC . Antibodies used for immunoblot include mouse anti-HCV Core ( 1:250 , Abcam ) , mouse anti-HCV NS3 ( 1:500 , Abcam ) , rabbit anti-HCV NS4A ( 1:1000 , Genscript [63] ) , mouse anti-HCV NS5A ( 1:500 , 9e10 , gift of Dr . Charles Rice ) , anti-Flag HRP ( 1:2500 , Sigma ) , rabbit anti-HA ( 1:500 , Sigma ) , and mouse anti-E1 ( 1:1000 , A4 , gift of Dr . Charles Rice ) . This protocol was adapted from the manuscript by Gentzsch and colleagues [24] . Briefly , cells electroporated with JFH1-p7+NS in vitro transcribed RNA were harvested at 48 hours post electroporation by scraping into cold proteinase K buffer ( 50 mM Tris-HCl pH 8 . 0 , 10 mM CaCl2 , 1 mM DTT ) . Cells were then lysed by five freeze/thaw cycles and aliquots of lysate ( 50 μL ) were either ( i ) left untreated , ( ii ) pretreated with 5 μL of 10% Triton-X-100 followed by proteinase K treatment ( 50 μg/mL ) for 30 minutes on ice , or ( iii ) treated with proteinase K only . Proteinase K treatment was terminated by incubation with 10 mM phenylmethane sulfonyl fluoride . The samples were mixed with 4X SDS sample buffer ( 1 M Tris ( pH 6 . 8 ) , 60% glycerol , 0 . 06% Bromophenol Blue , 12% SDS ) ) , incubated at 50°C for 5 minutes , and immunoblotted for HCV Core protein , as above . 300–500 μg of protein extracted as above was incubated with 50 μL anti-Flag M2 magnetic beads ( Sigma ) in 1X Tris buffered saline ( TBS ) at 4°C overnight with head over tail rotation . Beads were washed 3X in modified 1X RIPA buffer and eluted in 2X Laemmli Buffer ( BioRad ) at 50°C for 5 minutes . Protein was resolved by SDS/PAGE and immunoblotting , as above . In the anti-HA immunoprecipitations , protein was incubated with equivalent amounts of anti-HA or IgG antibodies in 1X TBS at 4°C overnight with head over tail rotation . Then Protein G Dynabeads ( Thermo-Fisher ) were added and rotated at 4°C for two hours . Beads were washed and eluted as above . In vitro transcribed genotype 1B HP-HCV RNA ( linearized by ScaI ) was electroporated as above into Huh7 . 5 cells and plated into 10 cm plates at 2*104 or 2*103 cells per plate , along with cells electroporated with a non-replicating control . Cells were washed thoroughly with 1X PBS and cDMEM at 4 hours post electroporation . cDMEM containing 0 . 4 mg/mL G418 ( Life Technologies ) was added to cells at 24 hours post electroporation to begin selection . Cells were fixed after 3 weeks under G418 selection and stained with crystal violet ( Sigma ) in 20% methanol . Colonies from triplicate plates were counted and each normalized to the average number of colonies on WT plates . Core protein was quantified from filtered supernatants from Huh7 . 5 cells 72 hours post-electroporation using the HCV Core Antigen ELISA kit according to the manufacturer’s instructions ( Cell Biolabs ) . Concentrated supernatants were purified over a 10–50% iodixanol gradient , as previously described [50] . Briefly , at 48 hours post electroporation of HCV RNA in Huh7 . 5 cells , supernatant was collected , mixed with polyethylene glycol ( PEG ) 8000 to a final concentration of 8% , and incubated with rocking at 4°C overnight . PEG supernatants were centrifuged at 11 , 000 X g for 30 minutes , supernatant was removed , and remaining pellets were suspended in cold 1X PBS . These resuspensions were layered over a 10–50% iodixanol gradient and centrifuged at 222 , 000 X g for 16 hours in a SW41 rotor in a Beckman Coulter ultracentrifuge . 10 equal fractions ( 1 ml ) were collected with a BioComp piston gradient fractionator , and then viral titer ( FFU/ml ) , HCV RNA copy number ( RT-qPCR ) , and Core protein ( immunoblotting ) were measured from each fraction . In some experiments ( indicated in the figure legend ) , the protein from the gradients was incubated for 30 minutes at 37°C in 2X Urea Loading Buffer ( 50 mM Tris-HCL , 1 . 6% SDS , 7% glycerol , 8 M Urea , 4% 2-mercaptoethanol , Bromophenol Blue ) with vortexing every 10 minutes . It was then boiled at 95°C for 10 minutes prior to SDS-PAGE and immunoblotting . Student’s unpaired t-tests and one-way analysis of variance ( ANOVA ) were used for statistical analysis of data . Values are presented as mean ± standard error of the mean for biological replicates or standard deviation for technical replicates ( n = 3 , or as indicated ) . *—P < 0 . 05 , **—P < 0 . 01 , ***—P < 0 . 001 , ****—P < 0 . 0001 .
|
RNA viruses , which encompass both established and emerging pathogens , pose significant public health challenges . Viruses in the family Flaviviridae , including Dengue virus , Zika virus , and hepatitis C virus ( HCV ) , continue to cause morbidity and mortality worldwide . One HCV protein , NS4A , acts in several steps of the viral lifecycle; however , how it contributes to viral particle production is not understood . Here , we investigated the role of one region of NS4A , the C-terminal acidic domain , in regulating the viral lifecycle . We found that amino acids within this domain are important for viral envelopment and the production infectious particles , specifically through interaction with the E1 glycoprotein . NS4A interacts with the first hydrophobic domain of E1 . Disruption of this interaction , by either mutation of E1 or NS4A , prevents the production of infectious virus particles and instead results in release of low-density Core protein complexes that lack HCV RNA into the cellular supernatant . Overall , our results reveal that NS4A is important for late stages of the HCV lifecycle and suggest that the interaction between NS4A and E1 may regulate the incorporation of viral RNA into the virion for the formation of infectious HCV particles .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"microbial",
"mutation",
"pathology",
"and",
"laboratory",
"medicine",
"hepacivirus",
"pathogens",
"enzymes",
"microbiology",
"enzymology",
"viral",
"structure",
"viruses",
"rna",
"viruses",
"molecular",
"biology",
"techniques",
"rna",
"sequencing",
"research",
"and",
"analysis",
"methods",
"lipids",
"viral",
"core",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"hepatitis",
"c",
"virus",
"hepatitis",
"viruses",
"viral",
"replication",
"molecular",
"biology",
"virions",
"biochemistry",
"flaviviruses",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"proteases",
"organisms"
] |
2019
|
The acidic domain of the hepatitis C virus NS4A protein is required for viral assembly and envelopment through interactions with the viral E1 glycoprotein
|
The human cytomegalovirus ( HCMV ) clinical strain Toledo and the attenuated strain AD169 exhibit a striking difference in pathogenic potential and cell tropism . The virulent Toledo genome contains a 15-kb segment , which is present in all virulent strains but is absent from the AD169 genome . The pathogenic differences between the 2 strains are thought to be associated with this additional genome segment . Cytokines induced during viral infection play major roles in the regulation of the cellular interactions involving cells of the immune and inflammatory systems and consequently determine the pathogenic outcome of infection . The chemokine RANTES ( Regulated on activation , normal T-cell expressed and secreted ) attracts immune cells during inflammation and the immune response , indicating a role for RANTES in viral pathogenesis . Here , we show that RANTES was downregulated in human foreskin fibroblast ( HFF ) cells at a later stage after infection with the Toledo strain but not after infection with the AD169 strain . miR-UL148D , the only miRNA predicted from the UL/b' sequences of the Toledo genome , targeted the 3′-untranslated region of RANTES and induced degradation of RANTES mRNA during infection . While wild-type Toledo inhibited expression of RANTES in HFF cells , Toledo mutant virus in which miR-UL148D is specifically abrogated did not repress RANTES expression . Furthermore , miR-UL148D-mediated downregulation of RANTES was inhibited by treatment with a miR-UL148D-specific inhibitor designed to bind to the miR-UL148D sequence via an antisense mechanism , supporting the potential value of antisense agents as therapeutic tools directed against HCMV . Our findings identify a viral microRNA as a novel negative regulator of the chemokine RANTES and provide clues for understanding the pathogenesis of the clinical strains of HCMV .
Human cytomegalovirus ( HCMV ) is a member of the β-herpesvirus family and a ubiquitous human pathogen . After a primary infection , HCMV establishes lifelong latency , which seldom causes illness in an immunocompetent host [1] , [2] . However , HCMV is an infectious pathogen that induces morbidity and mortality in immunocompromised individuals such as AIDS patients [3] . HCMV strains display different levels of virulence , tissue tropism , and pathogenicity depending on their degree of adaptation in fibroblasts . Injection of the low-passaged HCMV strain Toledo into healthy adults causes clinically apparent diseases [4] , whereas adults inoculated with the attenuated HCMV AD169 or Towne strains do not manifest any clinical symptoms [5] , [6] . These results indicate that the clinical Toledo strain is more virulent than the attenuated AD169 strain . Clinical and attenuated strains of HCMV also differ in their ability to render infected cells susceptible to the action of natural killer ( NK ) cells . Clinical strains confer a strong NK cell resistance , whereas high-passaged attenuated strains cause only marginal effects with respect to NK cell recognition [7] , [8] . This suggests that the mechanisms employed to evade NK cell lysis may be lost during in vitro passage of the attenuated viruses . The complete genome of the laboratory-adapted strain AD169 has been sequenced [9] . An additional 19 viral genes ( UL133 through UL151 ) , which are absent from AD169 , were found in low-passaged clinical isolates [10] . These genetic differences between attenuated strains and clinical strains may be implicated in HCMV-induced immunopathogenesis , as well as in strain-specific behaviors , such as tissue tropism and the ability to establish persistent or latent infections [11]–[13] . When such a virus infects its host , the host immune system is activated to remove the virus or virus-infected cells . One of the first lines of effector signals that attract circulating leukocytes to the site of viral infection is provided by chemokines [14] . Chemokines are small chemoattractant cytokines that are expressed and secreted during an inflammatory response . Chemokines attract specific immune cells during viral infection . In certain inflammatory reactions , proinflammatory cytokines , such as tumor necrosis factor alpha ( TNF-α ) and interferon gamma ( IFN-γ ) , stimulate the expression of RANTES , which is chemotactic for T cells , eosinophils and basophils [15] . The initiation stage of viral infection , which includes particle binding and internalization , activates various responses in the host cells [14] . Viruses can manipulate the cellular interactions involving cells of the immune systems for their benefit by regulating cellular chemokines . Regulation of RANTES by HCMV has also been reported . Billstrom et al . showed that the abortive HCMV added to endothelial cells was capable of both entering into the host cells and uncoating . However , the virus was found to be incapable of repressing the expression of RANTES . The degradation of RANTES mRNA was also found in endothelial cells infected with the clinical isolate HCMV 4010 [16] . In contrast , Michelson et al . showed that the level of extracellular RANTES was reduced at a later stage after infection of AD169 in fibroblasts without mRNA degradation . An apparent discrepancy regarding degradation of RANTES mRNA between these two studies might be due to the genetic polymorphisms among HCMV strains . In particular , because the predominant function of miRNA is to induce degradation of target mRNA for reduced protein output [17] , the differential expression of miRNAs among HCMV strains might account for this discrepancy . miRNAs are a type of small RNA that regulates a variety of cellular processes [18] , [19] . Mature miRNAs are single-stranded RNAs of 20–24 base pairs that are derived from longer transcripts processed by the enzymes Drosha and Dicer [20] , [21] . Mature miRNA is incorporated into the RNA-induced silencing complex ( RISC ) during targeting of transcripts [22] . In the case of complete homology between miRNA and a target transcript , the target transcript is cleaved , whereas partial homology can lead to RNA cleavage or translational inhibition [23] . miRNAs exist in virtually all organisms including animals , plants and viruses . HCMV expresses multiple miRNAs during its infectious life cycle . By small RNA cloning and sequencing technique it was shown that HCMV expresses 9 miRNA precursors [24] . A bioinformatics analysis predicted that 2 additional miRNAs are conserved between HCMV and chimpanzee CMV [25] . Collectively , these 11 miRNA precursors in HCMV lead to 14 mature miRNAs , but their physiological functions are largely unknown . Interestingly , miR-UL148D miRNA exists in the 15-kb segment of clinical strains [24] , [26] while it is not found in the attenuated AD169 strain . This suggests that miR-UL148D might play a role in the pathogenicity of the HCMV clinical strains . In this study , we show that miR-UL148D directly targets RANTES , thereby inducing degradation of RANTES mRNA . Accordingly , the level of secreted RANTES was reduced by miR-UL148D during infection with the Toledo strain . A mutant Toledo virus with a deleted miR-UL148D gene did not repress RANTES expression . A PNA-based antisense oligonucleotide specific to a miR-UL148D reverted miR-UL148D-mediated downregulation of RANTES during HCMV infection . These results reveal that the clinical HCMV strain employ an additional miRNA-based mechanism to modulate the host immune system .
RANTES secretion is reduced by the attenuated HCMV strain AD169 in HFF cells during the replication phase [27] . Taking diverse roles of RANTES in immune and inflammatory responses into consideration and better understanding the virulence and pathogenicity of the clinical HCMV strain , we tested whether RANTES expression is also regulated by Toledo , which is the predominant clinical isolate of HCMV . HFF cells were infected with wild-type AD169 ( WT-AD169 ) or wild-type Toledo ( WT-Toledo ) . We employed quantitative RT-PCR to examine the RANTES mRNA levels . In the WT-AD169-infected cells , the level of RANTES mRNA increased gradually and peaked at 48 h post-infection ( Figure 1A , triangles ) . Interestingly , WT-Toledo infection repressed the expression of RANTES mRNA throughout the infection with almost nondetectable levels at 48 h post-infection ( Figure 1A , circles ) . These data suggest that Toledo might have at least an extra functional gene , which is lacking in AD169 , that inhibits the transcription of RANTES or degrades RANTES mRNA . The primary function of miRNA is to destabilize target mRNA for reduced protein output [17] . To explore whether HCMV miRNAs are related to the regulation of the expression of RANTES , we performed bioinformatics-based target site screening . We utilized 2 previously-designed algorithms for identification of miRNA-mRNA target interaction pairs that exhibit favorable free energies ( ΔG ) . The favorable free energies of the interaction of viral miRNAs with RANTES 3′UTR were estimated using the computational prediction algorithm , RNA22 [28] . Highly favorable free energy is provided by complementary pairing of miRNA with the 3′UTR that further stabilizes target recognition . Among the 14 miRNAs tested , only miR-UL148D [24] exhibited highly favorable free energy ( ΔG = −35 . 4 kcal/mol ) . As a comparison , this interaction was predicted to be of higher stability than that observed between a pairing of a well-known let-7 miRNA with lin-28 mRNA ( ΔG = −31 . 0 kcal/mol ) ( data not shown ) . These results were confirmed by using another algorithm , RNAhybrid [29] . This RNAhybrid algorithm yields highly favorable free energies for miR-UL148D ( ΔG = −36 . 0 kcal/mol ) . On the basis of computational predictions , we concluded that miR-UL148D might target the 3′UTR of RANTES . miR-UL148D expression has been detected by RT-PCR analysis in other clinical strains [30] and the mature miR-UL148D sequence is completely conserved among all known clinical strains ( Figure 1C ) . We determined whether miR-UL148D is also expressed in Toledo . Total RNA was isolated from Toledo-infected HFF cells . The mature miR-UL148D was detected by RNase protection assay . RNase digestion yielded a mature miR-UL148D of ∼22 bp size that is shorter than the original miR-UL148D probe of 27-bp ( Figure 1D , first panel ) . The expression of miR-UL148D increased throughout the infection and peaked at 72 h post-infection ( Figure 1D , first panel , lanes 2–4 ) . RT-PCR analysis of the same samples revealed that the expression level of RANTES mRNA is inversely correlated with the expression level of miR-UL148D ( Figure 1D , third panel ) . These data suggest that Toledo-induced downregulation of RANTES may be a consequence of miR-UL148D expression . To test whether the 3′UTR of RANTES is a target of miR-UL148D , the entire RANTES 3′UTR was subcloned immediately downstream of the firefly luciferase open reading frame ( ORF ) , and miR-UL148D was cloned into the pSuper-retro vector as a 22-nucleotide mature form . A luciferase reporter vector containing the 3′UTR of RANTES was transfected into 293T cells with varying quantities of wild-type miR-UL148D ( miR-UL148D-WT ) or control seed binding mutant miRNA ( miR-UL148D-mut ) . The transfected cells were lysed and appropriate substrates were added into the lysates in order to measure luciferase activity . Dose-dependent experiments demonstrated that the relative luciferase activity was significantly decreased in the presence of miR-UL148D ( Figure 2A , black bars ) . In contrast , no reduction was observed in miR-UL148D-mut-transfected cells ( Figure 2A , gray bars ) . Thus , the 3′UTR of RANTES has a specific binding site for miR-UL148D , indicating that it may be a target of miR-UL148D . To confirm that the 3′UTR of RANTES contains a functional target site for miR-UL148D , we generated a seed-binding mutant of the 3′UTR of RANTES ( RANTES-3′UTR-mut ) ( Figure 2B ) . After co-transfection of both the RANTES 3′UTR-mut and miR-UL148D , the luciferase activity was measured . As expected , miR-UL148D-WT did not repress the luciferase activity of RANTES-3′UTR-mut-expressing cells ( Figure 2C , lane 5 ) , whereas the luciferase activity of wild type RANTES 3′UTR ( RANTES-3′UTR-WT ) -expressing cells was downregulated in the presence of miR-UL148D-WT ( Figure 2C , lane 2 ) . These data demonstrate that miR-UL148D targets RANTES specifically at the predicted target sequence identified in the 3′UTR . As demonstrated above , miR-UL148D represses luciferase activity in cells expressing the 3′UTR of RANTES downstream of a luciferase reporter gene ( Figure 2 , A and C ) . To ascertain whether miR-UL148D is sufficient to downregulate the level of RANTES protein , we examined the level of RANTES secreted into culture supernatants in the presence of miR-UL148D . We generated a RANTES-expressing vector which includes the 3′UTR region and tested whether the ectopic expression of miR-UL148D in 293T cells reduces the ectopic expression of RANTES . After transfection of miR-UL148D into 293T cells with the S35-radioisotope , secreted RANTES was immunoprecipitated with anti-RANTES antibody . Northern blot analysis validated the expression of miR-UL148D-WT and miR-UL148D-mut in transfected cells ( Figure 3A , bottom panels ) . The amount of precipitated RANTES decreased in the presence of miR-UL148D ( Figure 3A , compare lane 3 and lane 4 ) . In contrast , miR-UL148D-mut did not affect the secretion of RANTES ( Figure 3A , lane 5 ) . These results indicate that miR-UL148D downregulates the expression of RANTES protein through its miRNA function . Next , we determined whether the reduction of RANTES by miR-UL148D is caused by specific binding of miR-UL148D to the seed region of RANTES 3′UTR . After co-transfection of RANTES-3′UTR-WT or RANTES-3′UTR-mut with miR-UL148D-WT or miR-UL148D-mut , the labeled media of the transfected cells were immunoprecipitated with the anti-RANTES antibody . Co-transfection of both RANTES-3′UTR-WT and miR-UL148D-WT resulted in reduced levels of secreted RANTES . miR-UL148D-mut did not affect the expression of RANTES-3′UTR-WT ( Figure 3B , compare lane 1 and 2 ) . As designed , miR-UL148D-mut was able to prevent the expression of RANTES-3′UTR-mut ( Figure 3B , compare lane 3 and lane 4 ) . These results show that the 3′UTR of RANTES contains a specific target site for miR-UL148D . To elucidate whether miR-UL148D reduces RANTES expression by translation inhibition or mRNA degradation , we determined RANTES mRNA levels in 293T cells expressing either miR-UL148D-WT or miR-UL148D-mut . RANTES-3′UTR-WT and RANTES-3′UTR-mut were transfected into 293T cells along with either miR-UL148D-WT or miR-UL148D-mut . After RNA extraction from transfected cells , RANTES mRNA levels were measured by RT-PCR . The level of RANTES-3′UTR-WT transcripts was decreased in the presence of miR-UL148D-WT but not in the presence of miR-UL148D-mut ( Figure 3C , compare lane 3 and lane 4 ) . As expected , miR-UL148D-mut targeted only RANTES-3′UTR-mut mRNA but not RANTES-3′UTR-WT for degradation ( Figure 3C , compare lane 4 and lane 7 ) . These results indicate that miR-UL148D mediates RANTES mRNA degradation but not translational repression . We have shown that miR-UL148D encoded by an expression vector inhibits the secretion of RANTES . We determined whether the expression of RANTES is affected by miR-UL148D during viral infection . We initially utilized a mutant virus ( ToledoΔUL150 ) in which the UL150 ORF region containing miR-UL148D was deleted . The infected cells with ToledoΔUL150 showed an increased capacity to accumulate RANTES in culture media compared with the cells infected with Toledo-WT or a revertant of ToledoΔUL150 ( Figure S1 ) . To exclude the possibility that the deletion of the UL150 ORF affected RANTES level , we generated a mutant virus ( ToledoΔmiR-UL148D ) in which miR-UL148D was specifically abrogated but UL150 ORF was intact and its revertant virus ( Toledo-Revertant ) . To this aim two point mutations were engineered at the wobble positions of UL150 ORF that comprise the mature sequence of miR-UL148D ( Figure 4A ) . RNA protection analysis confirmed that miR-UL148D was expressed in HFF cells infected with Toledo-WT or Toledo-Revertant but not with ToledoΔmiR-UL148D ( Figure 4B , top panel ) . ToledoΔmiR-UL148D and Toledo-Revertant virus displayed phenotypes similar to the parental Toledo-WT in viral immediate-early ( IE ) and late ( gB ) gene expression and viral replication capacity ( Figure 4B , bottom panels and Figure 4C ) . The amount of RANTES protein secreted into culture supernatants of infected HFF cells was assessed by ELISA . We found that in ToledoΔmiR-UL148D–infected cells , the amount of accumulated RANTES was significantly higher at 24–72 h post-infection than that in the Toledo-WT–infected cells ( Figure 4D ) . qRT-PCR analysis revealed that RANTES mRNA level was also significantly higher in ToledoΔmiR-UL148D–infected cells than in Toledo-WT or Toledo-Revertant-infected cells ( Figure 4E ) . These data demonstrate that Toledo microRNA miR-UL148D inhibits RANTES secretion by mediating degradation of RANTES mRNA during infection . Peptide nucleic acid ( PNA ) , which is soluble , stable , specific to DNA or RNA and probably non-immunogenic , is known to inhibit miRNA function by an antisense mechanism through complementary binding of the PNA to the miRNA sequence [31] . To evaluate the therapeutic potential of PNA-based antisense oligonucleotides , we tested whether a miR-UL148D-specific PNA could restore RANTES expression during HCMV infection . We designed a PNA specific to miR-UL148D ( PNA-anti-miR-UL148D ) and a scrambled PNA ( PNA-control ) as a negative control . PNA-control or PNA-anti-miR-UL148D was transfected to HFF cells for 2 days before HCMV infection . At 48 h post-infection , total RNA and culture media were harvested for analysis by qRT-PCR and ELISA . In the presence of PNA-anti-miR-UL148D , secreted RANTES protein and RANTES mRNA were significantly increased ( Figure 5A and B ) . The reduction of miR-UL148D level in HFF transfected with PNA-anti-miR-UL148D was confirmed by an RNase protection assay ( Figure 5C ) . These results suggest that PNA-based antagonist to viral miRNA could be developed as a useful tool to counteract miRNA-based immune evasion strategies by viruses .
In this study , we report that infection with Toledo , a virulent clinical HCMV strain , induces degradation of RANTES mRNA in HFF cells , thereby downregulating the level of the secreted RANTES . Moreover , we show that the degradation of RANTES mRNA is mediated by HCMV-miR-UL148D , which resides in the additional DNA segment of ∼15 kb that is missing in the attenuated HCMV strains such as AD169 . Abortive HCMV added to endothelial cells is capable of both entering into the host cells and uncoating , but cannot repress the expression of RANTES [16] . In contrast , replicative HCMV infection induces cells to express significantly lower levels of RANTES [16] . These observations have led to the suggestion that HCMV encodes functional genes that can downregulate RANTES expression . In fact , HCMV appears to have developed several mechanisms for modulating RANTES at different stages of infection . The HCMV immediate-early 2 protein IE86 suppresses virus-induced RANTES expression [32] . HCMV also encodes US28 that is a receptor for RANTES . US28 binds and internalizes secreted RANTES , resulting in depletion of RANTES from the environment of HCMV-infected cells [33] , [34] . In addition , HCMV encodes pUL21 . 5 protein that functions as a decoy receptor for RANTES . By binding to the RANTES receptor , pUL21 . 5 blocks the interaction of RANTES with the RANTES receptor [35] . Moreover , HCMV significantly reduces the cell-surface expression of CC chemokine receptor 1 and RANTES receptor through the internalization of receptors [36] . The multiple strategies employed by HCMV ORF gene products for RANTES regulation paradoxically underscore a critical role for RANTES in host defense against HCMV infection . Our work identifies a viral miRNA , miR-UL148D , as a novel negative regulator for RANTES expression . To date , 14 miRNAs have been found in HCMV [24] , [25] , but their functions remain poorly understood . It is noteworthy that , among the 14 miRNAs , miR-UL148D is found only in clinical HCMV strains . Comparisons of nucleotide sequences revealed that the clinical Toledo genome contains a 15-kb DNA segment that is absent from the attenuated strain AD169 genome . This additional Toledo sequence encodes at least 19 ORFs and is found in all clinical strains [10] . This region is associated with strain-specific tissue tropism and immunopathogenesis [37] , [38] . Interestingly , miR-UL148D exists in this genomic region of clinical strains [26] . Our findings may provide an explanation for the previous seemingly contradictory observations made by others . Michelson et al . [27] showed that the levels of both RANTES mRNA and RANTES protein are increased immediately after infection of AD169 in fibroblasts . In addition , they observed that extracellular accumulation of RANTES protein is downregulated late during HCMV infection , whereas synthesis of RANTES mRNA remains unchanged . In contrast , Billstrom and Worthen [16] found that RANTES mRNA is degraded in endothelial cells infected with the clinical isolate HCMV 4010 . This result is consistent with our observations of the clinical Toledo strain ( Figure 1 ) . Our data demonstrate that the level of RANTES mRNA is downregulated in cells infected with Toledo but not with the AD169 in which an additional ∼15 kb sequence is absent . Unlike wild-type Toledo , ToledoΔmiR-UL148D virus did not cause a decrease in relative mRNA and secreted protein levels ( Figure 4 ) . Thus , our work shows that the apparent discrepancy in those studies is attributable to the genomic differences between the clinical strain and the attenuated strain , and more specifically , the presence of clinical strain-specific miR-UL148D . High-passaged laboratory strains ( AD169 and Towne ) and low-passaged clinical strains ( Toledo ) exhibit a significant difference in their pathogenicity . For instance , Toledo grows to high titers in implanted human thymus and liver tissues in SCID-hu mice , whereas AD169 failed to replicate in human thymus and liver implants [26] . The pathogenic differences between the 2 strains are believed to be due to the existence of this additional 15-kb DNA segment . Consistent with this notion , this 15-kb genomic region encodes three envelope proteins UL128 , UL130 and UL131 . They form gH/gL/UL128/UL130/UL131 protein complex and the complex is required for entry in epithelial and endothelial cells [39] . Stanton et al . reported that HCMV clinical strain Merlin acquired mutations in UL128 , UL130 or UL131 , which inhibited virus growth specifically in fibroblast cells [40] . It is noteworthy that a few genes whose functions were described among 19 ORFs to date are related to cytokines or chemokines . The UL146 and UL147 genes encoded in this region are α- ( CXC ) -chemokine viral homologues , which can influence host immune responses by impairing the trafficking of peripheral blood leukocytes , particularly neutrophils [41] . UL144 encodes a homologue of the herpes simplex virus entry mediator , a member of the tumor necrosis factor ( TNF ) -α-like receptor superfamily [42] . However , a ligand for UL144 has not yet been identified , and the function of UL144 is still unknown [43] . Given that cytokines and chemokines are the first line of host defense and subsequently govern the downstream events of immune responses , we propose that some of the additional ORFs in the 15-kb segment could possess cytokine-related functions . Although the physiological function of UL150 is unknown , it seems that at least UL150 is not directly involved in the RANTES regulation because RANTES was detected at comparable levels in both ToledoΔmiR-UL148D and ToledoΔUL150-infected cells ( Figure 4 and Figure S1 ) . Our results demonstrate that HCMV clinical strains encode a RANTES-targeting viral miRNA . This finding adds another dimension to the reported viral immunoevasive strategies by which viral proteins have been employed to subvert the host immune responses . Considering the specificity , non-immunogenicity and relative ease of manipulation of miRNA , a better understanding of the RANTES regulation by miR-UL148D could provide the means for immunosuppressive therapy .
The human embryonic kidney fibroblast cell line 293T ( 293T ) was maintained in DMEM medium ( Hyclone ) containing 7 . 5% fetal bovine serum ( HyClone Laboratories , Logan , UT ) , 2 mM L-glutamine , 50 U/mL penicillin , and 50 µg/mL streptomycin . Human foreskin fibroblast ( HFF ) cells ( passage 8–10 ) were grown in DMEM supplemented with 10% FBS under 5% CO2 at 37°C . Transient transfection was performed using Lipofectamine ( Invitrogen ) according to the manufacturer's instructions . HEK 293T cells seeded on 60-mm culture dishes were grown to 80% confluence and transfected with proper vector using Lipofectamine 2000 ( Invitrogen ) . After 24 h , the cells were harvested for the luciferase assay . Anti-RANTES monoclonal antibody ( sc-32246 ) was purchased from Santa Cruz Biotechnology ( Santa Cruz , CA ) . Anti-IE1 antibody ( MAB810R ) was purchased from Chemicon . Normal mouse IgG was purchased from Sigma-Aldrich . A BAC vector including the Toledo strain genome was introduced into human foreskin fibroblast cells ( HFF cells ) by electroporation ( Bio-Rad ) . After 3 weeks , culture supernatants were collected and used for re-infection of fresh HFF cells . After 3 weeks , culture supernatants were collected and virus titers were measured . The virus titers were determined as infectious units after the measurement of the IE1-positive cells in the infectious center assays using HFF cells . To detect the expression of RANTES , HFF cells were infected with the Toledo strain as 2 MOI ( multiple of infection ) . The full-length 3′-untranslated region ( 3′UTR ) of the RANTES gene was amplified by PCR from HFF cDNA using a pair of primers ( forward: GATGGAGAGTCCTTGAACCTGAAC; and reverse: TTTTTTTTTTATGGTTGCATTGAGAACTTT ) . Cells were co-transfected with 2 . 5 ng of the vector pGL3-Basic fused RANTES 3′UTR , 0 . 5 ng of Renilla luciferase vector , pRL-CMV ( Promega ) and a vector expressing miR-UL148D . Firefly activities were normalized for transfection efficiency using Renilla activity . After 16 h post-transfection , the cells were pelleted , washed in Dulbecco's phosphate-buffered saline and lysed in 1× passive lysis buffer ( Promega ) . The firefly and Renilla luciferase activities were determined according to manufacturer's instructions ( Dual-Luciferase Reporter Assay System , Promega ) using a luminometer ( Berthold Technologies ) . The relative luciferase activity was calculated from the firefly luciferase activity of RANTES 3′UTR/construct Renilla luciferase activity of co-transfected pRL-CMV ) / ( firefly luciferase of mock vector pGL3-Basic/Renilla luciferase activity of co-transfected pRL-CMV ) . Retroviruses encoding hairpin miRNA were constructed using the pSUPER . retro . puro vector system ( OligoEngine , Seattle , WA ) . A double-stranded hairpin oligonucleotide designed to express the miR-UL148D ( TCGTCCTCCCCTTCTTCACCG ) or miR-UL148D-mut ( TCGTGGACCCCTTCTTCACCG ) was cloned into the BglII/HindIII sites of pSUPER . retro . puro . A control miRNA was constructed , which contains a 19-bp target sequence to target the GFP molecule . Correct insertion of the oligonucleotides was confirmed by DNA sequencing . pSuper-retro viruses were produced in Phoenix-Ampho packaging cells . Extracted total RNA from HFF or 293T was resolved on 15% acrylamide , 8 M urea gel , transferred onto nylon membranes , and UV cross-linked . [γ-32P]ATP labeled antisense RNA as a probe for each miRNA was purchased from Cosmo Genetech ( Seoul , Korea ) . Hybridization was performed in the hybridization solution overnight at room temperature and the FUJI BAS film was exposed for 1 day . 5S rRNA bands stained with ethidium bromide are presented as a loading control for normalization . DNase-treated total RNA was reverse-transcribed with oligo dTs . The RT reaction was carried out at room temperature for 10 min . The reaction mixtures were then heated in a thermal cycler at 42°C for 15 min and 99°C for 5 min and then cooled at 5°C for 5 min . After adding specific primers and PCR reagents , the mixture was denatured at 95°C for 4 min . The mixture was then denatured at 95°C for 1 min , annealed at 55°C for 1 min , and extended at 72°C for 1 min . After 25 cycles , the PCR mixture was incubated at 72°C for 7 min . To confirm the identity of the amplified cDNAs , each insert product was directly sequenced using a specific primer sequence or asymmetrically cut with restriction enzymes . The primers for RT-PCR were as follows: HCMV IE1 ( forward , GTCAGGTCCACCACTGACAC; reverse , TCATATTAAAGGCGCCAGTG ) , HCMV gB ( forward , AACGCGGCTGTAAGAACTGT; reverse , ACGAGGGCATCATGGTAGTC ) , GAPDH ( forward , ATCATC CCTGCCTCTACTGG; reverse , GTCAGGTCCACCACTGACAC ) . Total RNA was isolated from Toledo-infected HFF cells at various post-infection times . The mirVana miRNA probe construction kit ( Ambion ) was used to synthesize the 32P-labeled miR-UL148D antisense probe ( CGGUGAAGAAGGGGAGGACGACCAGAG ) . This probe was designed so that an obvious size difference is detectable between the full-length undigested probe and the protected fragment after RNase digestion . Probe hybridization and RNase protection was then carried out using the mirVana miRNA detection kit ( Ambion ) according to the manufacturer's instructions . Briefly , the total RNA was incubated at 42°C for hybridization to the miR-UL148D-specific probe . After hybridization , unhybridized RNA and excess RNA probe were removed by RNase digestion . The double-stranded product was resolved in a 12% polyacrylamide , 8 M urea denaturing gel and visualized using autoradiography . RNA isolation from HCMV-infected cells was performed according to the manufacturer's instructions ( Invitrogen ) . Viral RNA was quantified by qRT-PCR using the iScript cDNA synthesis kit ( Bio-Rad ) for 2 h at 42°C . cDNA was amplified with human RANTES-specific primers ( forward , CTCATTTGCTACTGC CCTCTGCGCTCCTGC; reverse , GCTCATCTCCAAAGAGTTGATGTACTC ) , HCMV IE1-specific primers ( forward , GTCAGGTCCACCACTGACAC; reverse , TCATATTAAAGGCGCCAGTG ) , and GAPDH-specific primers ( forward , ATCATC CCTGCCTCTACTGG; reverse , GTCAGGTCCACCACTGACAC ) . Quantification of HCMV RNA was normalized with GAPDH RNA as an internal control . Reactions were performed by the Lightcycler PCR ( Roche ) using the following program: 50°C for 30 min , 95°C for 15 min and 45 cycles as follows: 95°C for 10 s , 53°C for 10 s , and 72°C for 25 s . The online target prediction algorithm RNA22 ( http://cbcsrv . watson . ibm . com/rna22 . html ) and RNAhybrid ( http://bibiserv . techfak . uni-bielefeld . de/rnahybrid/submission . html ) were used to predict potential target sites . HCMV 14 miRNAs were used as target miRNAs [24] , [44] . To generate the deletion mutant virus of the miR-UL148D ( ToledoΔmiR-UL148D ) , UL150 region encoding miR-UL148D was deleted in the Toledo bacterial artificial chromosome ( BAC ) using rpsL-neo cassettes . Briefly , rpsL-neo cassettes were PCR amplified by using primers containing homology arms consisting of 50 nucleotides upstream and downstream of the target gene plus 24 nucleotides homologous to the rpsL-neo cassette . The amplified DNA fragments were introduced into E . coli DH10b cells containing wild-type Toledo-BAC for recombination by electroporation using Gene Pulser II ( Bio-Rad ) . The intermediate Toledo-BAC construct containing the rpsL-neo cassette was selected on Luria broth ( LB ) plates containing kanamycin . UL150 region in Toledo-BAC was destroyed by insertion of rpsL-neo cassette . We also generated Toledo-Revertant BAC by the same method . Deletion of miR-UL148D was confirmed by RNase protection assay using specific probe . PNA miRNA inhibitors were purchased from Panagene Inc . The miR-UL148D-specific inhibitor sequence is 5′-RRRQRRKKR-OO-GTGAAGAAGGGGAGGACG-3′ and the scrambled control PNA sequence is 5′-RRRQRRKKR-OO-TAGAGCTCCCTTCAATCCAAA-3′ . One day before transfection , HFF cells were seeded onto a 24-well plate in appropriate complete growth medium without antibiotics . PNAs transfection was performed using Dharmafect 1 reagent ( Dharmacon ) . PNA miRNA inhibitor was added into the culture medium at a final concentration of 500∼2000 nM . Cells were incubated at 37°C for 48 h prior to viral infection . A miRNA isolation kit ( Invitrogen ) was used to isolate miRNA from total RNA . ELISA was used to measure the expression level of secreted RANTES . A RANTES ELISA kit was purchased from Thermo Fisher Scientific Inc . ( Rockford , IL ) . Culture media of transfected or infected cells were collected and incubated with RANTES antibody on the plate . Each well was washed with the washing buffer provided in the ELISA kit . After incubation of HRP-conjugated antibody , TMB substrate was added into each well . Absorbance was measured using a microplate reader , iMARK ( Bio-Rad ) at a wavelength of 450 nm . Cells were labeled with [35S]methionine and [35S]cysteine and lysed in 1% NP-40 in PBS supplemented with protease inhibitors . After preclearing , samples were incubated with the appropriate antibodies for 2 h at 4°C before Protein G-Sepharose beads were added . Beads were washed four times with 0 . 1% NP-40 in PBS and bound proteins were eluted by boiling in SDS sample buffer . Proteins were separated by SDS-PAGE .
|
Unlike the attenuated HCMV strain AD169 , the clinical isolates of HCMV , including the Toledo strain , are virulent and can cause disease in healthy adults . Toledo differs from AD169 in that Toledo contains a 15-kb DNA segment , encoding at least 19 ORFs and a single microRNA known as miR-UL148D . This 15-kb segment is believed to be a major determinant of the virulence and pathogenicity of the Toledo clinical strain . The CC–chemokine RANTES recruits immune cells during viral infection , suggesting that it may play a role in virus-related diseases . Here , we show that RANTES mRNA was degraded in human foreskin fibroblast cells during infection with Toledo but not during infection with AD169 . The degradation of RANTES mRNA was mediated by miR-UL148D , the only viral microRNA predicted from the 15–kb segment of the Toledo genome . Accordingly , the levels of secreted RANTES in infected cells with ToledoΔmiR-UL148D in which miR-UL148D was deleted were higher than those in infected cells with Toledo . Our results reveal that a viral microRNA could be a novel potential therapeutic target and provide important insights into understanding the differences in pathogenic potential between clinical and attenuated strains .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"immunology",
"biology",
"microbiology"
] |
2012
|
Human Cytomegalovirus Clinical Strain-Specific microRNA miR-UL148D Targets the Human Chemokine RANTES during Infection
|
Human African trypanosomiasis ( HAT ) , a major parasitic disease spread in Africa , urgently needs novel targets and new efficacious chemotherapeutic agents . Recently , we discovered that 4-[5- ( 4-phenoxyphenyl ) -2H-pyrazol-3-yl]morpholine ( compound 1 ) exhibits specific antitrypanosomal activity with an IC50 of 1 . 0 µM on Trypanosoma brucei rhodesiense ( T . b . rhodesiense ) , the causative agent of the acute form of HAT . In this work we show adenosine kinase of T . b . rhodesiense ( TbrAK ) , a key enzyme of the parasite purine salvage pathway which is vital for parasite survival , to be the putative intracellular target of compound 1 using a chemical proteomics approach . This finding was confirmed by RNA interference experiments showing that down-regulation of adenosine kinase counteracts compound 1 activity . Further chemical validation demonstrated that compound 1 interacts specifically and tightly with TbrAK with nanomolar affinity , and in vitro activity measurements showed that compound 1 is an enhancer of TbrAK activity . The subsequent kinetic analysis provided strong evidence that the observed hyperactivation of TbrAK is due to the abolishment of the intrinsic substrate-inhibition . The results suggest that TbrAK is the putative target of this compound , and that hyperactivation of TbrAK may represent a novel therapeutic strategy for the development of trypanocides .
Trypanosomiases belong to the major parasitic diseases spread throughout the world . Human African Trypanosomiasis ( HAT; sleeping sickness ) is a vector-borne parasitic disease transmitted by a protozoan parasite of the genus Trypanosoma via the bites of infected tsetse flies . Two different forms of HAT are known . The chronic form is caused by Trypanosoma brucei gambiense ( T . b . gambiense ) infection whereas Trypanosoma brucei rhodesiense ( T . b . rhodesiense ) is responsible for the acute form of the disease . Both forms of HAT develop in two stages . After the bite , parasites first distribute to the blood , lymph and peripheral organs ( stage 1 ) , then spread to the central nervous system ( stage 2 ) where they cause serious neurological disorders . Leaving infected people untreated , HAT is invariably fatal . Today , sleeping sickness threatens millions of people in 36 countries of sub-Saharan Africa [1] , [2] . The estimated number of deaths annually is currently between 50’000 and 70’000 [3] . Four drugs ( pentamidine , melarsoprol , eflornithine and suramin ) are registered for the treatment of sleeping sickness and provided free of charge to endemic countries through a World Health Organization ( WHO ) private partnership . The type of treatment depends on the stage of the disease . Pentamidine and suramin are used for first stage treatment of T . b . gambiense and T . b . rhodesiense sleeping sickness , respectively . The drugs capable to treat the second stage infections are more toxic and complicated to administer , and they need to cross the blood-brain barrier to reach the parasites within the central nervous system . Melarsoprol , which is effective against both forms of HAT , derives from arsenic and has many severe side effects , the most dramatic ( prevalence 5 to 15% ) being a reactive encephalopathy ( encephalopathic syndrome ) which can be fatal in 3–10% of affected patients . An increase of resistance to the drug has been observed in several foci particularly in central Africa . Eflornithine is less toxic than melarsoprol , but it is only effective against T . b . gambiense sleeping sickness , and a strict and complicated regimen has to be applied . It is evident that therapy of HAT relies on few drugs which are associated with severe side effects . There has been a revival of drug research and development regarding HAT compared to the last 15 years , and a number of drug development projects are currently ongoing . Unfortunately , the development of the only compound ( pafuramidine ) having advanced to phase III clinical trials [4] , [5] for stage one treatment was discontinued [6] . Thus , novel targets and new efficacious chemotherapeutic agents are urgently needed . Recently we reported the synthesis and evaluation of new 4-[5- ( 4-phenoxy-phenyl-2H-pyrazol-3-yl]morpholine derivatives against several parasites [7] . One of the compounds , 4-[5- ( 4-phenoxy-phenyl-2H-pyrazol-3-yl]morpholine ( Fig . 1 , compound 1 ) , exhibited good activity toward T . b . rhodesiense with an IC50 of 1 µM and low cytotoxicity [7] . This finding prompted us to address the question regarding the cellular target and the molecular mechanism underlying the observed toxicity toward the parasite . To this end , here we report a chemical proteomics approach that led to the identification of adenosine kinase as the putative target . Subsequent biochemical and biophysical characterization with respect to compound 1 binding as well as drug sensitivity tests on the corresponding knock-down strain allowed its validation and suggested hyperactivation of adenosine kinase as the molecular mechanism underlying the biological activity .
The parasites collected from mouse blood were lysed with lysis buffer consisting of 20 mM Hepes , 150 mM NaCl , 1% Triton X-100 , 2 mM tris ( 2-carboxyethyl ) phosphine ( TCEP ) , 10% glycerol , 1 µl/ml protease inhibitor cocktail , at pH 7 . 5 . To this end , the suspension of the purified cells ( 2 . 56×109 trypanosomes ) in 160 µl PBS were treated with 40 µl lysis buffer concentrate ( five fold ) . After short intervals of sonication the mixture was centrifuged for 10 min at 14’000 g , the supernatant was recovered and stored in aliquots at -20°C . The total protein concentration was measured using Bradford dye assay [8] , and the lysate was further diluted to the desired concentration using lysis buffer . All protocols and procedures used in the current study were reviewed and approved by the local veterinary authorities of the Canton Basel-Stadt ( Switzerland ) . Compounds 2 , 3 , and 4 ( Fig . 1 ) were immobilized via the primary or secondary amino groups ( Fig . S1 ) on epoxy-activated agarose providing a 12-atoms spacer ( 1 , 4-bis ( 2∶3-epoxypropoxy ) butane ) . Swollen and thoroughly washed matrix was resuspended in two to three volumes of 12 . 5–50 mM ligand dissolved in 50% DMF/50 mM Na2CO3 , 50 mM NaCl , pH 9 . 5 . Coupling was performed for 16–72 h at 25–40°C . After 4 to 5 washes with 50% DMF/50 mM Na2CO3 , 50 mM NaCl , pH 9 . 5 , remaining reactive groups were blocked with 1 M ethanolamine ( pH 8 ) and thoroughly washed with low/high pH buffers . A control matrix was prepared without ligand and treated as described above . Direct absorbance scans of the immobilized ligand on the matrix resuspended in 50% glycerol solution ( v/v ) clearly confirmed successful coupling ( data not shown ) . The amount of inhibitor bound to the matrix was determined by back calculation of amount of compound applied and amount recovered by UV determination . Routinely 20–45 µmol/ml compound were bound . The resin was incubated in 2 . 5 volumes of T . b . rhodesiense total cell lysate ( 2 mg/ml protein ) and incubated at 4°C for 2 hours . After washing with lysis buffer ( four times ) , the matrix was heated at 95°C for 5 min with Laemmli sample buffer , directly loaded and separated by 12% SDS-PAGE . Resolved proteins were visualized by silver staining . The control matrix was incubated with the same amount of lysate and treated equally . Proteins bands were excised from the gels and fragments resulting from trypsin digestion were analyzed by LC/ESI/MS/MS-QTOF mass spectrometry . Database searches were performed by using the ProteinLynx Global Server ( all species ) and Mascot ( other eukaryotes ) search programs . A score >46 indicates identity ( p<0 . 05 ) . TbrAK catalyzes the ATP-dependent phosphorylation of the 5′-hydroxyl moiety of adenosine to form AMP and ADP . The activity assay was based on HPLC separation of substrate and products ( adenosine , AMP , ADP , and ATP ) . The mobile phase consisted of 35 mM KH2PO4 , 6 mM tetrabutylammonium hydrogensulfate , ( pH 7 . 0 , adjusted with KOH ) , 125 mM EDTA , and 1% ( v/v ) acetonitrile . The separation was performed isocratically at a flow-rate of 1 . 5 ml/min and monitored at 254 nm . The standard reaction in 75 µl total volume containing 0 . 5 mM ATP , 0 . 5 mM adenosine , 0 . 5 mM MgCl2 , 1% DMSO in buffer E1 ( 20 mM Hepes , 150 mM NaCl , pH 7 . 0 ) was started with the addition of 2 µg TbrAK ( final concentration 0 . 7 µM ) and incubated at 37°C for 10 min while shaking vigorously ( 600 rpm ) . To measure the influence of compound 1–5 on TbrAK activity , DMSO was replaced in the assay mixture by compound 1–5 so that the final concentration of 1% DMSO was maintained while the concentration of compound could be varied in the range of 0–1 mM depending on solubility . All reactions were stopped by adding 75 µl mobile phase . 50 µl of the sample volume were injected for analysis and the resulting ADP/ATP ratios were calculated and used as a measure of activity . For comparative reasons the activity recorded in absence of compound was set to 100% . The mean of three independent measurements are reported . Isothermal titration calorimetry ( ITC ) was employed to elucidate the binding affinity of compound 1 toward TbrAK . All solutions were degassed for 5 min with gentle stirring under vacuum . TbrAK was dialyzed into buffer E2 ( 20 mM Tris , 150 mM NaCl , pH 7 . 5 , 5% glycerol ) , diluted to 7 µM and supplemented with 1% DMSO , filtered and filled into the sample cell . The ligand solution was prepared by diluting a stock solution of compound 1 or 5 in DMSO with buffer E2 to give a final concentration of 150 µM and 1% DMSO . The final ligand solution was filtered before use . A titration experiment consisted of a first control injection of 1 µl followed by 27 injections , each of 10 µl and 20 s duration , with a 4-min interval in between . Raw data were collected , corrected for ligand heats of dilution , integrated and fit to the two-set non-interacting binding sites model using the Origin® software supplied with the instrument . The measurements were performed at least in duplicate . Kinetic constants for adenosine phosphorylation by TbrAK were obtained by monitoring the conversion of [2-3H]adenosine to [2-3H]AMP . Reactions were executed at 37°C in a final volume of 30 µl containing 20 mM Hepes ( pH 7 . 0 ) , 50 mM NaCl , 167 µM ATP , 167 µM MgCl2 , 0 . 34% BSA , and 0 . 34% DMSO . To measure the influence of compound 1 to kinetic parameters , DMSO was omitted in the assay mixture and 1 was added from a 10 mM stock solution in DMSO to give a final concentration of 33 . 4 µM and 0 . 34% DMSO . The concentration of [2-3H]adenosine ( 0 . 1–12 µM ) was chosen in consideration of Michaelis–Menten conditions for initial velocity measurements . The reaction was started by the addition of 1 ng ( 0 . 88 nM ) TbrAK . The reactions were incubated and 5 µl aliquots were spotted on 5 mm diameter DE81-cellulose disks placed in a 96-well plate to stop the reaction . The disks were washed 3 times with 250 µl 5 mM ammonium formate , once with H2O , transferred to scintillation vials , and then soaked with 2 ml of a displacement solution ( 100 mM HCl , 200 mM KCl ) and gently shaken for 1 minute to elute the phosphorylated products [9] . After adding 10 ml scintillation liquid , the samples were counted in scintillation counter . Km and Vmax with respect to adenosine and Ki for adenosine substrate-inhibition in absence and presence of compound 1 were determined by non-linear fit of the data to the substrate-inhibition model described elsewhere [10] . The results are based on three independent series with each data point measured in triplicates . The construction of the T . b . brucei adenosine kinase gene knock-down mutant has been described recently [11] . Expression of a stem-loop construct targeting TbAK was induced by addition of 10 µg/ml tetracycline to the medium . Down-regulation of TbAK expression was verified by Northern blotting . Parasites were cultivated and drug sensitivity was tested using the Alamar blue test [12] . IC50 values were calculated by nonlinear fitting to the sigmoidal dose-response curve using Origin® software . The assays were performed in triplicates for at least four times . Origin of supplies as well as protocols for purification of recombinant proteins , the spectroscopic assay for TbrGAPDH , the radiometric assay for TbrAK , and the thermal denaturation assay are described in the Supporting Information ( Text S1 ) .
In order to isolate and identify the intracellular target ( s ) of compound 1 in T . b . rhodesiense we decided to immobilize the compound and to isolate their target ( s ) by an affinity chromatography approach . The binding mode of 1 to the putative target ( s ) being unknown , we immobilized three derivatives of compound 1 ( Fig . 1 , compound 2–4 ) , each containing an additional primary or secondary amino group at different positions [7] , and linked them to epoxy-activated agarose . The primary aromatic amine in compound 2 and 3 , and the secondary amino group in the piperazine moiety of compound 4 react with the terminal epoxy group of the activated agarose , providing an uncharged , hydrophilic and very stable 12-atoms spacer ( see Fig . S1 ) . The direct linkage of compound 1 to epoxy-activated agarose via the pyrazole moiety was not likely to occur under the applied experimental conditions due to the weak nucleophilicity of the heterocycle for epoxide-based alkylation which in general requires rather drastic conditions , i . e . prolonged heating at high temperature with the concomitant use of a strong base [13] . Direct absorbance scans of the immobilized ligand on the matrix resuspended in 50% glycerol solution ( v/v ) confirmed successful coupling ( data not shown ) . Routinely 20–45 µmol/ml compound were bound . Our recent evaluation of compound 1 and its derivatives clearly indicated the morpholine moiety to be important for the biological effect [7] which is corroborated by the fact that compound 5 ( morpholine removed ) only exhibits very low antiparasitic activity ( IC50 of >65 µM ) . Therefore compound 5 was chosen as negative control . The total parasite lysate was prepared , loaded on the matrices and unbound material was removed as outlined in the Methods section . The bound proteins were separated by SDS-PAGE and visualized by silver-staining ( Fig . 2 ) . The use of compound 2 derived matrix led to the detection of three distinct protein bands , two of which at higher molecular weights also being present in the control ( Fig . 2A ) . Trypsin digestion and peptide sequencing by LC/ESI/MS/MS-QTOF mass spectrometry identified the specific band at ∼40 kDa as adenosine kinase of T . b . rhodesiense ( TbrAK , expected size of 38 . 1 kDa; score 141 ) , whereas the other two corresponded to human keratin ( score 392 and 752 ) . Affinity matrix prepared with compound 3 did not isolate a protein different to the back ground signals identified as human keratin ( Fig . 2B ) . In contrast , immobilized derivative 4 behaved differently , and more distinct proteins were isolated . Subsequent protein analysis found human keratin in most cases . One distinct protein band not visible in the control was identified as glycosomal T . b . rhodesiense glyceraldehyde-3-phosphate dehydrogenase ( TbrGAPDH , expected size of 39 . 3 kDa; score 87 ) ( Fig . 2C ) . Protein bands isolated on the control matrices were considered to bind unspecifically and were not further examined . For chemical validation of both TbrGAPDH and TbrAK , proteins were produced recombinantly in sufficient amounts . The gapdh gene , kindly provided by Prof . Paul A . M . Michels ( University of Louvain , Brussels , Belgium ) , was subcloned into an expression vector that yielded 10–15 mg of highly pure ( >99% ) TbrGAPDH per liter culture ( Fig . S2A ) . Since the ak is a tandem gene coding for two almost identical ( 99% identity , 4 out of 345 amino acids different ) adenosine kinases ( SwissProt entries Q584S0 and Q584S6 ) it was necessary to determine which one had been isolated . The detailed analysis of the fragments resulting from the trypsin digestion revealed that peptide DIESTVLATK can be unambiguously allocated to the Q584S0 sequence , thus this isoform of TbrAK was cloned , expressed and purified to homogeneity and resulted in 40–50 mg of soluble and highly pure ( >99% ) TbrAK per liter culture ( Fig . S2B ) . Both TbrGAPDH and TbrAK were active and could be used for subsequent analyses . TbrAK activity in absence and presence of the compounds was analyzed using an HPLC protocol that allowed the separation of adenosine , AMP , ADP , and ATP . Surprisingly , the assay revealed a 2 . 5-fold ( absolute value: 245±3% ) and 1 . 7-fold ( 174±5% ) increase of enzyme activity in presence of 50 µM of compound 1 and 2 , respectively ( Fig . 3; Fig . S3 ) . A less pronounced but still significant effect ( two-tailed t-test , p<0 . 01 ) was observed for compound 3 and 4 at 50 µM with an apparent increase of activity to 110±1% and 109±1% , respectively . The negative control ( compound 5 ) did not have this activating effect on TbrAK activity ( see Fig . S3 ) . To validate the activity measured at a single concentration , the concentration dependence of the effect was analyzed . The hyperactivation effect is found to be concentration dependent for compounds 1 , 2 and 4 ( Fig . 3 ) . The results clearly show that under the experimental conditions a half maximal effective concentration ( EC50 ) of 38 . 9±0 . 9 µM for compound 1 is determined . Although the solubility limits prevented any EC50 determination regarding compounds 2 and 4 ( Fig . 3 ) , the concentration dependence , thus the specificity of the effect is proven also for these compounds . Interestingly , although compound 3 appeared to activate TbrAK at a low level ( 110±1% ) it does not exhibit any concentration dependence when assayed up to concentrations corresponding to its maximum solubility , thus behaved like the inactive control compound 5 ( Fig . 3 ) . The EC50 of found for compound 1 seemed to contrast the IC50 of 1 . 0 µM on the parasite . We reasoned that the experimental conditions of the HPLC assay may be at the origin of the high EC50 value , thus we analyzed the activating effect of compound 1 using the more sensitive radiometric assay , which allowed the analysis at low substrate and enzyme concentrations ( Supporting Information Text S1 ) , as an orthogonal method to the HPLC assay . Indeed , the hyperactivation effect was concentration dependent with an EC50 of 38±12 nM for compound 1 ( Fig . S4 ) . This value is well in line with the observed affinity ( KD of 75±20 nM and 497±34 nM for the high and low affinity binding site , respectively ) of compound 1 towards TbrAK , and the observed IC50 of 1 . 0 µM in the in-vitro whole cell assay . In contrast to adenosine kinase , TbrGAPDH activity was not influenced by any of the compounds with respect to forward and reverse reaction ( Fig . S5 ) when measured at concentrations up to 50 µM ( limit given due to strong absorption at the wavelength applied ) , suggesting that TbrGAPDH could be considered as a false positive hit . Specific binding of compound 1 with respect to the isolated proteins was analyzed by two methods . Thermal stability of proteins and protein complexes can be evaluated by CD spectroscopy . Due to the fact that the complex of a ligand bound to the native conformation of a protein will have a higher thermal stability than the empty protein [14] , an increased melting point ( Tm ) of the protein in presence of a ligand will give evidence for specific binding . Indeed , apo TbrAK melted at 43 . 1±0 . 4°C and was stabilized by the control compound adenosine with a ΔTm of 7 . 6°C ( Tm 50 . 7±0 . 2°C ) ( Table 1 ) . In a similar way , compound 1 and 2 were able to increase thermal stability by a ΔTm of 4 . 8°C ( Tm 47 . 9±0 . 1°C ) and a ΔTm of 2 . 8°C ( Tm of 45 . 9±0 . 4°C ) , respectively , thus confirming specific binding for compound 1 to the enzyme ( Table 1 ) . Compound 3 , 4 , and 5 did not stabilize the protein when added at 50 µM ( Table 1 ) . As an orthogonal method to CD we used isothermal titration calorimetry ( ITC ) to analyze binding of compound 1 to TbrAK . The titration of TbrAK with compound 1 revealed a complex and enthalpy driven binding mode with two molecules of 1 binding to one molecule of enzyme . Exothermic binding heats were corrected for heats of dilution , integrated and plotted against the molar ratio of 1 and enzyme ( trace I in Fig . 4 ) . The binding isotherm was best described using a non-linear least square fit assuming a two-sites non-interacting binding model . Subsequent analysis revealed a high affinity binding site with a KD of 75±20 nM and a ΔHbind of −3 . 05±0 . 77 kcal/mol , while the low affinity site exhibited a KD of 497±34 nM and a ΔHbind of −1 . 13±0 . 24 kcal/mol . The negative control using compound 5 showed no specific heat release ( see trace II in Fig . 4 ) . In contrast to TbrAK , TbrGAPDH was not significantly stabilized by any of the compounds , while in presence of the substrate NAD+ or DL-GAP thermal stability increased ( see Table S1 ) . In agreement with above TbrGAPDH activity measurement , no binding of the compounds to the protein was observed , clearly indicating that TbrGAPDH is a false positive hit . Therefore , this enzyme was not further validated . A common property of adenosine kinases from various organisms is their control via a substrate-inhibition mechanism [15]–[18] , and recently it was shown that T . brucei AK was inhibited by high adenosine concentrations to prevent non-physiologically high intracellular purine nucleotide levels [19] . This prompted us to analyze the influence of compound 1 toward TbrAK with respect to substrate transformation by determining the kinetic parameters in absence and presence of the activator . TbrAK activity was measured at increasing adenosine concentrations and a fixed concentration of 167 µM ATP . All data are given in Table 2 . Adenosine kinetics in absence of compound 1 displayed non-hyperbolic progress plots ( Fig . 5 ) . After increasing activity up to a maximum at 2–3 µM adenosine , the enzyme activity declined at higher substrate concentrations which is a typical finding for substrate-inhibition . Thus the observed kinetic data were fit to the substrate-inhibition model [10] . Indeed , a good fit was obtained for all data points , yielding a Km of 0 . 99±0 . 05 µM and a Vmax of 19 . 60±0 . 33 nM/min . kcat and catalytic efficiency were found to be 0 . 37±0 . 01 s−1 and 0 . 38±0 . 01 µM−1 s−1 , respectively . TbrAK was inhibited by adenosine with a Ki of 6 . 1±1 . 4 µM . In contrast , in presence of compound 1 substrate-inhibition was strongly reduced ( Fig . 5 ) , and adenosine inhibited TbrAK with a more than ten-fold increased Ki value ( 78 . 4±2 . 2 µM ) . While Km ( 0 . 65±0 . 04 µM ) , Vmax ( 14 . 20±0 . 14 µM ) and kcat ( 0 . 0 . 27±0 . 01 s−1 ) were slightly reduced in presence of compound 1 , the catalytic efficency kcat/Km ( 0 . 42±0 . 01 µM−1 s−1 ) did not change significantly ( Table 2 ) . The proposed mechanism of action for compound 1 , hyperactivation of TbrAK , implies that down-regulation of TbrAK activity counteracts compound 1 toxicity . This hypothesis was addressed by RNAi-mediated silencing of TbAK expression , using bloodstream-form T . b . brucei that express in a tetracycline-inducible manner a stem-loop construct targeted against TbAK [11] . TbAK knock-down has been confirmed recently on the RNA and protein level [11] , [19] . The addition of tetracycline reduced the sensitivity of TbAK RNAi cells to compound 1 , raising the IC50 from 131±43 nM to 271±25 nM ( two-tailed t-test , p<0 . 05; Fig . 6 ) . The negative control ( compound 5 ) did not exhibit any IC50 difference in the induced and non-induced TbAK RNAi cells .
A chemical proteomics approach was applied to isolate and identify the cellular target of compound 1 . To this end , three of its amino-derivatives where immobilized , and the corresponding affinity matrixes were used to pull down potential targets . Indeed , this approach afforded multiple proteins . Most of them could be detected in both the control and the isolation experiment ( Fig . 2 ) and thus could be considered as unspecific bound protein . Amongst other , human keratin was the predominant contamination which turned out to be present under any circumstances , and it was even found in new batches of matrix that were stored in sealed containers before first use . Remarkably , we observed differences regarding inter- and intrabatch reproducibility . Experiments were always reproducible with respect to background when the same batch of activated matrix was used while changing to a new batch led to altered background ( see lane a in Fig . 2A , B , C ) . However , two distinct proteins not appearing in the control reaction were identified as TbrGAPDH and TbrAK , two trypanosomal enzymes that are involved in parasite glycolysis and purine salvage , respectively . Both enzymes were cloned and recombinantly expressed . The subsequent chemical validation of both enzymes revealed that TbrGAPDH activity was neither affected by compound 1 ( Fig . S5 ) nor specific binding of 1 toward the enzyme was observed ( Table S1 ) , suggesting TbrGAPDH to be a false positive hit . As a matter of fact , due the highly charged nature and its high abundance in the parasite this particular target has a strong tendency to be isolated by affinity chromatography [20]–[22] . In contrast and unexpectedly , the HPLC activity assay revealed that compound 1 strongly activated TbrAK activity . A similar effect was measured for compound 2 , while derivatives 3 and 4 only exhibited very low though significant activity toward TbrAK ( Fig . S3 ) . Compound 1 , 2 and 4 activated TbrAK in a concentration dependent manner , yielding an EC50 of 38 . 9±0 . 9 µM for compound 1 ( Fig . 3 ) . While the EC50 of 2 and 4 could not be determined due to solubility problems , it can be estimated that they would be higher than the one observed with compound 1 ( Fig . 3 ) . The observed data for compound 1 and 2 correlate well with formerly determined in-vitro data that showed an IC50 of 1 . 0 µM and 17 . 7 µM against blood stage T . b . rhodesiense [7] , and also with the fact that compound 2 was able to bind the target . TbrAK activation by derivative 4 was concentration dependent ( EC50>750 µM , Fig . 3 ) which corresponds very well to the low activation capacity at 50 µM ( 109±1% , Fig . S3 ) , ( see text below for discussion of this issue ) . Specific binding could be verified for compounds 1 and 2 by the thermal stability assay as shown by the Tm increase ( Table 1 ) . For the least active compounds ( 3 and 4 ) no stabilization was observed . While this finding is in line with the lack of activity at all concentrations tested for compound 3 , it seems to contrast the concentration dependence of derivative 4 . However , taking into consideration the small activating effect , the concentration dependence and the unchanged Tm in presence of compound 4 , it is likely to assume that this derivative binds with low affinity to the target . Unfortunately , the limited solubility prevented further investigations at concentrations >750 µM regarding Tm as well as binding affinity due to the technical limits given by CD spectroscopy and ITC . As expected the negative control ( compound 5 ) did neither activate TbrAK nor improve thermal stability , giving further evidence for specific binding of compound 1 and 2 . In addition , binding of compound 1 to TbrAK could be confirmed by ITC which revealed enthalpy driven high affinity interaction . Interestingly , it appears that two molecules of the activator are bound per molecule of enzyme . Under the same conditions control compound 5 did not bind to the enzyme and produced only unspecific heat signals related to protein and ligand dilution in the sample cell ( Fig . 4 ) . On the target level it is evident that the additional amino group in derivative 2 only moderately interferes with the activating effect , thus binding to TbrAK seems disturbed but still productive . In contrast , replacing the oxygen in the morpholine moiety by nitrogen ( compound 4 ) strongly increases EC50 at least 20 fold , suggesting that binding to TbrAK is severely impaired with this derivative . Finally , a substitution on the pyrazole moiety ( compound 3 ) leads to very low and concentration independent activity . These observations are well in line with the fact that none of these two derivatives stabilized TbrAK in the thermal stability assay ( Table 1 ) . On the parasite level the situation is different . Regarding derivative 3 , the very low activity toward the isolated enzyme while exhibiting an IC50 of 10 . 3 µM on the parasite [7] , leads to the conclusion that toxicity is not conferred by TbrAK but by another yet to be determined target . For compound 4 the situation appears more complex . The observed concentration dependence gives strong evidence for it acting against TbrAK , but its potentially low affinity and the low activation capacity ( EC50 >750 µM ) toward the target would lead to reduced trypanocidal activity . However , derivative 4 is highly potent in the parasite assay ( IC50 of 1 . 1 µM ) [7] . It is therefore likely that the trypanocidal effect of compound 4 is based on a combination of target specific activity , and/or possible off-target activity , and increased general toxicity [7] . In addition , increased accumulation within the parasite compared to compound 1 could contribute to the strong effect , while reduced cell uptake may explain the difference in potency of derivative 2 on the isolated target . Thus , taking into consideration the above observations and the complete lack of activity of the control compound 5 in all experiments , we can state that the intact morpholine/pyrazole moiety represents an important part of the pharmacophore , while the introduction of basic amino groups may impair the activating effect due to altered physico-chemical properties on target and/or parasite level . To further validate TbrAK as the potential target of compound 1 we investigated parasite sensitivity toward compound 1 under conditions of reduced intracellular adenosine kinase levels . To this end , parasite viability of a knock-down mutant was measured . Indeed , as expected for a mechanism of action based on overactivation of TbrAK by compound 1 , the sensitivity of TbrAK silenced parasite cells decreased as shown by the IC50 raising from 131±43 nM to 271±25 nM ( two-tailed t-test , p<0 . 05 ) . Although the applied tetracycline-inducible system is leaky and not capable to provide a complete knock-out [11] the observed difference is statistically significant , demonstrating that the toxic effect of compound 1 is adenosine kinase dependent . Although we cannot rule out that compound 1 could interact with other targets in the cell that may not be amenable to the pull down approach ( e . g . cytoskeleton , DNA , interference with mitochondrial electron transport ) , the line of evidence strongly supports TbrAK to represent the putative cellular target . A first step toward the elucidation of the mechanism of action with respect to the activating effect was accomplished by determination and analysis of kinetic parameters with respect to substrate transformation in absence and presence of the activator 1 . TbrAK activity is strongly inhibited by its substrate adenosine ( Ki of 6 . 1±1 . 4 µM ) , thus it follows substrate-inhibition kinetics . This is a common characteristic of adenosine kinases isolated from various sources [15]–[18] and has also been described for TbrAK very recently [19] . Under the experimental conditions applied , the Ki of adenosine increased more than ten-fold to a value of 78 . 4±2 . 2 µM when compound 1 was present , resulting in strongly reduced substrate-inhibition ( Fig . 5 ) while the kinetic parameters remained almost identical . Taken together , the results suggest that the mechanism for trypanocidal activity functions via hyperactivation of adenosine kinase . There are two principally different explanations for the toxic effect of hyperactivation . Uncontrolled TbrAK activity could lead to purine imbalance within the parasite , thus interfering with the vital purine salvage pathway , the nucleotide pool and subsequently nucleic acid formation . Precedence for this mechanism comes from E . coli , where high adenine concentrations cause a cytotoxic [ATP]/[GTP] imbalance [23] . Alternatively , excessive adenosine kinase activity may use up the existing adenosine/ATP pools and lead to adenosine depletion and ATP burn-out . Interestingly , cell death upon hydrolysis of ATP reserves due to mislocalization of glycosomal hexokinase to the cytosol has already been observed for T . brucei [24] , [25] . Whereas hyperactivation as a mechanism of action is well known from drugs targeting cell signaling , e . g . acetylcholine receptor agonists , it represents a novel and hitherto unexplored concept for compounds targeting metabolic enzymes . The cytotoxic hyperactivation of adenosine kinase does not only provide an opportunity for the chemotherapy of sleeping sickness , but when explored against other pathogens or tumor cells , hyperactivation of metabolic key enzymes may well find further pharmacological applications .
|
Human African trypanosomiasis ( HAT ) , a devastating and fatal parasitic disease endemic in sub-Saharan Africa , urgently needs novel targets and efficacious chemotherapeutic agents . Recently , we discovered that 4-[5- ( 4-phenoxyphenyl ) -2H-pyrazol-3-yl]morpholine exhibits specific antitrypanosomal activity toward T . b . rhodesiense , the causative agent of the acute form of HAT . Here we applied a chemical proteomics approach to find the cellular target of this compound . Adenosine kinase , a key enzyme of the parasite purine salvage pathway , was isolated and identified as compound binding partner . Direct binding assays using recombinant protein , and tests on an adenosine kinase knock-down mutant of the parasite produced by RNA interference confirmed TbrAK as the putative target . Kinetic analyses showed that the title compound is an activator of adenosine kinase and that the observed hyperactivation of TbrAK is due to the abolishment of the intrinsic substrate-inhibition . Whereas hyperactivation as a mechanism of action is well known from drugs targeting cell signaling , this is a novel and hitherto unexplored concept for compounds targeting metabolic enzymes , suggesting that hyperactivation of TbrAK may represent a novel therapeutic strategy for the development of trypanocides .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"chemical",
"biology",
"pharmacology",
"biochemistry/protein",
"chemistry",
"biochemistry/drug",
"discovery"
] |
2009
|
Adenosine Kinase of T. b. rhodesiense Identified as the Putative Target of 4-[5-(4-phenoxyphenyl)-2H-pyrazol-3-yl]morpholine Using Chemical Proteomics
|
Bothrops , Crotalus and Lachesis represent the most medically relevant genera of pitvipers in Central and South America . Similarity in venom phenotype and physiopathological profile of envenomings caused by the four nominal Lachesis species led us to hypothesize that an antivenom prepared against venom from any of them may exhibit paraspecificity against all the other congeneric taxa . To assess this hypothesis , in this work we have applied antivenomics and immunochemical methods to investigate the immunoreactivity of three monovalent antivenoms and two polyvalent antivenoms towards the venoms from different geographic populations of three different Lachesis species . The ability of the antivenoms to neutralize the proteolytic , hemorrhagic , coagulant , and lethal activities of the seven Lachesis venoms was also investigated . A conspicuous pattern of immunorecognition and cross-neutralization for all effects was evident by the polyspecific antivenoms , indicating large immunoreactive epitope conservation across the genus during more than 10 million years since the Central and South American bushmasters diverged . Despite the broad geographic distribution of Lachesis , antivenoms against venoms of different species are effective in the neutralization of congeneric venoms not used in the immunization mixture , indicating that they can be used equivalently for the clinical treatment of any lachesic envenoming . This study demonstrates that antivenoms raised against venom of different Lachesis species are indistinctly effective in the neutralization of congeneric venoms not used in the immunization mixture , indicating that antivenoms against conspecific venoms may be used equivalently for the clinical treatment of envenomings caused by any bushmaster species .
Snakebite envenoming is a public health issue and a neglected disease in many tropical and sub-tropical regions of Africa , Asia , Latin America and Oceania , especially affecting the most impoverished and geopolitically disadvantaged rural communities [1–4] . Between 1 . 2 and 5 . 5 million people are victims of snakebites every year , leading to 95 . 000–125 . 000 deaths and leaving more than 400 . 000 people with permanent physical and psychological sequelae [4–7] . In Central and South America most accidents are caused by pitvipers of the Viperidae family , subfamily Crotalinae , with Bothrops , Crotalus and Lachesis being the most medically relevant genera . Genus Lachesis comprises the longest pitvipers in the world , with adults ranging in length from 2 to 2 . 5 m . The four nominal species of this genus , L . stenophrys ( Central American bushmaster ) , L . melanocephala ( Black-headed bushmaster ) , L . acrochorda ( Chocoan busmaster ) and L . muta ( South American bushmaster ) inhabit remote forested areas of Central and South America , and on the island of Trinidad [8–10] . Central and South American populations of Lachesis diverged around 18 . 0–6 . 5 Mya , with a later split between L . melanocephala and L . stenophrys taking place 11–4 Mya , while differentiation of South American lineages occurred 800 . 000 to 300 . 000 years ago [8] . L . stenophrys is distributed through the Caribbean coast of Nicaragua , Costa Rica and Panama; L . melanocephala is found in the Pacific versant of southwestern Costa Rica , and the extreme western regions of Panama; L . acrochorda inhabits both the Atlantic and Pacific versants of western Panama and into northwestern Colombia , on the Atlantic coast , where it extends southward into the Cauca and Magdalena rivers valleys , and along the Pacific versant of Colombia into northwestern Ecuador . L . muta is the most widely distributed species of the genus , including the equatorial forest east of the Andes , from Colombia , eastern Ecuador , Peru , northern Bolivia , eastern and southern Venezuela to Guyana , Surinam , French Guiana and most of northern Brazil [10] . Two subspecies of L . muta are reported: L . m . muta and L . m . rhombeata , with an exclusive distribution of the latter subspecies in the Atlantic forest of eastern-center of Brazil . Based on morphology , some authors consider that there are populations of L . m . rhombeata in the Amazonia basin [11] . Human bites by Lachesis species are not frequent but when occur cause severe envenoming due to large amount of venom ( 200–411 mg ) injected into the victim and also owing to its toxicity in humans , as reported for snakebites in Brazil , Colombia and Costa Rica [11–21] . Common local effects include agonizing burning-throbbing pain , mild hemorrhage , edema , and blister formation . These signs and symptoms are accompanied by systemic alterations , such as hemorrhage , coagulopathy , cardiovascular collapse , and by the so-called “Lachesis syndrome” , an alteration of the autonomic nervous system which manifests with profuse sweating , abdominal colic , nausea , recurrent vomiting , watery diarrhea , diastolic and systolic hypotension , and sinus bradycardia , together with sensorial disorders ( uncoordinated march , lapses of unconsciousness ) and serious hemodynamic alterations within 15–20 min after a bite [12–19 , 22] . Comprehensive transcriptomic and proteomic studies across Lachesis [23–26] have revealed remarkably similar venom phenotypes comprising seven or eight toxin families , including bradykinin-potentiating/C-type natriuretic peptide ( BPPs/C-NP ) , Zn2+-dependent snake venom ( SV ) metalloproteinase ( SVMP ) , serine protease ( SVSP ) , phospholipase A2 ( PLA2 ) , L-amino acid oxidase ( LAOs ) , C-type lectin-like ( CTL ) , and in venoms of the South American species , also cysteine-rich secretory protein ( CRISP ) . Ontogenetic changes in the toxin composition of L . stenophrys venom result in the net shift from a BPPs/C-NP-rich and SVSP-rich venom in newborns and 2-years-old juveniles to a ( PI>PIII ) SVMP-rich venom in adults [24] . The high conservation of the overall composition of Central and South American bushmaster venoms and their qualitatively similar pathophysiological profile observed in clinical settings [14 , 17 , 27–29] , suggested that antivenoms generated against any conspecific Lachesis venom may exhibit paraspecific protection against the toxic activities of all other Lachesis species [24 , 25] . The aim of the present work was to assess this hypothesis . To this end , we carried out a comparative study of the cross-reactivity , neutralization of toxic activities and immunoaffinity antivenomic profiles towards a panel of Lachesis venoms of two commercial polyspecific antivenoms ( BCL ) , manufactured at Instituto Clodomiro Picado , Costa Rica , against a mixture of L . stenophrys , Bothrops asper , and Crotalus simus venoms , and antivenom produced at Instituto Vital Brazil , Brazil , against venoms from L . m . rhombeata and five bothropic species ( BL ) , and experimental monospecific antivenoms AL , AB , and AC , generated , respectively , against venoms of adult Costa Rican L . stenophrys , B . asper , and C . simus .
All the procedures involving the use of animals in this study were approved by the Institutional Committee for the Care and Use of Laboratory Animals ( CICUA ) of Universidad de Costa Rica ( approval number CICUA 028–13 ) , and meet the Animal Research Reporting in vivo Experiments ( ARRIVE ) guidelines , and the International Guiding Principles for Biomedical Research Involving Animals of the Council of International Organizations of Medical Sciences ( CIOMS ) . Lachesis venoms were obtained from different geographic areas of Central and South America . Venom from L . stenophrys ( Central American bushmaster ) was pooled from more than 25 adult ( >5 years old ) snakes maintained in the herpetarium of Instituto Clodomiro Picado ( ICP , San José , Costa Rica ) . Venom from L . melanocephala ( black-headed bushmaster ) was pooled from two adult specimens maintained at Instituto Nacional de Biodiversidad ( San José , Costa Rica ) . L . muta muta ( South American bushmaster ) venoms pooled from adult specimens from Colombia , Peru , and Cascalheria and Tucurui regions of Brazil were kindly provided by Dr . María de Fatima D . Furtado ( Instituto Butantan , São Paulo , Brazil ) . Samples of L . muta rhombeata ( Atlantic forest bushmaster ) venom pooled from adult specimens were a generous gift from Dr . María de Fatima D . Furtado of Instituto Butantan and from Instituto Vital Brazil ( IVB , Niterói , Rio de Janeiro , Brazil ) . All venoms were lyophilized and stored at -20°C until used . Commercial polyspecific BCL antivenom ( batch 4800611POLQ ) was manufactured by the Industrial Division of Instituto Clodomiro Picado ( San José , Costa Rica ) from the plasma of horses hyperimmunized with a mixture of venoms of Costa Rican Bothrops asper , Crotalus simus and Lachesis stenophrys [30 , 31] , and consists of whole IgGs purified by caprylic acid fractionation [32] . BL antivenom ( batch 125901 ) from Instituto Vital Brazil ( Niterói , RJ , Brazil ) was produced in horses hyperimmunized with a mixture of venoms from L . m . rhombeata and a mixture of five bothropic species , B . jararaca ( 50% ) , B . jararacussu ( 12 . 5% ) , B . moojeni ( 12 . 5% ) , B . alternatus ( 12 . 5% ) and B . neuwiedi ( 12 . 5% ) , and consists of purified F ( ab' ) 2 fragments generated by digestion with pepsin of ammonium sulfate-precipitated IgG molecules [33] . Experimental monospecific AB , AC , and AL antivenoms were prepared by the Industrial Division of Instituto Clodomiro Picado from plasma of horses subjected to a single round of immunization with venoms of Costa Rican adult B . asper ( from the Pacific and Caribbean versants of Costa Rica ) , adult C . simus , and adult L . stenophrys , respectively , as described [33] . These monospecific antivenoms are also whole IgG preparations prepared by caprylic acid precipitation [33] . BCL and BL antivenoms are used therapeutically in the clinical management of Lachesis envenomings in Central America and Brazil , respectively . Monospecific AC , AB and AL antivenoms were developed for experimental use . For in vitro and in vivo assays the protein concentration of antivenoms was adjusted to 50 mg/mL . 96-well plates ( Dynatech Immulon , Alexandria , VA ) were coated overnight at 4°C with Lachesis venoms ( 0 . 5 μM/well ) in 0 . 1 M Tris , 0 . 15 M NaCl , pH 9 . 0 buffer . The plates were blocked for 1h with 2% bovine serum albumin ( BSA ) in 20 mM phosphate , 135 mM NaCl , pH 7 . 4 ( PBS ) at room temperature . Purified antivenom immunoglobulins were serially diluted by a factor of 3 ( starting from a dilution of 1/500 ) in PBS containing 1% BSA , and added to the wells for 1 h at room temperature . The plates were washed four times with washing buffer ( 50 mM Tris , 150 mM NaCl , 20 μM ZnCl2 , 1 mM MgCl2 , pH 7 . 4 ) , and anti-horse IgG-phosphatase-conjugate ( Sigma , St . Louis , MO , USA ) , diluted 1:20 , 000 with PBS containing 1% BSA , was added and incubated for 1 h at room temperature . The plates were washed and developed with p-nitrophenylphosphate in diethanolamine buffer ( 1 mM MgCl2 , 90 mM diethanolamine , pH 9 . 8 ) . Absorbance at 405 nm was recorded after 90 min using a microplate reader ( Multiskan Labsystems Ltd . , Helsinki , Finland ) . L . stenophrys venom proteins were separated by two-dimensional electrophoresis ( 2DE ) using an Ettan IPGphor III instrument ( GE Healthcare Bio-Sciences AB , Uppsala , Sweden ) . For isoelectric focusing , 300–350 μg of total venom proteins in 200 μL DeStreak Rehydration Solution ( GE Healthcare Bio-Sciences AB , Uppsala , Sweden ) including 10 mM DTT and 0 . 5% IPG buffer pH 3–10 NL ( GE Healthcare Bio-Sciences AB , Uppsala , Sweden ) were loaded on a 11 cm IPG strip , pH 3–10 ( GE Healthcare Bio-Sciences AB , Uppsala , Sweden ) and then focused using the following electrophoretic conditions: 500 V for 30 min , 1000 V for 30 min and 5000 V for 80 min . After isoelectric focusing , SDS-PAGE was performed under reducing conditions in 4–15% Criterion TGX precast 11 cm gels ( Bio-Rad , USA ) . An unstained protein molecular weight marker ( Fermentas ) was included in the analysis . Gels were stained using Bio-Safe Coomassie Stain ( Bio-Rad , USA ) or PlusOne Silver Staining Kit ( GE Healthcare AB , Uppsala , Sweden ) following the manufacturer´s instructions , and images were taken with Chemidoc XRS imaging system ( BioRad , USA ) . Spot identification was done using the collaborative bioimage informatics platform Icy [34] and quantified as relative density percentage using ImageJ software [35] . 2DE gels of 350 μg L . stenophrys venom proteins were transferred to polyvinylidene fluoride ( PVDF ) membranes at 50 mA in a Criterion Blotter instrument ( Bio-Rad , USA ) overnight . To assess transfer efficiency , PVDF membranes were previsualized by reversible Ponceau-S Red staining . Unoccupied membrane protein-binding sites were blocked with 2% casein in TBS-T ( Tris-buffered saline with Tween 20 , pH 7 . 6 ) for 30 min at room temperature , and the membranes were incubated for 1 h with 1/1000 dilution of antivenoms in TBS-T containing 1% casein . After five washing steps ( 5 min each ) with TBS-T , the membranes were incubated for 1 h at room temperature with rabbit anti-horse IgG-peroxidase conjugate ( 1:15000 dilution; Sigma-Aldrich , St . Louis , USA ) . Purified antibodies from non-immunized horses were used as control . After washing off unbound secondary antibodies , the immunoreactive spots were visualized using a chemiluminescence substrate ( Invitrogen , USA ) . Images were taken with Chemidoc XRS imaging system ( BioRad , USA ) and protein spots of interest were analyzed using ImageJ software . 2DE protein spots were excised and subjected to reduction ( 10 mM dithiothreitol ) , alkylation ( 50 mM iodoacetamide ) , and overnight in-gel digestion with sequencing grade trypsin ( Sigma ) , in 50 mM ammonium bicarbonate at 37°C . Tryptic peptide digests were extracted in 50% acetonitrile containing 1% trifluoroacetic acid ( TFA ) , and analyzed by MALDI-TOF-TOF MS using an AB4800-Plus Proteomics Analyzer ( Applied Biosystems ) . To this end , tryptic digests were mixed with an equal volume of α-cyano-hydroxycinnamic acid saturated in 50% acetonitrile , 0 . 1% TFA , and 1 μL spotted onto an Opti-TOF 384-well plate , dried , and analyzed in positive reflector mode . TOF MS spectra were acquired using 500 shots at a laser intensity of 3000 . TOF/TOF fragmentation spectra were acquired ( 500 shots at a laser intensity of 3900 ) for the ten most intense precursor ions . External calibration in each run was performed with CalMix standards ( ABSciex ) spotted onto the same plate . Fragmentation spectra were searched against the UniProt/SwissProt database ( taxonomy: Serpentes ) using the ProteinPilot v . 4 and the Paragon algorithm ( ABSciex ) at ≥95% confidence , or manually interpreted and the deduced sequences BLASTed against the NCBI ( http://blast . ncbi . nlm . nih . gov ) non-redundant database for protein class assignment by similarity . The immunoreactivity of poly- and monospecific antivenoms towards the different Lachesis venoms was assessed using a second-generation antivenomics approach [36] . To prepare the antivenom affinity column , 200 μL of NHS-activated Sepharose 4 Fast Flow ( GE Healthcare Bio-Sciences AB , Uppsala , Sweden ) matrix was packed in a Pierce centrifuge column and washed with 15 matrix volumes of cold 1 mM HCl followed by two matrix volumes of 0 . 2 M NaHCO3 , 0 . 5 M NaCl , pH 8 . 3 ( coupling buffer ) to adjust the pH of the column to 7 . 0–8 . 0 . Antivenoms were dialysed against MilliQ water , lyophilised , and reconstituted in coupling buffer . The concentration of the antivenom stock solutions was determined spectrophotometrically using an extinction coefficient of 1 . 36 for a 1 mg/mL concentration of Fab at 280 nm using a 1 cm light pathlength cuvette . Twenty milligrams of polyspecific BCL antivenom , 15 mg of BL antivenom , and 35–50 mg of monospecific AC , AB and AL antivenoms were dissolved in a half matrix volume of coupling buffer and incubated with the matrix for 4 h at room temperature . Antivenom coupling yield was estimated measuring the non-bound antivenom by quantitative band densitometry of SDS-PAGE ( MetaMorph software , MDS Analytical Technologies ) using as standard for the linear range the pre-coupled antivenom . After the coupling , any remaining active groups were blocked with 200 μL of 0 . 1 M Tris–HCl , pH 8 . 0 at 4°C overnight using an orbital shaker . The affinity column was washed alternately at high and low pH , with three volumes of 0 . 1 M acetate buffer , 0 . 5 M NaCl , pH 4 . 0–5 . 0 and three volumes of 0 . 1 M Tris–HCl buffer , pH 8 . 5 . This treatment was repeated six times and the column was equilibrated in binding buffer ( 20 mM phosphate , 135 mM NaCl , pH 7 . 4 , PBS ) . For the immunoaffinity assay , 200 μg of venoms from L . stenophrys ( Costa Rica ) , L . melanocephala ( Costa Rica ) , L . m . rhombeata ( Recife , Brazil ) , and L . m . muta from Colombia , Peru and Brazil ( Tucurui and Cascalheira regions ) , dissolved in 1⁄2 matrix volume of PBS , were loaded and incubated for 1 h at room temperature with the affinity matrix , followed by incubation in an orbital shaker overnight at 4°C . As specificity controls , 200 μL of Sepharose 4 Fast Flow matrix , without or with 8 . 5 mg of immobilized pre-immune IgGs , were incubated with venom and developed in parallel to the immunoaffinity columns . Non-retained fractions were collected with 5 matrix volumes of PBS , and the immunocaptured proteins were eluted with 5 matrix volumes of elution buffer ( 0 . 1 M glycine-HCl , pH 2 . 0 ) and neutralised with 150 μL 1 M Tris-HCl , pH 9 . 0 . The non-retained and the immunocaptured venom fractions were lyophilized , reconstituted in 40μl of MilliQ water , and fractionated by reverse-phase HPLC using a Discovery BIO Wide Pore C18 ( 15 cm x 2 . 1 mm , 3 μm particle size , 300 Å pore size ) column using an Agilent LC 1100 High Pressure Gradient System equipped with a DAD detector and micro-auto sampler . The column was developed at a flow rate of 0 . 4 mL/min and proteins eluted with a linear gradient of 0 . 1% TFA in MilliQ water ( solution A ) and 0 . 1% TFA in acetonitrile ( solution B ) : isocratic at 5% solution B for 1 min , followed by 5–25% solution B for 5 min , 25–45% solution B for 35 min , and 45–70% solution B for 5 min . Protein was detected at 215 nm with a reference wavelength of 400 nm . Polyspecific and monospecific antivenoms were assessed for their ability to neutralize the lethal , hemorrhagic , coagulant and proteolytic activities of venoms . The protein concentration of all antivenoms was adjusted to 50 mg/mL , as determined using a NanoDrop 2000 ( Thermo Fischer Scientific , DE , USA ) . For the neutralization assays , a fixed dose of venom ( “challenge dose” ) , dissolved in PBS , was incubated with various dilutions of antivenom . Controls including venom solutions incubated with PBS instead of antivenom were used . The venom/antivenom mixtures and controls were incubated for 30 min at 37°C and then tested in the experimental systems described below and detailed in previous publications [37 , 38] . Neutralizing ability was expressed as Median Effective Dose ( ED50 ) , defined as the μL antivenom/mg venom ratio in which the activity of venom was reduced by 50% [39] . In the case of coagulant activity , neutralization was expressed as Effective Dose ( ED ) , defined as the antivenom/venom ratio in which the clotting time of plasma was prolonged three times when compared with clotting time of plasma incubated with venom alone [40] . All the in vivo experiments were performed in CD-1 mice , and the protocols were approved by the Institutional Committee for the Care and Use of Laboratory Animals ( CICUA ) of the University of Costa Rica . Lethality was assessed by the intraperitoneal route in 16–18 g mice and a challenge dose corresponded to 3 Median Lethal Doses ( LD50 ) was used for the neutralization tests [37] . An arbitrary level of 500 μL antivenom/mg venom was selected to evaluate the efficacy of antivenoms for neutralizing lethality . Only this antivenom/venom ratio was used owing to the scarcity of some venoms and also for reducing the number of mice used . Death of mice was recorded at 48 h . Hemorrhagic activity was evaluated by using the rodent skin test using 18–20 g mice and a challenge venom dose corresponding to 10 Minimum Hemorrhagic Doses ( MHD ) [41] . Coagulant activity was assessed on citrated human plasma and the challenge dose used was 2 Minimum Coagulant Doses ( MCD ) [40] . Proteolytic activity was determined using azocasein ( Sigma , USA ) as substrate , as described by Gutiérrez et al . [42] . For neutralization tests , a challenge dose was selected , corresponding to the amount of venom that induced a change in absorbance of 0 . 75 at 450 nm . A summary of reference venom activities ( Median Lethal Dose , Minimum Hemorrhagic Dose , Minimum Coagulant Dose and challenge dose for proteolytic activity ) of Lachesis venoms are listed in Table 1 . The results of neutralization assays of venom activities were compared by ANOVA , followed by Tukey test for specific comparisons between means of pairs of groups . A p value <0 . 05 was regarded as statistically significant . For data not following the assumptions of parametric tests , a Kruskal-Wallis test was used , followed by Dunn test . The analysis was performed using the Minitab ( v 16 . 1 . 0 , 2010 ) statistic program .
Initial assessment of the immunoreactivity of the commercial polyspecific BCL and BL antivenoms , and the experimental monospecific B , C and L antivenoms , against antigens present in the venoms of Costa Rican L . stenophrys and L . melanocephala , Brazilian L . m . rhombeata ( Recife ) and L . m . muta from different geographic locations ( Colombia , Peru , and Brazil [Cascalheira and Tucurui] ) were done by ELISA and 2DE immunoblotting analysis . No significant differences were found in the levels of specific antibodies against Lachesis venoms present in the BCL antivenom , and the AB and AL antivenoms ( S1 Fig ) . The highest titer corresponded to the binding of BL antivenom to L . stenophrys , L . m . muta ( Colombia ) , L . m . muta ( Cascalheira ) , and L . m . rhombeata ( Recife ) venoms , whereas the titer of this antivenom against venoms from L . melanocephala , L . m . muta ( Peru ) and L . m . muta ( Tucurui ) was indistinguishable from that of the BCL antivenom ( S1 Fig ) . Monospecific AC antivenom exhibited the lowest reactivity against the seven Lachesis venoms analyzed ( S1 Fig ) . The spectrum of L . stenophrys toxins immunorecognized by the poly- and monospecific antivenoms was investigated by 2DE and immunoblot analysis . Fig 1A displays a 2DE reference map and the MALDI-TOF-TOF MS protein assignments are listed in S1 Table . In concordance with ELISA results , Western blot analyses revealed extensive protein spot recognition by all the five antivenoms ( Fig 1B and S2 Table ) , particularly for spots in the range of 25–35 kDa ( serine proteinases , SVSPs ) and 14–16 kDa ( phospholipases A2 ( PLA2 ) and Galactose-binding lectin ) . Polyspecific BL and monospecific AB antivenoms showed also strong immunoreactivity towards protein spots of apparent molecular mass 55–80 kDa , which were identified as snake venom metalloproteinases of class PIII ( PIII-SVMP ) and L-amino acid oxidase ( LAO ) molecules . However , all the antivenoms showed weak immunostaining of spot 64 containing the major SVMP of class PI ( PI-SVMP ) . Weak immunorecognition of PI-SVMPs has been also reported for other antibothropic antivenoms [52 , 53] . Second generation antivenomics [36] was applied to complement the ELISA and Western blot analyses of the immunoreactivity of the lachesic antivenoms towards the panel of Lachesis venoms used for this study . Figs 2–8 display the immunoaffinity chromatography-based antivenomic profiles of commercial BCL and BL polyspecific antivenoms and monospecific B , L , and C experimental antivenoms towards the venoms of L . stenophrys and L . melanocephala from Costa Rica , L . muta muta from Colombia , Peru , and the Brazil regions of Cascalheria , Tucurui , and L . muta rhombeata from Recife , Brazil . The results show impaired immunocapturing ability of the early eluting chromatographic fractions comprising bradykinin-potentiating-like peptides ( BPP-like ) by all the antivenoms . Although together these fractions account for about 1/3 by weight of total venom components , previous investigations have shown that the intraperitoneal administration of an amount of BPP-like peptides contained in 10–24 LD50s of venom induced neither a significant change in the mean arterial blood pressure of mice , nor signs of abnormal behavior , or histopathological alterations in heart and lungs [25] . These observations strongly suggest that , despite being a major venom component , the BPP-like peptides by themselves may not represent a serious clinical concern in the treatment of Lachesis envenomings . The interpretation of these results has to consider that certain antigens may become denatured during reverse-phase separation and , henceforth , some conformational epitopes might be lost . Except for the BPPs , both polyspecific antivenoms efficiently immunocaptured all the components from L . stenophrys ( Fig 2 ) , L . m . muta ( Colombia ) ( Fig 4 ) and L . m . rhombeata ( Fig 8 ) venoms . In addition , the BL antivenom immunocaptured the venom components of L . m . muta from the Brazilian localities Cascalheira ( Fig 6 ) and Tucurui ( Fig 7 ) . The apparent low recovery of PI- and PIII-SVMPs ( eluting from the RP-HPLC column at 40–42 min ) in the immunoaffinity captured fractions of the BCL and BL affinity columns ( Figs 2–8 , panels B and D , respectively ) may be ascribed to the high affinity of these venom proteins for the antivenom molecules , as has been demonstrated in a previous work [25] . The worst immunocapturing profile of BCL and BL antivenoms was obtained using L . m . muta from Peru ( Fig 5 ) , where Gal-lectin [Q9PSM4] eluting in peak 9 ( Fig 5 ) was essentially ( >85% ) found in the non-binding fraction . The BCL antivenom also showed limited binding capability towards Gal-lectin [Q9PSM4] and serine proteinase [P33589] from L . m . muta from Cascalheira ( peaks 9 and 11 , respectively , Fig 6A; 65% of each proteins found in the not retained fraction ) and Tucurui ( peaks 8 and 10 , respectively , Fig 7A , 53% not immunocaptured ) , and the PLA2 molecule eluting in peak 3 of L . m . muta from Tucurui ( Fig 7A ) . 27% of this protein was not immunocaptured by the BCL antivenom . Monospecific antivenoms showed significantly more limited immunorecognition profiles than BCL and BL antivenoms toward venoms of all Lachesis taxa investigated . The three monospecific antivenoms , but particularly the anti-crotalic ( AC ) antivenom , exhibited poor binding ability towards most venom proteins , including PLA2s , CRISP , Gal-lectin , SVSPs , PI- and PIII-SVMPs and LAO . The average toxin immunocapturing activity of this monospecific antivenom was 16% ( L . stenophrys ) , 21% ( L . melanocephala , ) , 21% ( L . m . muta Colombia ) , 9% ( L . m . muta Peru ) , 9% ( L . m . muta Cascalheira ) , 17% ( L . m . muta Tucurui ) , and 19% ( L . m . rhombeata ) ( panels I of Figs 2–8 , respectively ) . Although a comparison of the levels of immune recognition gathered from antivenomics with the in vivo neutralization capacity of an antivenom is not straightforward , since both experiments involve radically different protocols , in our experience , an immunocapturing capability of ≥25% of total viperid venom proteins correlates with a good outcome in in vivo neutralization tests [48–51] . As a whole , the antivenomics evidence reinforce our view that both polyspecific BCL and BL antivenoms are likely to be effective in the neutralization of heterologous congeneric venoms , thus supporting their use for the treatment of Lachesis envenomings throughout the range of distribution of these snakes . In addition , the fact that antivenoms BCL , and particularly BL , are more effective than the monospecific AL antivenom , even against the homologous L . stenophrys venom , seems to indicate that the inclusion of botropic venoms in the immunization mixture aided in the generation of antibodies exhibiting paraspecificity against Lachesis toxins . This combination of immunogens seems to be a more appropriate formulation than a single venom for the treatment of envenomings by Lachesis species . Standard neutralization assays were performed to assess the extent of neutralization of proteolytic , hemorrhagic , procoagulant and lethal activities [39–41] . Despite the fact that all antivenoms were standardized for having a protein concentration of 50 mg/mL , two aspects need to be considered when comparing the values of neutralization experiments: ( a ) Most antivenoms are made of whole IgG molecules , whereas one of them is made of F ( ab’ ) 2 fragments; hence , for the same amount of protein , the number of molecules present in an IgG antivenom is 1 . 5 times lower than in a F ( ab` ) 2 antivenom; and ( b ) only an unknown proportion of all IgGs or F ( ab’ ) 2 fragments are specific against venom components . Hence , quantitative conclusions drawn by comparing the neutralizing abilities of different type of antivenoms should be regarded as gross estimates . The BCL and the BL therapeutic antivenoms , and the monospecific AL antivenom effectively neutralized the proteolytic activity of venoms from the 7 Lachesis taxa investigated ( Table 2 ) . The BL antivenom showed higher neutralization activity than the other antivenoms used in this study ( Table 2 ) . The AC monospecific antivenom was only able to neutralize the proteolytic activity of L . melanocephala venom ( Table 2 ) . The AB monospecific antivenom was unable to neutralize the proteolytic activity of any of the venoms ( Table 2 ) . The BCL and BL antivenoms , and the monospecific L antivenom effectively neutralized the hemorrhagic activity of all the Lachesis venoms studied ( Table 3 ) . The BL antivenom showed the highest neutralization capacity of the hemorrhagic activity than any of the other antivenoms used in this study ( Table 3 ) . The monospecific AB antivenom was only able to neutralize the hemorrhagic activity of L . stenophrys and L . melanocephala venoms , whereas the monospecific AC antivenom was unable to neutralize the hemorrhagic activity of any of the venoms ( Table 3 ) . The BCL and BL polyspecific antivenoms , and the monospecific L antivenom effectively neutralized the coagulant activity of Lachesis venoms from the seven bushmaster taxa sampled ( Table 4 ) . The BL antivenom showed the highest coagulant neutralization activity than any of the other antivenoms use in this study ( Table 4 ) . On the other hand , neither the AB nor the AC monospecific antivenoms were able to neutralize the coagulant activity of any of the Lachesis venoms used in this study ( Table 4 ) . These data agree with a previous work showing the inefficacy of monospecific bothropic antivenom in the neutralization of the coagulation activity of L . m . muta venom [54] . At the antivenom/venom ratio of 500 μL antivenom/mg venom , the BCL and BL polyspecific antivenoms and the monospecific L antivenom , effectively neutralized the lethal activity of the seven Lachesis venoms investigated ( Table 5 ) . The monospecific AB antivenom only neutralized the lethal activity of L . stenophrys , L . muta muta ( Cascalheira ) and L . muta rhombeata ( Recife ) ( Table 5 ) , while the monospecific AC antivenom was unable to neutralize the lethal activity of any of the Lachesis venoms studied at the ratio of 500 μL antivenom/mg venom ( Table 5 ) . Snakes from the Lachesis genus cause severe envenomings in humans and are widely distributed in a variety of habitats ranging from the Caribbean coast of Central America to the Atlantic rainforest of Brazil . Based on the high conservation of the overall protein composition of Lachesis venoms and their qualitatively similar pathophysiological profile observed in experimental envenomings and clinical settings we have suggested that antivenoms generated against any conspecific Lachesis venom may exhibit paraspecific protection against the toxic activities of all other Lachesis species . Combining immunochemical methods , second generation antivenomics , and venom neutralization tests we have unveiled the efficacy of two therapeutic polyvalent antivenoms and three experimental monospecific antivenoms to recognize the complete proteomes and neutralize the hemorrhagic , coagulant , proteolytic and lethal activities from three different Lachesis species from different geographic populations . The results demonstrate that antivenoms raised by immunizing horses with the venoms of different Lachesis species are effective in the neutralization of congeneric venoms not used in the immunization mixture , indicating that they could be used equivalently for the clinical treatment of any lachesic envenoming . Owing to the similar clinical presentations of envenomings by Lachesis sp . and Bothrops sp . , the use of polyvalent antivenoms which include the Lachesis component is therefore recommended in Latin America .
|
Snakebite envenoming is a neglected public health problem in many developing countries and antivenom administration constitutes the mainstay in the treatment of such envenomings . Therapeutic antivenoms contain animal-derived antibodies against venom toxins and are produced by immunizing animals with the venom from one or several snake species from a defined geographical area . Defining the geographic boundaries of the efficiency of an antivenom therefore has implications for its rational and efficient use . In Central and South America most accidents are caused by pitvipers of the genus Bothrops , Crotalus and Lachesis . There are four Lachesis species distributed in a variety of habitats ranging from the Caribbean coast of Central America to the Atlantic rainforest of Brazil . Lachesis species cause severe envenomings in humans due to the toxicity of their venoms and also to the large amount of venom they inject into their victims . In this work we investigate the capability of several antivenoms to neutralize the toxic activities of a panel of Lachesis venoms . The results demonstrate that antivenoms raised by immunizing horses with the venoms of different Lachesis species are effective at neutralizing congeneric venoms not used in the immunization , indicating that they could be used equivalently for the clinical treatment of any lachesic envenoming .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"costa",
"rica",
"toxins",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"enzyme-linked",
"immunoassays",
"immunology",
"geographical",
"locations",
"tropical",
"diseases",
"vertebrates",
"animals",
"toxicology",
"toxic",
"agents",
"north",
"america",
"reptiles",
"neglected",
"tropical",
"diseases",
"antibodies",
"immunologic",
"techniques",
"snakebite",
"research",
"and",
"analysis",
"methods",
"central",
"america",
"venoms",
"immune",
"system",
"proteins",
"south",
"america",
"proteins",
"immunoassays",
"brazil",
"snakes",
"people",
"and",
"places",
"biochemistry",
"colombia",
"squamates",
"physiology",
"biology",
"and",
"life",
"sciences",
"amniotes",
"organisms"
] |
2017
|
Cross-reactivity, antivenomics, and neutralization of toxic activities of Lachesis venoms by polyspecific and monospecific antivenoms
|
The complexity of the eukaryotic parasite Trypanosoma ( T . ) cruzi manifests in its highly dynamic genome , multi-host life cycle , progressive morphologies and immune-evasion mechanisms . Accurate determination of infection or Chagas’ disease activity and prognosis continues to challenge researchers . We hypothesized that a diagnostic platform with higher ligand complexity than previously employed may hold value . We applied the ImmunoSignature Technology ( IST ) for the detection of T . cruzi-specific antibodies among healthy blood donors . IST is based on capturing the information in an individual’s antibody repertoire by exposing their peripheral blood to a library of >100 , 000 position-addressable , chemically-diverse peptides . Initially , samples from two Chagas cohorts declared positive or negative by bank testing were studied . With the first cohort , library-peptides displaying differential binding signals between T . cruzi sero-states were used to train an algorithm . A classifier was fixed and tested against the training-independent second cohort to determine assay performance . Next , samples from a mixed cohort of donors declared positive for Chagas , hepatitis B , hepatitis C or West Nile virus were assayed on the same library . Signals were used to train a single algorithm that distinguished all four disease states . As a binary test , the accuracy of predicting T . cruzi seropositivity by IST was similar , perhaps modestly reduced , relative to conventional ELISAs . However , the results indicate that information beyond determination of seropositivity may have been captured . These include the identification of cohort subclasses , the simultaneous detection and discerning of other diseases , and the discovery of putative new antigens . The central outcome of this study established IST as a reliable approach for specific determination of T . cruzi seropositivity versus disease-free individuals or those with other diseases . Its potential contribution for monitoring and controlling Chagas lies in IST’s delivery of higher resolution immune-state readouts than obtained with currently-used technologies . Despite the complexity of the ligand presentation and large quantitative readouts , performing an IST test is simple , scalable and reproducible .
Chagas’ disease ranks as the most important parasitic infection in the Western hemisphere , as measured by disability-adjusted life years lost , surpassing malaria more than 7-fold [1–5] . It is found predominantly in Latin America , but migration of infected individuals has increased its geographic distribution into Europe , North America , Japan , and Australia [5–8] . It is globally the leading cause of infectious myocarditis [8–10] . Inconsistent with this significant health burden , resources for its control are considered highly limited [11 , 12] . The etiologic agent , Trypanosoma ( T . ) cruzi , is a flagellated protozoan transmitted predominantly via blood-feeding triatomine insects into mammalian hosts , where it can penetrate and multiply in a wide range of nucleated cell types . Other modes of dissemination include blood transfusion or congenital and oral routes [13] . Here we explore the potential application of the ImmunoSignature Technology ( IST ) to the diagnostic challenges of Chagas’ disease . A week or so after infection by the trypomastigote stage protozoan , hosts experience an acute phase characterized by microscopically visible blood-parasites and tissue parasitism . Symptoms are usually mild and non-specific such that this phase is often undiagnosed; however , rare cases manifest with disease-specific periorbital swelling or ulcerative lesions at the entry site . In less than 1% of cases , the acute phase is severe and life-threatening [5] . Survivors transition into a chronically-infected phase in which the host and parasite are immunologically balanced and symptoms resolve . Most patients will remain in this clinically indeterminate stage for life , typically accompanied by the loss of detectable parasitemia , though low levels of intracellular parasites may remain measurable in certain tissues [13 , 14] . From 10 to 30 years later , a third or more of these individuals will progress into a symptomatic stage of chronic disease . They suffer severe manifestations of cardiac , gastric , or other organ-related disease that lead to irreversible muscular lesions and often death [15–17] . Based on current estimates of the true prevalence of T . cruzi infection , the World Health Organization ( WHO ) has recently estimated that approximately 200 , 000 people will die from Chagasic cardiomyopathy in the next five years . A similar number of women are forecast to die in the US from breast cancer in the same timeframe [18] . The only two drugs available for Chagas treatment [7 , 19] have shown limited efficacy [20 , 21] and notable side-effects [19 , 22] . Discovery of new drugs against T . cruzi infections that are safer and more effective [23] has been hampered by the lack of reliable , practical methods to assess efficacy in either subclinical or clinical stages of the chronic phase . There are many challenges to measuring infection status and therapeutic impact in this phase [24] . For example: i ) parasitemia is subpatent and low levels of organ-concealed parasites are anatomically scattered , ii ) other endemic parasites such as Leishmania and Plasmodium spp . share similar antigens with T . cruzi and iii ) there are no reliable biomarkers of incipient or active disease [25] . In summary , there is need for a test that stratifies T . cruzi seropositive individuals into clinically distinct groups . These strata include distinguishing individuals who remain infected from those that manage to resolve a chronic phase infection . It would also be desirable to predict which indeterminate-stage individuals will remain asymptomatic from those that will progress to suffering life-threatening complications . A recent study identified four biomarkers that added predictive value of short-term mortality in patients suffering severe cardiomyopathy [26] . Another study identified a set of inflammatory cytokines and cardiac biomarkers that were elevated in patients with severe disease [25] . A transcriptome analysis identified a set of genes differentially expressed between patients with moderate versus severe Chagas cardiomyopathy [27] . Together , these results suggest that molecular stratification of clinically distinct patient groups is possible . The WHO recommends diagnosis of disease to be based on epidemiological risk and laboratory testing [28] . In the acute phase , direct detection of parasite is possible; however , this is not reliably sensitive in the chronic phase . For these individuals , diagnosis is based on indirect positive detection by two serologic tests [29] . In 2007 , US blood centers began serologically screening blood and organ donors . The FDA has approved the Ortho T . cruzi ELISA [30] and Abbott PRISM [31] tests . These report a signal to cut off value ( S/CO ) that quantifies levels of lysate-antigen binding in blood plasma and reflects antibody titers . A confirmatory radio-immunoprecipitation assay ( Quest Diagnostic T . cruzi RIPA ) [32] is employed routinely by many blood banks [30] . Newer generation assays are under development based on various mixtures of recombinant proteins . In a study to screen for new antigen candidates , a tiled peptide array spanning 457 T . cruzi proteins was probed against pooled immunoglobulin G ( IgG ) preparations purified from T . cruzi-antibody positive sera . Binding analysis identified 97 novel candidate-antigens [33] . Since this eukaryotic pathogen has a complex proteome and life cycle [34] , many targets may be needed to fully capture human immune responses to it [8] . A pre-requisite for developing a test with new Chagas’ disease-related diagnostic capacities would seem to be access to a reliable and scalable platform for measuring the diversity of host responses to T . cruzi infections . The ImmunoSignature Technology has shown applicability to the classification of many immune-mediated diseases [35–40] . It is based on diverse yet reproducible patterns of peripheral antibody binding to an array of >100 , 000 peptides that were selected to provide an unbiased sampling of all possible amino acid ( a . a . ) combinations rather than to correspond to any biological sequences . The assay is performed with a small sample of blood , plasma , or serum [41] . A peptide bound by an antibody is presumably not the original antibody recognition sequence but rather mimics the sequence or structure of the epitope . Since the diversity of chemical sequences is several orders of magnitude greater than proteomic sequences , a broad range of mimicry is afforded . For example , a chemical-library peptide may be mimicking a linear sequence , a structure , a mutated sequence such as found in tumors , or a non-peptidic biomolecule such as carbohydrate . Even if the cognate epitope is peptidic and linear , the probability of any mimetic-peptide ( mimotope ) exactly matching the epitope is low because the arrayed library of peptides only samples chemical sequence space . Each IST peptide sequence that is selectively bound by an antibody is a functional surrogate of an epitope that the antibody recognizes in vivo . When the mimotope signal is unique to a health state , the bound antibody becomes a biomarker . Collectively , all specifically-bound antibodies correlated with a given health state are informative for both detecting and monitoring it . A key advantage of this platform is that it is not designed to accommodate any single disease , such that the same library is universally appropriate . In principle , it should be informative for any condition that elicits a specific immune response once a unique , multivariate algorithm is generated . Application-specific accommodations can be reserved as optimization steps , if desired . In demonstration of this universality , we conducted two studies with the same combinatorial peptide library . In the first , a binary disease positive versus negative contrast was evaluated . In the second , the ability to simultaneously distinguish between multiple infectious diseases with one algorithm was explored . For this work , we chose to evaluate the ability of IST to classify individuals who were seropositive but asymptomatic for T . cruzi , hepatitis B virus ( HBV ) , hepatitis C virus ( HCV ) and West Nile virus ( WNV ) or seronegative for all four . Blood specimens from these donors were available that had been retrospectively collected using the same protocols , and declared positive or negative by the same testing regimens . The main goal of this work was to establish the feasibility of using IST to distinguish serologic T . cruzi positivity in asymptomatic blood donors that either were T . cruzi seronegative or were seropositive for a different disease . The studies presented here demonstrate IST as a viable diagnostic approach for T . cruzi and lay the groundwork for its use in new areas of clinically relevant discovery .
Plasma samples were collected by United Blood Services ( http://www . unitedbloodservices . org ) , and obtained from Creative Testing Solutions ( CTS , Tempe , AZ ) in 2015 as samples that were blood panel tested as being serologically-positive for T . cruzi-specific antibodies and corresponding age and gender matched samples that were determined to be T . cruzi antibody-negative . A second cohort of T . cruzi seropositive and seronegative samples were obtained from CTS in 2016 . These specimens tested negative for all other blood panel diseases . Additional plasma samples that serologically tested positive for HBV , HCV and WNV were obtained from CTS in 2016 . These were all collected by United Blood Services in the US , from the blood donor population; ethnicity is provided for those for which it was reported to CTS . Upon receipt the specimens were thawed , a portion of each was removed and mixed 1:1 with ethylene glycol as a cryoprotectant , then aliquoted into single-use volumes . Aliquots were stored at -20°C until needed . The remaining undiluted sample volume was stored neat at -80°C . Identities of all samples were tracked using 2D barcoded tubes ( Micronic , Leystad , the Netherlands ) . In preparation for the assay , sample aliquots were warmed on ice to 4°C and diluted 1:100 in primary incubation buffer ( Phosphate Buffered Saline with 0 . 05% Tween 20 ( PBST ) and 1% mannitol ) . Microtiter plates containing the 1:100 dilutions were then diluted to 1:625 for use in the assay . For the subset of samples selected for evaluating platform performance across wafer lots , the 1:100 dilutions were aliquoted into single-use microtiter plates and stored at -80°C . All aliquoting and dilution steps were performed using a BRAVO robotic pipetting station ( Agilent , Santa Clara , CA ) . All procedures , which used de-identified , banked plasma samples , were reviewed by the Western Institutional Review Board ( protocol no . 20152816 ) . A combinatorial library of 125 , 509 peptides with a median length of 9 residues and range from 5 to 13 a . a . ’s was designed to include 99 . 9% of all possible 4-mers and 48 . 3% of all possible 5-mers of 16 a . a . ’s . Methionine and cysteine were excluded because of their oxidation and cyclization potential , which would cause probe variability . Isoleucine and threonine were excluded because valine and serine , respectively , are generally considered chemically and structurally similar . Valine and serine were selected for inclusion because they are structurally smaller than their excluded counterpart . Limiting the number of amino acids reduced fabrication complexity and cost while maintaining a survey of a similar diversity of chemical and structural space . Several categories of control peptides added 6 , 203 features . For example , 500 features corresponded to the established epitopes of five different well-characterized murine monoclonal antibodies ( mAb ) , each replicated 100 times . Another 935 features corresponded to four different sequence variants of three of the five established epitopes , each replicated from 100 to 280 times . Another 500 control peptides were designed with a . a . compositions like those of the library peptides , but were uniformly 8-mers in length and present in triplicate . The median signals of these 1500 8-mers were quantitated and treated as part the library when developing the IST models; the other control peptides were not . The binding of purified murine antibodies , antibody-spiked human plasma , and cohorts of normal donor plasma samples to all of these control features were analyzed in platform validation studies . They were also monitored in all library-directed assays conducted with IST study samples to ensure data quality . The remaining 3 , 268 controls include fiducial markers to aid grid alignment , analytic control sequences and linker-only features . Except for the fiducials , all features are distributed evenly across the array . The distribution of replicate features enabled spatial variability to be analyzed by measuring the coefficient of variance of peptides with many or few replicates; it also enabled the detection of any signal gradients or other aberrancies . Finally , the library peptides themselves served as a category of control features . Namely , on and off target specificity assessments were made by analyzing the binding of mAbs with known epitope recognition to the diverse panel of library peptides . The peptides were synthesized on a 200 mm silicon oxide wafer using standard semiconductor photolithography tools adapted for tert-butyloxycarbonyl ( BOC ) protecting group peptide chemistry , using methods described previously [40] . Briefly , an aminosilane functionalized wafer was coated with BOC-glycine . Next , photoresist containing a photoacid generator , which is activated by UV light , was applied to the wafer by spin coating . Exposure of the wafer to UV light ( 365nm ) through a photomask allows for the fixed selection of features on the wafer to be exposed with any given mask . After exposure to UV light , the wafer was heated , allowing for BOC-deprotection of the exposed features . Subsequent washing , followed by the application of an activated a . a . would complete a cycle . Each cycle added a specific a . a to the N-terminus of peptides located at masked-defined locations within the array . These cycles were repeated , varying the mask and a . a . ’s that were coupled at each feature , to obtain the combinatorial library of chemical sequence peptides . The arrays used in the Chagas study were synthesized on a hydrophilic wafer surface , ( 3-glycidoxypropyl ) trimethoxysilane-2 , 2’- ( ethylenedioxy ) bis ( ethylamine ) , with a polyethylene glycol linker . The arrays in the multi-disease study were synthesized on a less hydrophillic wafer surface , ( 3-glycidoxypropyl ) trimethoxysilane -poly ( allylamine ) , with a SGSG linker . Each completed wafer was diced into 13 rectangular regions each having the dimensions of standard microscope slides ( 25mm x 75mm ) . Each of the slides contained 24 arrays arranged in eight rows by three columns . Finally , the protecting groups on several a . a . side chains were removed using a standard cocktail [42] . The fully prepared slides were stored in a dry nitrogen environment until used in assays . Several quality tests were performed to ensure arrays were manufactured within process specifications including the use of 3σ statistical limits for each step . Wafer batches were sampled intermittently by MALDI-MS to verify that each a . a . was being coupled at the intended step with efficiencies of >97% ( 95%-100% ) , thus ensuring that the final combinatorial synthesis products were correct . Wafer manufacturing was tracked from beginning to end via an electronic custom Relational Database . Data typically tracked include chemicals , recipes , time and technician performing tasks . After a wafer was produced the data was reviewed and the records were locked and stored . Finally , each lot was evaluated in a binding assay to confirm performance , as described below . Our current automated system , using commercially available wafer-synthesis tools , is capable of manufacturing 1 , 248 arrays in three days . Production could be scaled by adding more shifts or purchasing additional automated systems . Prior to conducting the IST assays with donor plasma , the binding activity of commercial , murine monoclonal antibodies ( mAb ) was evaluated against control peptides corresponding to the established recognition sequence of each mAb . The IST arrays were probed separately , in triplicate , with 2 . 0 nM of antibody clones 4C1 ( GeneTex , Inc . , Irvine , CA ) , p53Ab1 ( EMD Millipore , Billerica , MA ) , p53Ab8 ( EMD Millipore ) and LnkB2 ( Absolute Antibody , Ltd . , Cleveland , United Kingdom ) in primary incubation buffer ( 1% mannitol , PBST ) , as detailed below . Binding was detected with 4 . 0 nM goat anti-mouse IgG conjugated to DyLight 549 ( KPL , Inc . , Gaithersburg , MD ) and signal was quantified , as detailed below . Production quality manufactured microarrays were rehydrated prior to use by soaking with gentle agitation in distilled water for 1 h , PBS for 30 min and primary incubation buffer ( 1% mannitol , PBST ) for 1 h . The microarray slides were next briefly rinsed in distilled water to prevent a surface salt residue and centrifuged to remove excess liquid . The slides were loaded into an ArrayIt microarray cassette ( ArrayIt Corporation , Sunnyvale , CA ) to adapt the individual slides into a microtiter plate footprint . Using a liquid handler , 90μl of each sample was prepared as a 1:625 dilution in primary incubation buffer ( 1% mannitol , PBST ) and then transferred to the cassette . This mixture was incubated on the arrays for 1 h at 37°C with mixing on a TeleShake95 ( INHECO , Martinsried , Germany ) to drive antibody-peptide binding . Following incubation , the cassette was washed three times in PBST using a BioTek 405TS Select microtiter plate washer ( BioTek Instruments , Inc . , Winooski , VT ) . Bound antibody was detected using either 4 . 0 nM goat anti-human IgG ( H+L ) conjugated to AlexaFluor 555 ( Invitrogen-Thermo Fisher Scientific , Inc . , Carlsbad , CA ) , or 4 . 0 nM goat anti-human IgA conjugated to DyLight 550 ( Novus Biologicals , Littleton , CO ) in secondary incubation buffer ( 0 . 5% casein in PBST ) for 1 h with mixing on a TeleShake95 platform mixer , at 37°C . Following incubation with the secondary antibody , the slides were again washed with PBST , followed by distilled water; once removed from the cassette , the slides were sprayed with isopropanol and centrifuged dry . Quantitative signal measurements were obtained by determining a relative fluorescent value for each addressable peptide feature . These assay conditions were identified in several design of experiment ( DOE ) matrices in which plasma dilution , secondary concentration , blocking reagent , incubation times , and shaking fluidics were permutated . Separately , ELISAs were conducted to assess cross-reactivity between the anti-IgG and anti-IgA secondary antibody products used in this work . A low level of cross-reactivity was noted for the anti-IgG product against the IgA secondary; no reactivity was detectable by the anti-IgA product against the IgG secondary reagent . Assayed microarrays were imaged using an Innopsys 910AL microarray scanner fitted with a 532nm laser and 572nm BP 34 filter ( Innopsys , Carbonne , France ) . The Mapix software application ( version 7 . 2 . 1; Innopsys ) identified regions of the images associated with each peptide feature using an automated gridding algorithm . Fluorescent intensities were acquired from 14 laser line-scans at 1μm resolution for each feature of the array . Median pixel intensities for each peptide-feature were calculated and saved as a tab-delimitated text file , and stored in a database for analysis . The median feature intensities were log10 transformed after adding a constant value of 100 to improve homoscedasticity . The log-transform was applied because the data were log-normally distributed; the precision of the measurements was approximately proportional to the intensities . The intensities on each array were normalized by subtracting the median intensity of the combinatorial library features for that array . For analysis of the monoclonal assays , selective binding of each mAb to its cognate epitope was assessed using a Z-score , calculated as: Z=mean ( ImAb ) −mean ( I2° ) sd ( I2° ) where ImAb and I2o are the transformed peptide intensities in the presence of mAb and secondary or secondary reagent only , respectively . Binding intensities of all four mAbs to all features corresponding to the four epitope sequences were measured . Calculated scores were based on scanner measurements with a range from 0 to 65 , 535 RFI units . For analysis of the IST assays , binding of plasma-antibodies to each library feature , as detected by fluorescently labeled secondary antibodies , was measured for all donor samples . A complete list of all library peptides and the raw fluorescent intensities measured at each library peptide-feature for each of the donor samples used in classifier development are reported in the Dryad Digital Repository ( http://dx . doi . org/10 . 5061/dryad . p6882 ) . From these data , peptides were identified as displaying differential signal levels between different health groups by using a t-test of mean peptide intensities of a peptide within a group to that of the contrasting group . The Welch adjustment for unequal variances was applied . For the 2015 Chagas cohort , T . cruzi seropositive donors ( n = 146 ) were compared to seronegative donors ( n = 189 ) , and peptides with significantly differential signal were identified . In a separate experiment , a set of peptides was identified that co-discriminated Chagas and three viral-positive donor samples . The mean intensities were compared of Chagas positive ( n = 88 ) , HBV positive ( n = 88 ) , HCV positive ( n = 71 ) and WNV positive ( n = 88 ) donor samples , as determined by blood panel testing protocols . For both the Chagas and the multi-disease experiments , peptides that showed significant discrimination were identified based on 5% threshold for false positives after applying the Bonferroni correction for multiplicity ( i . e . , p <4e-7 ) . Even though peptide selection is done at the peptide level , our classification analysis does not depend on the performance of individual peptides , but rather on the power of the multivariate peptide classifier . In another analysis of the 2015 cohort results , a Pearson correlation was calculated for the transformed peptide intensities of Chagas-positive donors to their median signal over cut-off value ( S/CO ) from three serially conducted T . cruzi ELISA assays . Library peptides correlated to S/CO were identified using a 10% false discovery rate ( FDR ) criterion by the Benjamini-Hochberg method [43] or by the Bonferroni correction for multiplicity . To construct the Chagas and multi-disease experiment classifiers , features were ranked for their ability to discriminate samples based on the p value associated with a Welch’s t-test that compared Chagas positive to negative donors or that compared all four different disease classes , respectively . The number of peptides selected was varied from 5 to 4000 features progressively . The transformed intensities of each peptide were mean centered and scaled to unit variance . To train a classifier , these data were input to a support vector machine ( SVM ) [44] with a linear kernel and cost parameter of 0 . 01 . A four-fold or five-fold cross validation procedure was repeated 100 times and used to assess model performance . This was estimated as the area under the receiver-operating characteristic ( ROC ) curve , which incorporated both feature selection and classifier training steps to avoid bias . A classifier was fixed that comprised an optimal number of ranked input peptides based on the cross-validated performance estimates of the 2015 Chagas cohort . This fixed SVM classifier model was evaluated without further optimization ( fixed coefficients ) by using it to predict the classification of samples in the independent 2016 Chagas cohort , which served as a test set for verification of the algorithm . The blood bank’s positivity assignments of these test samples were blinded to the researchers conducting the IST assay until the results were obtained . This classifier was also used in assessing signal precision and performance reproducibility of the platform . All analyses were performed using R version 3 . 2 . 5 [45] . SVM classifiers were developed using the e1071 package [46] . Library peptides were aligned to the T . cruzi CL Brener [47] and Sylvio proteomes downloaded October 2016 from UniProt . A BLAST strategy [48] , requiring a seed of 3 a . a . ’s , a gap penalty of 4 a . a . ’s , and a scoring matrix of BLOSUM62 [49] modified to reflect the a . a . composition of the array [50] . These modifications increased the score of conserved substitutions , removed penalties for a . a . ’s absent from the array and scored all exact matches equally . An alignment score is assigned to each a . a . position of a protein sequence to which a given set of peptides align; the overlap score is the sum of these a . a . alignment scores . To correct this score for library composition , another overlap score was calculated using the identical method but for a list of all array peptides . This allowed for the calculation of a peptide overlap difference score , s , at each a . a . position via the equation: s=a− ( b/d ) *c In this equation , a is the overlap score from the classifying peptides , b is the number of classifying peptides , c is the overlap score for the full library of peptides and d is the number of peptides in the library . To convert these s scores ( which were at the a . a . level ) to a full-protein statistic , the sum of these scores for every possible 20-mer epitope tiled acrossa protein was calculated . The final protein epitope score , S , was the maximum score along the rolling window of 20-mers for each protein . A similar set of scores was calculated for 100 iterative-rounds of randomly selected peptides from the library , equal in number to the number of classifying peptides . The p-value for each score S was calculated by permutation test from the number of times this score was met or exceeded among the randomly selected peptides , controlling for the number of iterations . The precision of antibody binding to the array features was characterized using a set of eight plasma samples to measure the fluorescent signals of the library peptides that comprised the fixed model for classifying T . cruzi seropositive ( Chagas positive ) versus seronegative ( Chagas negative ) donor samples . The fixed classifier was fit using data from the 2015 cohort and then applied to these slides without any further optimization . Four Chagas positive samples associated with a range of median ELISA S/CO values and three Chagas negative samples were selected from the 2015 cohort of donors , and assayed in triplicate . A well-characterized in-house plasma sample from a healthy donor was also included in the slide design , assayed in duplicate . As a negative control , one array was assayed in the absence of plasma during the primary incubation step but in the presence of the secondary detection antibody . These 24 samples were assigned to array positions on a single slide such that replicates were evenly distributed . This slide layout was repeated across slides according to the designs specified below . Signal precision of the peptides comprising the fixed classifier was determined for each study . The normalized readouts were fit to a mixed effects model from which inter-array , inter-slide , inter-wafer , inter-day , and inter-wafer batch CVs were calculated . Each donor sample was treated as a fixed effect . The nested factors ‘wafer’ or ‘wafer batch’ , ‘slide’ , and ‘array’ were crossed with ‘day’ , and these were treated as random effects . Models were fit in R using the lme4 package [51] to derive coefficients of variance ( CV ) . To evaluate precision within a manufacturing batch , three wafers from a single batch were selected . Twelve of the thirteen slides from each of these wafers were evaluated using the one-slide precision design described above . The 36 slides were evaluated across three ArrayIt cassettes , which each carry four-slides , on three different days . Slides from each wafer were assigned evenly across the three days such that each cassette contained two slides from one of the three wafers and one slide each from the remaining two wafers . To measure precision across wafer batches , one wafer was selected from each of four different manufacturing batches . Twelve of the thirteen slides from each wafer were evaluated using the precision study sample-set described above . These slides were distributed for testing across four cassettes per day , spanning three days . Slides from each wafer were distributed evenly across the three days such that each cassette contained two slides from two of the four wafers . As an internal QC assessment of the robustness of the fixed Chagas classifier across many wafer manufacturing batches and assay days , a quality control ( QC ) sample panel was designed for testing on a single slide . The layout comprised a representative panel of 11 Chagas positive samples , 11 Chagas negative samples , a well characterized in-house healthy donor sample and a secondary-only array . This panel was assayed on a slide from 22 different wafers manufactured across 10 different synthesis batches . For each of the 22 wafer-slides tested , the fixed model classifier developed in the Chagas 2015 trial was applied to the QC set of 22 Chagas positive or negative samples to estimate area under the ROC curve ( AUC ) . One of these wafers included in this analysis was used for the Chagas study ( Chagas positive and negative ) and another was used for the multi-disease study ( T . cruzi , HBV , HCV , & WNV ) trial .
A peptide synthesis protocol has been developed in which many a . a . coupling reactions are performed in parallel directly on silicon wafers using masks and photolithographic techniques . In the work presented here , arrays displaying a total of 131 , 712 peptides ( median length of 9 a . a . ) in situ synthesized at features of 14 μm x 14 μm were utilized to query antibody-binding events . The array layout includes 125 , 509 library-peptide features and 6203 control-peptide features attached to the surface via a common linker ( see Methods ) . The library peptides were designed to provide a sampling of all possible a . a . combinations and were not designed to match any proteomic sequences . Several sets of control-features were included in the array . These controls provided synthesis verification of intended peptide sequences , precision and reproducibility measurements , and fiducials for manufacturing and data processing . Experiments were conducted using mAbs that functionally evaluated the quality of final array-synthesized products with respect to ligand presentation and antibody recognition . A panel of four murine antibody clones ( 4C1 , p53Ab1 , p53Ab8 , and LnkB2 ) were selected with recognition sequences that correspond to four of the five epitope control-peptides designed within the array layout . These four epitopes collectively include all 16 a . a . that were used to synthesize the library . Fig 1 presents the results from a binding assay conducted as described ( see Methods ) in which each antibody was individually applied to an array with competitor agent , in triplicate . For each mAb , the control feature intensities were used to calculate a Z score for both the peptide sequence corresponding to its epitope , and the three non-cognate sequences . Each of the cognate sequences were bound with high signal intensity whereas the non-cognates displayed little or no signal above background values ( secondary antibody only ) . The dynamic ranges ( 95th percentile/5th percentile ) of the raw intensities of the library features were ~1 . 04 , reflecting binding to less than 1% of the library peptides . This indicates that the microarrays carry peptides suitable for specific antibody recognition and binding . Two cohorts of plasma samples collected from asymptomatic donors were obtained from a blood bank repository ( Creative Testing Solutions ( CTS ) , Tempe , AZ ) , and are described in S1 Table . CTS tests all blood collected by United Blood Services ( UBS ) against a panel of infectious diseases . The 2015 cohort is comprised of 335 donors that were each serologically tested for T . cruzi-specific antibodies using the blood bank’s algorithm . Three ELISAs were serially performed to assay plasma against T . cruzi whole cell lysate ( Ortho ELISA Test System ) . If any one of these scored positive by a signal to cutoff value ( S/CO > 1 . 0 ) , then a confirmatory test was performed . The confirmatory test was an immunoprecipitation assay ( Quest Diagnostic RIPA test ) that uses plasma to precipitate radiolabeled T . cruzi lysates . By these criteria 146 donors were determined to be T . cruzi seropositive , and were declared “Chagas positive” by the blood bank; 189 were determined to be T . cruzi seronegative , and were declared “Chagas negative” . This designation is widely used , although they are subclinical and it is understood that these Chagas positive individuals may or may not have remained parasite-infected at the time of the draw . A more accurate designation would be T . cruzi-antibody positive or negative . Literature sources consider an S/CO score above 4 . 0 to be strong positivity [52]; this assigns 49 of the 146 ( 33 . 5% ) seropositive donors into a strongly positive S/CO subgroup and 66 . 5% as weakly positive . The distributions of gender , age , and ethnicity were those typically observed in the UBS blood donor population ( http://www . unitedbloodservices . org/aboutUs . aspx ) , although not all demographic information was reported . The 2016 cohort comprised 116 donors that were tested for T . cruzi-antibodies with the same protocol of ELISA and RIPA testing described above . This second cohort was designed to contain 58 Chagas positive and 58 Chagas negative participants . A somewhat higher proportion of the seropositive individuals scored into the higher S/CO subgroup , 31 of 58 ( 53% ) . The distributions of gender , age , and ethnicity were like those of the 2015 cohort , with ethnicity reporting being more complete ( 89% versus 62% ) . The blood donors described here were located predominantly in the US West , reflective of the bias for Hispanic over black minority participants . The study trial presented here was conducted by using the 2015 cohort as an algorithm-training set to be used for developing a classifier that distinguishes T . cruzi seropositive from seronegative individuals . This classifier was fixed and then applied to the prediction of positivity for the 2016 cohort donor samples . Thus , analysis of the 2016 cohort served as a training-independent verification trial . All IST assays were performed as described ( Methods ) and scanned to acquire signal intensity measurements at each feature . Application of Welch’s t-test identified 356 peptide sequences that had significant differences in mean signal between those donors who were blood-bank scored as seropositive versus seronegative for T . cruzi- specific antibodies . The volcano graph in Fig 2 plots the ratio of mean signal intensities between seropositive and negative samples versus the significance of the differential , for each peptide on the array . A white dashed line demarcates the Bonferroni-corrected p value limit . Most , though not all , of the significantly different peptide signals displayed higher binding intensities in the seropositive as compared to seronegative donor samples . Approximately half of these class-distinguishing peptides had signal levels that were also positively correlated to the median T . cruzi S/CO value of donor samples declared Chagas positive ( shown as blue and green circles ) . This is consistent with the possibility that some library peptides bind the same or related plasma-antibodies as those bound by antigen in the ELISA screen . Conversely , there were 14 peptides that were significantly correlated to the S/CO value but did not meet the Bonferroni threshold for IST discrimination of Chagas positivity ( circles below white dashed line ) . The remaining half of the 356 peptides that showed the strongest discrimination by IST are notable for not being significantly correlated to ELISA S/CO values ( red dots above the white dashed line ) . This finding indicates that in addition to antibodies measured in both assays , unique antibody interactions were detected by IST . The 370 ( 356 + 14 ) library peptide sequences and associated raw signal data are tabulated in S2 Table . A support vector machine ( SVM ) classifier of Chagas positivity was developed in the 2015 cohort . The best average performance by cross-validation was achieved when the top 500 peptides , as ranked by Welch t-test , were input to the model . This number is greater than the 356 that met the Bonferroni significance cutoff , indicating that additional information was contained in some of the peptides that did not meet this stringent threshold . As few as 200 and as many as 4 , 000 library peptides could be used in the model with only modest reductions in performance ( S1 Fig ) . Fig 3A shows the relationship between mean sensitivity and specificity of 100 iterations of five-fold cross validations , using these 500 top-scoring peptides as a function of diagnostic threshold . The AUC estimate of 0 . 98 means that any donor chosen randomly from within the 2015 cohort would have a 98% probability of being classified as seropositive if it was a blood-bank positive , and a 2% of being called positive if it was a blood-bank negative , with a 95% confidence interval ( CI ) of 97%-99% . At the threshold where sensitivity equaled specificity , the accuracy was 93% ( CI = 91%-95% ) . The cross-validation estimates were confirmed by application of the 500 peptide SVM classifier developed with the 2015 cohort , to the independent 2016 cohort . The observed performance within this verification test set ( AUC = 97%; accuracy = 91% ) was within the 95% CI of the cross-validation estimates ( Fig 3B ) . The ROC curves take advantage of the continuous signal intensity values . Alternatively , a single , pre-specified positivity cut-off for Chagas positivity can be applied . Using 50% probability as an arbitrary diagnostic threshold to the predictions of the training set’s fixed model , performance metrics were calculated for the 2016 test set . Accuracy was determined to be 87% ( 95% CI = 81%-93% ) , sensitivity was 76% ( 95% CI = 63%-86% ) , specificity was 100% ( 95% CI = 94%-100% ) and Cohen’s kappa was 76% ( 95% CI = 64%-87% ) . No test results were excluded . Table 1 presents the 2x2 matrix for these results . There were no false positives , and 14 false negatives . Of these false negatives , 8 had S/CO values of less than 2 . 0 . A diagram of the sample IST testing flow and the blood bank’s ( CTS ) assignment of the same samples is provided in S2 Fig . This same fixed classifier was used to assess the binding precision of the assay using a protocol in which a set of samples: four Chagas positive , three Chagas negative , and one control , was repeatedly assayed as described in the Methods section . These precision measurements are presented in Table 2 . The values for inter-slide , inter-wafer , and inter-day are comparable to intra- and inter-assay CVs obtained with other T . cruzi ELISA tests . In addition to these precision studies , reproducibility of classification accuracy was determined across 22 different wafers using the QC sample layout described in the Methods; these analyses indicated AUCs >0 . 98 ( median AUC = 1 . 0 ) . The results in Fig 4 explore the heterogeneity of antibody binding across the 2015 Chagas cohort . The relative signal intensities are displayed for the 370 ( 356 + 14 ) peptides described in Fig 2 that provided significant discrimination of Chagas positivity by t-test , by correlation to the ELISA S/CO levels or both criteria . In Fig 4 , each peptide ( x axis ) for each donor ( y axis ) is represented , and is shaded relative to the difference in its intensity compared to the mean intensity of the same peptide in all seronegative donors , which serve as controls . The heatmap color scheme is scaled by the standard deviation ( sd ) of a feature’s signal from that of the seronegative controls . The legend has been truncated at 7 sd’s to permit smaller , but significant variations to be visualized . The donors were ordered by their median ELISA S/CO measurements , and these data are plotted along the left side of the heatmap . The peptides have been clustered as indicated by the dendrogram at the top . The distinction between ELISA seropositive and negative donors is evident as visualized in the IST heatmap , as are correlations within the ELISA seropositive samples to some peptides’ IST signal levels . The Chagas positive samples displayed at least three distinct binding profiles for a subset of the informative IST peptides: those with i ) uniformly lower signal than seronegative samples , ii ) marginally but uniformly higher signal than seronegative samples and iii ) signal that increases as S/CO values increase . Peptide signal heterogeneity in the Chagas negative samples is relatively minor . In addition to measuring IgG antibodies bound to the IST peptide array , IgA binding activity was measured by detecting plasma-antibody binding events with a fluorescently-labeled anti-IgA specific secondary reagent . Signal intensities of IgA binding events were generally lower than those of IgG binding , and fewer library peptides ( 224 versus 356 ) passed the Bonferroni cutoff for significantly different signal levels between the seropositive and negative donors . This can be visualized in a volcano plot ( S3 Fig ) , in which the ratio of mean signal intensities between seropositive and negative samples was plotted versus the significance of the differential , for each peptide on the array . Within the 224 IgA-bound distinguishing peptides , 53 overlapped with those detected by probing with the anti-IgG secondary reagent ( S4A Fig ) . The non-overlapping IgA peptide reactivities generally showed lower fold changes than that of the IgG reactivities . When measuring plasma IgA , all 23 peptides that met the Bonferroni threshold ( p <4e-7 ) for correlation to T . cruzi S/CO values were among the 26 peptides correlated to S/CO when measuring IgG binding events . The effect sizes of the T . cruzi positive versus negative serogroups as measured by anti-IgA secondary were generally lower when compared to those measured by anti-IgG secondary . The peptides with the best effect sizes were predominantly those that overlapped between both detection methods , anti-IgG and anti-IgA secondary reagents ( S4B Fig ) . The performance of the IgA classifier was strong , with a AUC of 0 . 94 and peak model size of 4 , 000 peptides ( S5 Fig ) , though modestly less than that of the IgG classifier ( AUC = 0 . 98 ) . Combining the peptide lists for feature selection did not improve classifier performance ( AUC = 0 . 98 ) , suggesting some redundancy in the binding event information . Although the peak model size was 2 , 000 in the combined feature selection , whereas the IgG and IgA selection peaks were 500 and 4 , 000 , respectively . The 356 library peptides that displayed significantly ( p <4e-7 ) different signal intensities between Chagas positive and negative donors were combined with the 14 library peptides that were significantly correlated to S/CO values , but did not meet the Bonferroni cutoff for IST . Together , these 370 peptides were considered informative relative to Chagas positivity , and were used to explore possible informatic alignment to the T . cruzi proteome . A modified BLAST algorithm and scoring system was developed that used a sliding window of 20-mers ( Methods ) . This yielded a ranked list of candidate protein regions ( targeted 20-mers ) shown in Table 3 , with redundant hits stripped . The informative library peptides yielding these candidate T . cruzi proteome targets displayed alignment scores that greatly exceed the maximum scores obtained by performing the same analysis ten times with equally-sized ( 370 ) sets of peptides that were randomly selected from the library ( Fig 5 ) . For example , the maximum score obtained with the iterative sets of randomly selected peptides ranged from less than 2000 to a maximum of 2500; whereas the informative peptides generated maximum alignment scores from 3300 to 3500 . Four of these top ten targets are protein regions from members of three families previously shown to be antigenic in Chagas patients . These are marked with asterisks in Table 3 . The top-scoring alignments of the Chagas informative peptides mapped to the C terminus of the mucin’s Muc II subfamily of surface glycoproteins . Only one protein ID is represented here; however , this C terminus is found with high sequence similarity in all 662 Muc family members ( see Discussion ) . This library-peptide aligned region of Muc II includes a glycosylphosphatidylinositol ( GPI ) attachment site and corresponds to a highly immunogenic epitope found in Chagas patients [54] . The a . a . ’s most frequently identified in the 63 aligned library peptides are summarized in Fig 6 as a modified WebLogo [55] . The corresponding T . cruzi sequence ( TrTrypDB Gene ID = TcCLB . 509753 . 300 ) is displayed along the x axis . Amino acid substitutions at any one position are shown vertically and the proportional coverage within the mapped library peptides is depicted by the height of the one-letter code . The total height of a letter-code bar indicates the absolute number of peptides aligning to the Muc II a . a . position . Another member of the Muc II protein subfamily is the sixth ranked target candidate , and it also maps to its C terminus . A member of a different T . cruzi surface glycoprotein family , the dispersed gene family proteins ( DGF-1 ) [56] , ranked eighth by the aligning algorithm . The library peptides mapped to its C-terminal region , which corresponds to the DGF-1 family’s consensus sequence that is shared by all 565 members . Other candidate targets showing high library-peptide alignment scores included proteins involved in calcium signal transduction ( calmodulin ) , vesicle trafficking ( vacuolar protein sorting-associated protein , Vps26 ) [57] or uncharacterized proteins . The ten top-ranked T . cruzi candidates , along with their subfamily and family members , were targeted by 222 of the aligned Chagas informative peptides . The modified WebLogos for each of the other 9 top proteome alignment targets is provided in S6 Fig . A study was designed to determine whether IST could discriminate Chagas positive samples from samples of other infectious diseases . A subset of 88 T . cruzi seropositive samples from the full Chagas 2015 cohort was re-assayed , alongside 88 HBV , 71 HCV , and 88 WNV disease-positive plasma samples . The virus samples were assigned positivity by both indirect serologic and direct nucleic acid testing at CTS . All study samples were reported positive for only one of the four diseases . The demographic data showed a mix of genders and ethnicities , and a range of ages ( S3 Table ) . Although demographic data are missing for many of the samples , there is a higher prevalence of Chagas positivity among Hispanic donors , consistent with disease prevalence in Central and South America . This higher prevalence was also seen within the full Chagas cohort ( S1 Table ) . All IST assays for this study were performed on the same day and scanned immediately to acquire signal intensity measurements at each feature . The raw data was imported into R for analysis . In a first analysis , a set of binary classifiers were developed to differentiate each of the four diseases from a combined class of the remaining three ( Table 4 ) . Performance metrics for each disease contrast and their corresponding 95% CI’s were determined by four-fold cross-validation analysis . The models generated similarly strong AUC’s , which ranged from 0 . 94 to 0 . 97 , and corresponded to accuracies of 87%-92% . Nominally , the binary contrast of Chagas versus the combined class ( Other ) was best performing and HBV versus Other was the weakest; however , the parenthetically shown CI’s overlapped . While the number of optimal SVM input peptides for each contrast varied widely from 50 to 16 , 000 peptides , the number of peptides used did not significantly change performance . For instance , an SVM model size ( k ) of 50 for T . cruzi versus Other generated an AUC of 0 . 97; a model size of 4 , 000 generated an AUC of 0 . 96 , indicating model size was robust . In the next analysis , a model was developed to classify all four serologic states simultaneously , with one set of selected peptides ( k = 75 ) , and one algorithm . This multiclass model had marginally improved performance over the binary classifiers shown in Table 4 . Namely , the four-fold cross validation analysis yielded multiclass AUC’s of 0 . 98 for Chagas , 0 . 96 for HBV , 0 . 95 for HCV , and 0 . 97 for WNV classifications . Table 5 presents the performance metrics of the assignments of each sample to a class based on its highest predicted probability . These probabilities assigned most samples to the same infectious disease class confirmed by CTS testing . In this confusion matrix , the performance of each binary contrast using the single multi-disease classifier is presented . The estimated overall multi-disease classification accuracy achieved 87% . Cohen’s unweighted kappa for the agreement between the true classes and the predictions of the multiclass model was calculated and determined to be 0 . 84 ( 95% CI , 0 . 79–0 . 89 ) , indicating significantly greater concordance than expected by chance . A heat map is presented in Fig 7 that illustrates the mean probabilities of class membership for out-of-bag cross-validation predictions using the multiclass model ( shown in Table 5 ) . This was determined for each of the 335 study cohort samples , encompassing all four disease classes . The blood bank assignments are considered here as “true” , and the IST probabilities of being assigned to a disease class as “predicted” . The map presents the probabilities of each sample’s class assignment on a color scale from black to white; the samples are ordered along the x-axis by their true assignment . This enables the varying levels of IST-based distinction between classes to be visualized versus true classification . For example , prediction of true T . cruzi seropositive samples as T . cruzi versus each of the three viral diseases was strong , as the high probabilities ( bright colors ) in the correct class indicate . Even the few true T . cruzi samples with high probabilities for WNV prediction also displayed modest probabilities of a correct T . cruzi assignment . The predicted assignments of true HCV samples as HCV versus another liver virus , HBV , were weaker than those versus T . cruzi and WNV .
We have demonstrated the feasibility of using the ImmunoSignature Technology to detect T . cruzi-antibody positive individuals within a population of asymptomatic blood donors . The IST assignment of Chagas positivity closely matched that of the blood bank testing algorithms . A library of maximally-diverse peptides arrayed on a microchip were differentially bound by peripheral blood-antibodies in T . cruzi seropositive versus negative donors . These distinguishing peptides carried sequence-motifs with similarity to known immunogenic regions of several T . cruzi protein families , as well as several undescribed proteins that present here as possible antigens . In another study using the same library , both binary and simultaneous classifiers were developed that distinguished Chagas donors from those individuals who were positive but asymptomatic for three other blood panel tested diseases: hepatitis B , hepatitis C and West Nile virus . These exploratory diagnostic studies were conducted on a well-characterized microarray platform . The use of photolithography and masks for the in situ process provides the opportunity for production-level scaling and reproducibility . The binding assay has a workflow analogous to an ELISA , and is similarly amenable to electronic tracing and robotic workstations . The output from an IST binding assay is the quantified fluorescent measurements of antibody binding events that have occurred at every feature on an arrayed-library of peptides , which serve as epitope mimics . The diagnostic power derives from the identification of binding profiles , based on numerous selective antibody binding events , that are consistent within a disease class and consistently different relative to another . Notably , the same library design can be used to identify peptides that distinguish any disease class or health contrast , and a single assay can be used to simultaneously detect multiple serologic states . While we have optimized array manufacturing and assay conditions , we recognize that alternative library designs may provide improved overall performance or improvements for a particular diagnostic application . For example , increasing the number of peptides in the library will enable denser sampling of sequence space but would require either reducing the number of arrays manufactured per wafer or reducing feature size . Increasing the length of the library-peptides may enable longer linear and conformational epitopes to be mimicked; on the other hand , it may enable two or more antibodies with distinct recognition sites to find a mimic on the same peptide and thereby confuse interpretation of the collective signal . Our exploration of these parameters is in progress . The first Chagas study demonstrated that signal intensities of some of the library peptides which significantly distinguished the T . cruzi seropositive from seronegative disease states were also significantly correlated to ELISA-derived S/CO values . This finding provides an orthogonal verification of the IST test results relative to the current diagnostic standards . This finding also suggests that similar antibodies are being detected by the two different test platforms . However , there were many array binding events that did not correlate with the donor’s S/CO value yet were determined to differ significantly between classes and strengthened the IST classifier . This suggests that in addition to commonly measured antibodies , there are also unique antibody-binding events measured only by the IST platform . While beyond the scope of the work presented here , these unique events hold the potential for contributing to further assessments of T . cruzi-elicited immune activity . Next steps will be to investigate this possibility relative to clinically important stratifications of T . cruzi-related disease states . The peptides comprising the distinctive and consistent binding profiles of Chagas positive versus negative donors in the 2015 cohort were input to a machine learning algorithm to yield an estimated performance . A classifier was fixed , then verified by using it to predict the Chagas positivity status of an independent 2016 cohort of plasma samples . The accuracy of the IST assay evaluated here approached but was modestly reduced compared to that of the ELISA standard , as calculated in a high risk donor population trial [58] ( sensitivity = specificity at 91% versus 97% ) . However , since the ELISA was used to designate the “true” serologic class assignment , we cannot determine from this work whether the IST method was either more or less accurate than the ELISA . In an alternative analysis of the SVM classification results , a pre-specified cut-off of positivity was applied to each of the samples in the 2016 test cohort . Samples with a predicted Chagas probability of greater than 0 . 5 were classified as positive; those with probabilities of less than 0 . 5 were classified as negative . Cross tabulation of these assignments with the blood banks revealed zero false positives . The 14 differential assignments were all IST false negatives . We suggest that this result indicates either a lower IST sensitivity , or its higher specificity . This will be explored in future work using longitudinally collected samples from T . cruzi seropositive individuals with ultimately different clinical outcomes . The IST method may hold immune response information not extractable from current tests . To address this hypothesis , a high-resolution view of a deeper structure within the class-specific binding profiles was attempted . Within the set of peptides that were found to bind with significantly different intensities to the Chagas positive versus Chagas negative donor samples , intragroup heterogeneity was observed . For example , there were peptides that displayed low intensities and another set that showed high signal intensities versus the Chagas negative donors . In addition to these two T . cruzi seropositive class clusters of either uniformly low or high signal , about half of the peptides that distinguished the positive class from the negative displayed a range of signal intensity that increased with increasing S/CO values . These are the S/CO correlated peptides . There is a sub-cluster within these correlated peptides that increase gradually while another appears to transition to higher signal more suddenly at an S/CO of ~4 . 0 . Further work will be needed to determine whether there is any clinical significance to these different clusters , such as possibly indicating infection status or likelihood of progression to symptomatic disease [52] . As another means to potentially capture additional information about the T . cruzi-elicited antibody response , binding signals were detected using an anti-IgA , instead of anti-IgG , secondary antibody for detection . Given the mucosal route of parasite entry and evidence of protective mucosal immune responses to Chagas [59 , 60] , differential binding activity was anticipated and indeed found . However , no additional disease classifying peptides were identified in these interactions , suggesting that the same T . cruzi epitopes dominate both antibody isotype responses . An alignment strategy was attempted to explore whether the class-distinguishing peptides could be mapped to any T . cruzi proteome targets . The diverse sequence design of the peptides in our arrayed library is directed at using mimotopes of true epitopes to broadly sample antibody reactivities , without requiring knowledge of the antigen responsible . However , we hypothesized that some of these mimotopes might sufficiently resemble a previously characterized linear epitope to be identified by an alignment algorithm . This was a further challenge as T . cruzi genome harbors 12 , 000 genes , half of which carry repetitive sequences and mainly comprise large multigene families and retrotransposons [61] . Although overall synteny between T . cruzi and its most-related kinetoplastid , Leishmania major , is 75% , species-specific islands have been identified within the encoded surface glycoproteins [34] . These glycoproteins are suggested to be diagnostically important as cross-reactivity , especially with Leishmania spp . , is a significant source of false positive diagnosis in co-endemic regions [62] . In our analysis , the top ten scoring targets included the T . cruzi mucin ( Muc II ) and DGF-1 surface glycoprotein families . These linear peptidic hits support the interpretation that the IST assay is truly measuring T . cruzi antibody binding events , and leave open the possibility that the other distinguishing but unmapped peptides may mimic non-linear or non-peptidic ligands such as structural or glycosyl moieties . Alternatively , the unmapped peptides that differentially bound Chagas positive plasma may be mimicking non-parasitic self-antigens . While controversial , autoimmunity has been hypothesized as contributing to Chagas disease [63] . Mucins ( Muc I and Muc II ) comprise the third largest protein family of the T . cruzi proteome , holding 662 members , which together make up the protective glycoconjugate coat that covers the parasite surface . All Muc proteins contain an N terminal signal sequence for secretory pathway targeting , a C terminal GPI anchor attachment site , and a central region that is hypervariable ( HV ) [64] . The central domain of variable sequences stimulates a myriad of immune responses that lead to T cell anergy . By contrast , the GPI anchor region is both conserved and consistently immunogenic [34 , 54] . The highest scoring IST peptide-motif alignments mapped to the C terminus of the Muc II protein subfamily . Although this region is nearly identical in all members , another Muc II family member with a variant C terminus ranked sixth . A mucin-like glycoprotein called tryptomastigote small surface antigen ( TSSA ) has been reported to elicit strong antibody responses [65] yet was not targeted by our best aligning peptides . It is possible that it was missed because the combinatorial library used here was not designed for proteome mapping , even though we explored it . Alternatively , it is because TSSA is expressed only by the infective , bloodstream tryptomastigotes [66] . The samples analyzed here were from asymptomatic donors; these people were likely to be harboring intracellular amastigotes , associated with the chronic infection . The DGF-1’s are the fifth largest protein family with 565 N-glycosylated members [56] . Their conserved adhesion motifs suggest that they function similarly to integrins , and may play roles in extracellular matrix interactions and bidirectional signal transduction [67] . The consensus C terminus of the DGF-1 protein family ranked eighth in alignments by the IST peptide motifs . Members of two much smaller protein families , also involved in signal transduction and protein trafficking , were alignment targets: calmodulin and vacuolar protein sorting-associated protein 26 ( Vps26 ) . T . cruzi and human calmodulins display a high degree of similarity; the IST classifying peptides that aligned to T . cruzi calmodulin also aligned to the human protein . Human calcium/calmodulin-dependent protein kinase is known to play a critical role in cardiomyocyte survival and cardiomyopathy [68–70] . Finally , the identification of Vps26 is noteworthy as the protozoan transforms from tryptomastigote to amastigote within the vacuole of an infected host [10] . Together these protein families , collectively well over 1200 proteins , could account for 222 of the 370 mimetic peptides displaying highly significant , Chagas-distinguishing binding signals . While this represents a large portion of the antibody profile , there remains 148 mimetic responses to be explored . None of these protein families are currently in any purified antigen-based Chagas test . However , a large project conducted several years ago started with 400 recombinant Chagas proteins and identified 16 candidates in a bead-based screening assay for serologic reactivity [71] . Both a calmodulin and a DGF-1 protein were found to be consistently reactive in T . cruzi seropositive Chagas patients . However , full length or large protein fragments were used such that the epitopes within these targets were not defined for our comparison here . An earlier mentioned antigen discovery study used a tiled peptide array spanning known or predicted T . cruzi antigens to probe purified IgG’s from nine T . cruzi seropositive individuals [33] . Their results showed positive binding to peptides corresponding to the same region of the Muc II consensus sequence that we identified here . We suggest that a Chagas proteome-targeted array holds great promise for new T . cruzi antigen discovery and that the IST mimotope array may be complimentary in providing additional T . cruzi-epitope screening capacity . However , IST also provides a single technology for developing a scalable , diagnostic platform applicable to Chagas , in addition to other diseases or disease stratifications . The finding that a single test algorithm could distinguish Chagas positivity from that of several other blood panel diseases supports the interpretation that disease-specific antibody binding profiles can be identified on the mimetic peptide arrays . It is possible that a general state of immune activation such as inflammation might also be detected on the arrays , perhaps indirectly by some general change in IgG binding activity . However , any such antibody-bound peptides would not be anticipated to specifically distinguish one disease state from another and therefore would not be identified in the analysis . IST disease specificity was suggested by early studies with small cohorts ( n = 10 ) in which different sets of peptides were identified that bound serum samples from patients with an infectious disease or cancer [39 , 40] . Shown here , the multi-disease classifier of four blood panel diseases suggests that a single IST test , with a single set of peptides , might be developed for the detection of a panel of diseases . The ability to successfully co-classify diseases also suggests that IST may provide an opportunity for building a diagnostic with reduced cross-reactivity . The larger number of biomarker ligands may provide the improved specificity . Clinically relevant disease contrasts to assess next on this platform will be T . cruzi versus parasitic infections endemically found in the same regions such as Leishmania spp . [72] and Plasmodium spp [73] , and the nonpathogenic T . rangeli [74] . Whereas blood banks test samples from asymptomatic individuals , clinical settings require the diagnosis of both asymptomatic and symptomatic patients , in acute or chronic stages of disease . Evaluating the potential for IST to contribute in these contexts will require ongoing sample collections and experimentation . The subclasses of binding profiles within the Chagas positive group may contain information that we have not yet learned to extract or interpret . With the relevant sample cohorts and clinical annotation for training and testing , this might be achieved . While the express purpose in the presented work was not to discover new T . cruzi antigens or drug targets , or to develop a diagnostic for use in Chagas endemic areas , the results direct attention to these potential applications . Additional work will be required to assess the capability of the IST for addressing these clinical needs . Relative to the main objective of this work , the blood donor studies presented here show that IST is a viable tool for exploring the complexity of human immune responses to T . cruzi , and have highlighted platform-specific attributes that may hold advantages for Chagas diagnostics , prognostication , and monitoring of disease progression or resolution .
|
The most prevalent parasite infection in Latin America causes Chagas’ disease . Diagnosis is based on serologic tests because direct detection of T . cruzi is very difficult beyond its brief acute phase . The complexity of this parasite’s pathophysiology and the human immune system’s elaborate engagement with it makes current indirect diagnosis methods suboptimal . We applied the ImmunoSignature Technology to Chagas diagnosis . In addition to approximating the accuracy of current tests , a single IST test simultaneously distinguished four blood panel diseases , including Chagas . Other unique aspects include the demonstrated ability to identify subclasses within seropositive individuals . Unlike other tests , IST uses chemically-diverse peptides , not proteomic sequences , as antibody-capture ligands to identify antibody binding patterns that differ between health-states . Some of the Chagas-distinguishing ligands were enriched in sequence motifs that aligned to known T . cruzi epitopes; unmapped sequences suggest additional targets exist . While IST is also an indirect method , it holds greater capacity for delivering high-resolution serologic feedback . Relative to providing this test to the millions of individuals estimated to be infected with T . cruzi , IST is both highly scalable and reproducible .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"chemical",
"characterization",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"tropical",
"diseases",
"microbiology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"viruses",
"protozoans",
"rna",
"viruses",
"peptide",
"libraries",
"neglected",
"tropical",
"diseases",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"sequence",
"alignment",
"bioinformatics",
"proteins",
"medical",
"microbiology",
"immunoassays",
"microbial",
"pathogens",
"proteomics",
"protozoan",
"infections",
"binding",
"analysis",
"biochemistry",
"trypanosoma",
"cruzi",
"trypanosoma",
"chagas",
"disease",
"west",
"nile",
"virus",
"post-translational",
"modification",
"flaviviruses",
"viral",
"pathogens",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"signal",
"peptides",
"organisms"
] |
2017
|
An ImmunoSignature test distinguishes Trypanosoma cruzi, hepatitis B, hepatitis C and West Nile virus seropositivity among asymptomatic blood donors
|
Pulmonary emphysema is a connective tissue disease characterized by the progressive destruction of alveolar walls leading to airspace enlargement and decreased elastic recoil of the lung . However , the relationship between microscopic tissue structure and decline in stiffness of the lung is not well understood . In this study , we developed a 3D computational model of lung tissue in which a pre-strained cuboidal block of tissue was represented by a tessellation of space filling polyhedra , with each polyhedral unit-cell representing an alveolus . Destruction of alveolar walls was mimicked by eliminating faces that separate two polyhedral either randomly or in a spatially correlated manner , in which the highest force bearing walls were removed at each step . Simulations were carried out to establish a link between the geometries that emerged and the rate of decline in bulk modulus of the tissue block . The spatially correlated process set up by the force-based destruction lead to a significantly faster rate of decline in bulk modulus accompanied by highly heterogeneous structures than the random destruction pattern . Using the Karhunen-Loève transformation , an estimator of the change in bulk modulus from the first four moments of airspace cell volumes was setup . Simulations were then obtained for tissue destruction with different idealized alveolar geometry , levels of pre-strain , linear and nonlinear elasticity assumptions for alveolar walls and also mixed destruction patterns where both random and force-based destruction occurs simultaneously . In all these cases , the change in bulk modulus from cell volumes was accurately estimated . We conclude that microscopic structural changes in emphysema and the associated decline in tissue stiffness are linked by the spatial pattern of the destruction process .
Emphysema is a chronic obstructive pulmonary disease ( COPD ) that commonly occurs in conjunction with chronic bronchitis . While tobacco smoke is believed to be the primary risk factor for emphysema [1] , other factors such as environmental pollutants [2] , senescence [3] , [4] , nutrition [5]– and genetic predispositions [8] can also cause emphysema . Each of these risk factors triggers a series of interconnected biochemical processes that lead to cell death and the degradation of protein fibers that reinforce alveolar walls . Consequently , alveolar walls rupture and abnormally enlarged airspaces appear . Over time , the destruction of alveolar walls become progressive and patients experience increased difficulty in breathing [1] . In clinical settings , doctors rely on spirometric indices such as the amount of air that can be forcefully exhaled in 1 second ( ) to detect and characterize the progession of this disease [9] . However , significant destruction of tissue can occur at the microscopic scale before emphysema can be detected using [10] . In order to develop better diagnostic methods and to understand how emphysema progresses , it is essential to first understand the link between microscopic changes in structure and global measures of function . Emphysematous patients differ widely in the rate of decline in lung function as well as their morphometric characteristics . Despite this apparent heterogeneity , there is data suggesting that different morphological changes affect differentially lung function . For instance , in smokers , the decline in lung function is faster for patients with lesions concentrated in the upper zones of the lung as compared to patients with a more uniform destruction pattern [11] . In patients with -antitrypsin deficiency , a rare genetic form of emphysema , was found to correlate better with the extent of destruction evaluated from CT when the destruction is in the basal part of the lung [12] , [13] . Further , these macroscale patterns have also been linked to different microscale structures [14] , which in turn , have also been found to influence functional parameters such as lung compliance [15] . These findings suggest that there is a possible link between emphysema pathology , patterns of tissue destruction and decline in lung function . However , such a relationship between microscopic patterns of destruction and loss of function has not been identified . Previous studies have shown that lung compliance can vary significantly even when structural measurements made from two dimensional ( 2D ) sections of the lung did not show any significant change [15] , [16] . This apparent lack of structure-function relationship maybe due to the fact that 2D sections do not accurately represent the true three-dimensional ( 3D ) geometry of lung tissue and are significantly more error prone because of the noise introduced by sectioning [17] . Indeed , recent direct measurement of 3D alveolar structure has shown that the actual size of an alveolus measured in 3D is 1 . 5 times its current 2D estimate [18] . Therefore , the problem of relating structural and functional changes in emphysema is best analyzed in 3D . In this study , we developed a 3D model of lung tissue which incorporates the interdependent nature of the lung parenchyma– where upon destruction of an alveolar wall , the microscopic structure of tissue rearranges to balance forces [19] . Specifically , we modeled a block of lung tissue as a tessellation of space filling polyhedra , with each of the unit cells representing an idealized alveolus ( Fig . 1 ) . The model can be pre-strained and various destruction patterns can be mimicked by eliminating walls in a spatially random or in a correlated manner in which elimination is based on force carried by a face .
Fig . 2A shows a “cut-away” view of a pre-stressed cubic network which serves as the initial network from which faces were removed during the simulation of tissue destruction in emphysema . Fig . 2B , Fig . 2C and Fig . 2D show three different structures that emerged after random cutting , force-based cutting and a mixed cutting pattern , respectively . Despite the different geometries , these networks have similar values of . The results in panels A and B of Fig . 3 show how the mean and variance of alveolar airspace volumes change as the number of faces eliminated from the network shown in Fig . 2A is gradually increased . For the random destruction of faces , the mean volume increases at a faster rate than for the force-based pattern; however , this trend is reversed for the variance of cell volumes with the force-based destruction resulting in a much faster increase in the variability of airspace sizes . The early rise in the variability in the force-based simulations is similar to the experimentally observed early increase in airspace size variability in animal models of emphysema [18] , [21] . Fig . 3C shows the corresponding declines in : the force-based cutting method results in a fast and apparently linear decrease in whereas in the random cutting method , the decay in is curved and significantly slower . Fig . 4A and Fig . 4B show the change in with respect to the mean and variance of cell volumes . In Fig . 4A , there is a considerable spread between the values of for the random and the force-based cutting implying a significantly lower macroscopic stiffness in the force-based case at the same mean cell volume . Therefore , an important result is that a given macroscopic stiffness does not correspond to a well defined mean cell volume . However , the spread decreases considerably when is plotted against the variance of cell volumes ( Fig . 4B ) . This suggests that changes in higher order moments of the cell volumes may provide a better predictive relation between structural changes and decline in . We considered four moments of the cell volume distribution: the mean and three moments defined by . Due to the interconnected nature of the lung parenchyma , we reasoned that the way change with respect to one another during the cutting process , specifically the cross correlations between the time series , obtained from the model , would be indicative of the spatial pattern in the destruction process . To analyze the cross correlations in , we form the matrix , where each column in corresponds to a moment and each row corresponds to a step in the cutting process . To reduce dimensionality of the problem , we remove redundant information in using the principal component analysis [22]–[25] . This is done by projecting onto a new basis given by the eigen vectors of . Fig . 5 shows that 99% of the variability in is explained by the first two eigen vectors , and of , so that we only need to consider 2 dimensional data given by the projection of onto the subspace defined by and . We will refer to this new transformed data as and . Note that and are merely the first four moments of the cell volumes projected onto a 2D subspace . We now proceed to obtain structure-function relationships using the transformed variables and . Fig . 6 shows the decline in plotted against and for random ( green circles ) and force-based ( red squares ) destruction patterns . Also shown in Fig . 6 is the minimum mean square error fit of the equation ( 1 ) to the data obtained from the random and force-based simulations . Note that a single equation was able to fit both destruction patterns . The parameters of the fit were , , . To test our hypothesis that it is the pattern of destruction that determines changes in microscopic structure and decline in function , in Fig . 7 we plot the moments from independent simulations of force-based destruction projected on to the same , basis vectors as shown in Fig . 6 . First , we considered force-based destruction on a network where the prestrain imposed was 1 . 5 times higher than the original set ( magenta squares ) , we then considered a force-based destruction on a network made up of 14-hedral unit cells ( blue squares ) and finally we considered a nonlinear spring network where the springs developed force in response to strain as + , . Although we only tested one particular nonlinear force-strain relation , the destruction pattern does not depend on since the order in which springs are chosen for removal in the force-based cutting should not change for monotonously increasing . The trajectory of ( , , ) in all these cases lie very close to the original force-based cutting simulations ( red squares ) . We have also added to the plane , data from a random cutting simulation with the initial network strained to a value 1 . 5 times higher than the original set ( orange circles ) . Finally , we also considered a simulation with a mixed cutting pattern with = 0 . 1 ( yellow triangles ) . In all these cases , the trajectories of ( , , ) are well approximated by the plane defined by Eq . 1 . To examine the how well decline in stiffness can be estimated from structural measurements , we calculate the relative estimation error given by ( 2 ) where , given by Eq . 1 , is an estimate of stiffness from structural measurements and is the actual measured value . In order to identify the destruction patterns in which structural changes do not yield information about decline in stiffness ( i . e . , simulations with maximum error ) , we examine the box plot of estimation error for each destruction pattern ( Fig . 8 ) . We found that when the destruction pattern is independent ( spatially random ) , the estimation of decline in stiffness from structure lead to very high prediction errors ( 15% ) . However , when the pattern of destruction was spatially correlated ( mixed and force based ) , the maximum error was less than 8% and the median error was less than 4% .
Traditionally , emphysema is subdivided into two major categories based on the location of destruction within the pulmonary acinus [26] . In centrilobular emphysema , which is more common and often associated with smoking , the destruction occurs mainly in the distal part of the proximal acinus . In panacinar emphysema , which is associated with antitrypsin deficiency , more of the destruction occurs in the distal regions . At the scale of the whole lung , centrilobular emphysema shows signs of tissue destruction in the upper zones of the lung ( the upper lobe and the superior segment of lower lobe ) while in panacinar emphysema the destruction mostly involves lower zones and the anterior margins of the lung . At the microscopic scale , these two categories have different appearance on histological sections [14] . Further , they have also been shown to have different functional properties [27] . Saetta et al . [15] found that when human emphysema patients were classified into four categories based on patterns observed in 2D histological images ranging from a very homogeneous destruction pattern to a highly heterogeneous pattern , the subjects showed a significant difference in static compliance . These findings suggest that there is a strong link between the patterns of destruction at the microscopic scale , the geometry observed on histological sections and macroscopic lung function . Interestingly however , Saetta and coworkers found no differences in the mean linear intercept between the different groups in their study . A similar lack of correlation between tissue structure , as quantified by the mean linear intercept , and lung compliance has also been noted in animal models of emphysema [16] . In this study , we found that the decline in lung tissue stiffness was significantly influenced not only by the amount of tissue loss , but also by the spatial pattern of the destruction process ( see Fig . 3C ) , whereby to achieve the same 60% drop in , nearly twice as many faces had to be removed in the random destruction as compared to the force-based destruction . This conclusion is in agreement with previously published observations from 2D models [10] . We also found that these two cutting methods resulted in very different geometries ( Fig . 2 ) , with the correlated destruction leading to more heterogeneous structures . This finding has important implications on the characterization of emphysema from histological sections . Our results indicate that the heterogeneity in microscopic structure observed in the early stages of emphysema [18] , [21] , is an indicator of the pattern of destruction and hence is also indicative of the extent of decline in tissue stiffness . One possible reason for the disconnect between structure and compliance noted above maybe due to the fact that currently accepted standards for quantifying structural changes in emphysema do not account for the patterns in tissue destruction [28] . In a recent study [29] , we examined the relation among alveolar structure , tissue composition and lung function . Specifically , respiratory compliance C was correlated to biochemical and structural parameters of the mouse lung before and after elastase-induced emphysema . Interestingly , C did not correlate with bulk measures of soluble type I collagen , type III collagen or elastin . There was , however , a strong association between C and the mean equivalent diameter of airspaces ( ) , and a much stronger relation between C and the area weighted mean diameter ( ) with R values of 0 . 675 ( ) and 0 . 933 ( ) , respectively . Since includes higher order moments of the distribution of diameters [30] , it is highly sensitive to structural heterogeneities and hence patterns . Thus , there is now experimental data showing that it is not the mean airspace size , but its heterogeneity that determines function in agreement with the network analysis we presented here . Several previous publications have used different 3D models to examine the elasticity of normal lung tissue . Kimmel and Budiansky [31] employed a dodecahedral model to calculate elastic moduli for small deformations about a state of uniform expansion . More sophisticated models have later been proposed to examine non-uniform , large deformations [32]–[34] . Denny and Schroeter [33] also examined , using 3D models , changes in tissue elasticity when the relative amount of collagen versus elastin is perturbed as happens in the early stages of emphysema . However , to the best of our knowledge , the change in tissue elasticity associated with destruction of alveolar walls and its relation to structural changes have not been examined thus far . In order to examine how our results compare to observations in real emphysema , it is important to consider the factors that influence in real lung tissue and the limitations of the present model . In this study , we considered a small block of tissue far away from the major airways and devoid of ducts . The tissue network of the lung is usually classified into 3 interdependent compartments: a peripheral tissue system consisting of the pleural membrane and the interlobular membranes , an axial system which forms the alveolar ducts and surrounds the mouth of alveoli where they join the ducts and the fiber network that forms the alveolar septa [35] , [36] . Since the predominant structural change in emphysema is destruction of alveolar septa [37] , [38] , it is only this part of the lung tissue that we considered in this study . In our model , we only considered the recoil forces provided by the protein fibers that make up the ECM . However , the walls of the alveoli are coated with a liquid layer that provides surface tension at the air liquid interface . The value of surface tension is lowered by surfactant released by epithelial cells [39] . Surface tension forces act in two ways , they provide a recoil pressure [40] and , additionally , they distort the parenchymal geometry thereby providing an indirect contribution to the recoil forces [41] . The problem of how surface tension changes in emphysema may affect functional properties has been studied using models [42] . However , a recent experimental study suggested no change in surface tension in the lung due to emphysema [43] . Since the airspace sizes are generally larger in emphysema , surface forces likely decrease and the effect of surface tension may not be important in affecting the process of tissue destruction . Hence , we neglected the contribution of surface tension to elastic recoil . We simulated tissue destruction in a pre-strained network with the outer boundary fixed . In this case , during the destruction process , the total volume of the network is conserved . If we changed the boundary condition to a pressure boundary condition , which is perhaps more realistic , then , as tissue is destroyed , the whole network would expand outward thereby increasing the total volume . In this case , a network with such a boundary condition would distribute stresses differently after an alveolar wall is destroyed and the pattern of force-based destruction would be somewhat different from a fixed boundary condition . Nevertheless , the pressure boundary is also not fully consistent with physiology since the distending stress around the lung is maintained by the balance of the nonlinear chest wall elasticity and the changing lung recoil during the progression of emphysema . These issues represent significant additional computational challenges and will have to be examined in future studies . Before concluding , we note the following implication of the modeling results to disease progression . As Fig . 3C demonstrates , the pattern of tissue destruction has a significant impact on the rate of stiffness decline . Whether or not correlated patterns develop depends on the presence or absence of mechanical forces . In the normal lung , mechanical forces are present everywhere due to the negative pleural pressure surrounding the lung . However , these forces are not sufficient to rupture the tissue . In the diseased lungs , enzymatic damage weakens the tissue and lowers its failure threshold allowing mechanical forces of breathing to rupture the alveolar septal walls [20] . Thus , a lack of heterogeneity in the emphysematous tissue structure implies that mechanical forces are small or not dominating the destruction process and , consequently , the progressive nature of the disease must be driven by biological mechanisms that produce strong enzymatic tissue digestion . This may have further implication for treatment since enzymatic activity may be attenuated pharmacologically whereas eliminating mechanical forces is not feasible since it would lead to lung collapse . To summarize , we have developed a 3D computational model and a general framework to relate structural changes characterized by the cell volume distribution to functional changes . Our simulations demonstrate that different destruction mechanisms lead to grossly different microscopic destruction patterns which in turn result in different cell volume distributions and macroscopic declines in function for the same amount of tissue loss . It is therefore important to consider both the total amount as well as the spatial history of the destruction process in an attempt to relate structure to function . Further , our results suggest that changes in variability and higher order moments of the alveolar dimensions are not only important in determining changes in function but observing the corresponding structural patterns may also provide insight into the mechanism of disease progression . Finally , appropriate application of the uncovered structure-function relations to real lungs may in the future help evaluate the efficacy of therapies or novel drugs .
To model the elastic properties of the alveolar wall , each face of an idealized alveolus has linear hookean springs connecting its vertices to its centroid ( Fig . 1A and Fig . 1C ) . Additionally , springs are also placed along the edges of the polyhedron . These face and edge springs represent the combined effect of collagen and elastin fibers that are considered to be the two major force-bearing components that make up the alveolar wall . For small strains , we will assume that these springs develop a force in response to an applied strain as ( 3 ) The assumption here is that all the springs are made of the same material so that the constants is a property of the material analogous to the Young's modulus . It should be noted that springs in the model do not support compression . Consequently , the networks shown in Fig . 1 are inherently unstable and will collapse when subjected to a shear deformation . In order to stabilize the model , it is necessary to apply a pre-strain to the structure . The model is capable of being pre-strained in 3 different ways . 1 . Fixed Boundary: The entire network is subjected to a uniform expansion and the vertices along the faces that make up the exterior boundary ( boundary nodes ) are fixed . When the entire network is stretched out uniformly , individual springs become stretched and the equilibrium configuration is then determined by minimizing the total energy of the network given by ( 4 ) where s are the unstretched lengths of the springs and the summation is carried out over all the springs in the network . 2 . Force boundary: To each boundary node , , an external force is applied . The set of forces determine the boundary condition . The equilibrium configuration is calculated by minimizing the free energy , which takes into account the internal energy of the spring network and the work done by the external forces . The minimization is carried out on , changes in which are defined by: ( 5 ) where is given by Eq . 4 and is the position vector of the boundary node . 3 . Pressure Boundary: A negative external pressure ( ) can be applied to the entire network and the equilibrium configuration can be calculated by minimizing the free energy which accounts for the total internal energy of the network and the work done by the applied pressure . The change in free energy is given by ( 6 ) where is given by Eq . 4 and is the current volume of the expanding network . In the model , a face is composed of a set of non-overlapping triangles , so that the pressure acting on a triangular facet with outward normal area vector generates a force to act on the three nodes that make up the triangular facet . The network is then allowed to reach an energy minimum as described below . As the network equilibrates , the change in geometry of the network causes the vectors to change . The new set of area vectors and forces on the boundary nodes are updated . The minimization procedure is repeated until the change in between two successive minimization steps falls below a preset error bound . Changes in the microscopic structure of the network as a result of tissue destruction are tracked by recording the volume of every cell in the network . The change in macroscopic mechanical properties of the network are also tracked by measuring the bulk modulus , which indicates the ability of the material to resist a small uniform expansion and the shear modulus , which is a measure of the materials ability to resist small iso-volume shape distortions . Principal Component Analysis ( PCA ) also known as the Karhunen-Loève transform or the Hotelling transform [22]–[25] is a common method for removing correlations in an input data set . This is done by projecting the input data onto a new basis which is derived from the original data set . Let represent the original data set with the vector representing different observations of correlated input variables . We first form the covariance matrix given by ( 9 ) is a square matrix whose diagonal elements are the variance of the i variable in and are the covariance of variables corresponding to and . Next we find the eigen vectors and eigen values of . The original data set , , can be transformed to a new data set by projecting on to the new basis defined by the eigen vectors . The advantage in doing this is that , depending on the level of correlation in , usually decrease extremely fast and only the first few vectors in the new basis of need to be considered , so that the new projected data is usually of lower dimensionality than the original set .
|
Current standards for characterizing microscopic structural changes in emphysema are based on estimating the amount of tissue loss using stereological techniques . However , several previous studies reported that , in emphysema , there is a lack of correlation between stereological indices of tissue structure and increases in lung compliance , which is the inverse of tissue stiffness . In this study , we developed a novel three-dimensional computational model to show that the amount of tissue loss is not the sole determinant of increased lung compliance in emphysema . A key component that needs to be considered is the pattern of tissue destruction , which we demonstrate has a significant effect on the rate of decline in stiffness . Our findings also indicate that the heterogeneity observed at the microscopic scale in emphysema is a signature of the spatial history of the destruction process . These results highlight the importance of characterizing the heterogeneity of lung tissue structure in order to be able to relate microscopic structural changes to macroscopic functional measures such as lung compliance .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"physics",
"respiratory",
"medicine/copd",
"and",
"allied",
"disorders",
"physiology/respiratory",
"physiology"
] |
2011
|
Linking Microscopic Spatial Patterns of Tissue Destruction in Emphysema to Macroscopic Decline in Stiffness Using a 3D Computational Model
|
In vertebrate neurons , the axon initial segment ( AIS ) is specialized for action potential initiation . It is organized by a giant 480 Kd variant of ankyrin G ( AnkG ) that serves as an anchor for ion channels and is required for a plasma membrane diffusion barrier that excludes somatodendritic proteins from the axon . An unusually long exon required to encode this 480Kd variant is thought to have been inserted only recently during vertebrate evolution , so the giant ankyrin-based AIS scaffold has been viewed as a vertebrate adaptation for fast , precise signaling . We re-examined AIS evolution through phylogenomic analysis of ankyrins and by testing the role of ankyrins in proximal axon organization in a model multipolar Drosophila neuron ( ddaE ) . We find giant isoforms of ankyrin in all major bilaterian phyla , and present evidence in favor of a single common origin for giant ankyrins and the corresponding long exon in a bilaterian ancestor . This finding raises the question of whether giant ankyrin isoforms play a conserved role in AIS organization throughout the Bilateria . We examined this possibility by looking for conserved ankyrin-dependent AIS features in Drosophila ddaE neurons via live imaging . We found that ddaE neurons have an axonal diffusion barrier proximal to the cell body that requires a giant isoform of the neuronal ankyrin Ank2 . Furthermore , the potassium channel shal concentrates in the proximal axon in an Ank2-dependent manner . Our results indicate that the giant ankyrin-based cytoskeleton of the AIS may have evolved prior to the radiation of extant bilaterian lineages , much earlier than previously thought .
Polarized neurons with functionally distinct axons and dendrites are the foundation of the complex neuronal circuits in vertebrate nervous systems . The axon initial segment ( AIS ) plays a pivotal role in the function of polar vertebrate neurons as both a plasma membrane diffusion barrier that maintains the separate molecular identity of the axon , and as the site of action potential initiation . The presence of a plasma membrane diffusion barrier at the AIS that restricts lipid movement was first demonstrated in cultured neurons in 1992 [1] . Proteins were shown to have limited mobility at the AIS in 1999 . In this latter study , the barrier was found to depend on actin: when actin was depolymerized , axonal and dendritic plasma membrane proteins leaked across the barrier into the other compartment [2] . Since then other elegant experiments on cultured neurons have confirmed the existence of the plasma membrane diffusion barrier [3] , and have identified some of its molecular underpinnings . Giant isoforms of the cytoskeletal crosslinker ankyrin G ( AnkG ) localize to the AIS and nodes of Ranvier in myelinated axons [4] , and the longest 480 kD variant , formed by inclusion of an unusual 7 . 8 Kb exon [4] , is required for AIS organization [5] . Increasing amounts of AnkG are associated with reduced diffusion [3] , and targeting AnkG by RNAi allows increased diffusion of membrane tethered quantum dots in the AIS [6] . AnkG coordinates the AIS diffusion barrier in part by linking membrane proteins with the spectrin submembrane cytoskeletal network [7–9] . The high density of membrane proteins anchored at the AIS is believed to sterically limit the diffusion of other proteins and lipids through the plasma membrane [3 , 8 , 10] . Membrane proteins directly bound to AnkG are critical for AIS function as they include cell adhesion molecules and ion channels that control axon potential initiation . Thus AnkG is positioned as the central regulator of the AIS . While ankyrins are present across the bilateria , AnkG itself is a vertebrate specific gene [11] . It is therefore unclear which , if any , aspects of the AIS exist outside vertebrates . AnkG is one of three vertebrate ankyrin paralogs ( AnkB , AnkG and AnkR ) that are believed to have arisen from a single ancestral ankyrin gene during two large scale duplications early in vertebrate evolution [5 , 11] . Additional support for the idea that the AIS is a recent innovation comes from the observation that the AnkG binding sites on two vertebrate AIS-localized voltage-gated ion channels , voltage-gated Na+ channels and KCNQ K+ channels , can be traced back only as far as chordates and early vertebrates , respectively [12] . The AIS is thus frequently cited as a key evolutionary innovation within the vertebrate lineage that , together with myelin , allows the rapid , precise signaling necessary for our complex nervous systems [5 , 11–13] . Bilaterian ankyrins share a common motif structure that includes an N-terminal domain of 24 ankyrin repeats involved in transmembrane protein tethering , a central core involved in spectrin binding that consists of a tandem array of two ZU5 domains and one UPA domain , followed by a death domain ( DD ) of uncertain function . The ankyrin gene family can be traced as far back as cnidarians , a sister group to the bilaterians , although it has been suggested that this complete domain structure is absent in cnidarians and thus is bilaterian-specific [11] . The evolutionary history of the long exon present in vertebrate AnkG , which has been shown to be critical to AIS function [5] , is less clear . The current view is that the long exon was inserted into a vertebrate AnkG/AnkB ancestor only after the first of the two genome-scale rounds of duplications in vertebrates [5 , 11 , 14] . It is not present in vertebrate AnkR , which was a direct product of the first vertebrate genome-scale duplication , and has been reported as absent from ankyrin in the tunicate Ciona intestinalis [5] . Tunicates separated from the vertebrate lineage early in chordate evolution prior to the vertebrate ankyrin gene duplications . Interestingly , the ankyrin binding domain present on chordate voltage-gated Na+ channels predates these vertebrate gene duplications [12] , suggesting that ankyrins may have already been playing a role in the clustering of at least some axonal ion channels prior to the emergence of vertebrate AnkB and AnkG . Giant neuronal isoforms including unusually long exons have been found in the Drosophila neuronal ankyrin , Ank2 [15 , 16] , and the lone C . elegans ankyrin ( unc-44 ) [17 , 18] , but it has been postulated that these long exons have a separate evolutionary origin from the vertebrate long exon because 1 ) they do not share significant sequence homology with the long exon of vertebrate AnkG , 2 ) they are inserted downstream rather than upstream of the DD , and 3 ) they have not been found in intervening species such as Ciona [14] . The giant isoform of C . elegans unc-44 is required for normal axonal growth [19] , and an unidentified isoform of unc-44 is also required to maintain axon identity and concentrate the microtubule organizer CRMP to the proximal axon [20] . Drosophila Ank2 contains two long exons , XL and L , and the giant isoforms they encode work together to regulate axon diameter and synapse stability through microtubules [15 , 16 , 21] . Interestingly , genetic deletion of Ank2 leads to ectopic branching in the proximal axon of Drosophila sensory neurons [22] . However , a possible role for these long invertebrate ankyrins in generating AIS-like structures has not been examined to date . A late vertebrate origin for the AIS would contrast with findings that other fundamental features of polar neurons evolved much earlier , at least prior to the divergence of protostomes and deuterostomes , the two major bilaterian lineages . Many families of neuronal ion channels and receptors have origins in basal metazoans or even protozoans [23–29] , and diversification of the major neuronal voltage-gated ion channels was complete prior to the divergence of bilaterians and cnidarians [26 , 30–34] . Polar neurons with distinct axons and dendrites can be found in protostome invertebrate model organisms such as Drosophila and C . elegans [35] . Axons and dendrites in vertebrates , Drosophila and C . elegans can all be distinguished by shared differences in microtubule polarity: axons contain almost exclusively plus-end-out microtubules , while dendrites contain high percentages of minus end-out microtubules [35–39] . In addition to axon-dendrite polarity , γ-neurons in the Drosophila mushroom bodies have patterned concentrations of ankyrin and K+ channels in a proximal domain reminiscent of the vertebate AIS [40] . We therefore reasoned that at least some structural features of the vertebrate AIS may predate the divergence of the protostome and deuterostome lineages . In particular , the maintenance of distinct axons and dendrites in invertebrates suggests a possible role for an ankyrin-based diffusion barrier . Although it is not known whether invertebrate neurons have an AIS-like diffusion barrier , they do at least have the capacity to establish plasma membrane diffusion barriers . It was recently shown that Drosophila motor neurons in culture generate a diffusion barrier half way out their nascent axon as they develop [41] . The relationship between this barrier and the mammalian AIS diffusion barrier or its dependence on ankyrins is not known . We re-examined the possible evolutionary origins of the ankyrin-based AIS diffusion barrier taking a two-pronged approach . Here we report a broad evolutionary analysis of ankyrins , including giant isoforms encoded by long exons , and a functional analysis of the proximal axon of a model Drosophila sensory neuron . We find substantial evidence for a common evolutionary origin for giant ankyrins in a bilaterian ancestor . Furthermore , we present evidence for an AIS-like diffusional barrier in vivo in the proximal axon of a Drosophila sensory neuron , and patterns of ion channel localization , that depend on giant Ank2 isoforms . These results suggest that at least some of the neuronal functions of giant ankyrins evolved much earlier than previously thought , before the divergence of the major bilaterian lineages .
We first explored the evolutionary origin of ankyrins using TBLASTN searches of available sequence data from early metazoan lineages ( cnidarians , placozoans , sponges , ctenophores ) and choanoflagellates , the most closely related protozoans to the metazoan lineage . S1 Table lists species and data type searched ( genome or transcriptome ) , and a summary of ankyrin genes found , and amino acid sequences of the ankyrins are provided in S2 Table . We found one complete ankyrin with 24 ankyrin repeats , the ZU5-ZU5-UPA cassette and the DD in each of four cnidarian species ( Nematostella vectensis , starlet sea anemone [26]; Orbicella faveolata , star coral; Acropora millepora , stony coral [42]; and Exaiptasia pallida , brown anemone [43] ) , and in the placozoan Trichoplax adherens [44] . Fig 1 shows an amino acid alignment of the core domains between mouse AnkG , Drosophila Ank2 and the ankyrin orthologs from a cnidarian ( Nematostella vectensis , starlet sea anemone ) and a placozoan ( Trichoplax adhaerens ) . 336/1205 positions ( 28% ) in the alignment are identical in all 4 sequences and a further 345 positions ( 29% ) are identical in 3 of 4 sequences . Drosophila Ank2 and mouse AnkG share 64 . 4% amino acid identity across the Ank-UPA core , while Nematostella shares ~50% identity to the bilaterian ankyrins . Lower sequence identity for Nematostella ankyrin is expected given that cnidarians and bilaterians diverged well before the radiation of bilaterians; a similar small drop in sequence identity has been consistently observed for cnidarian ion channels compared to their bilaterian orthologs despite high conservation of functional properties [30–33 , 45–47] . Similarly , Trichoplax ankyrin , shares only 42 . 5–44 . 7% identity with the other sequences , consistent with the hypothesis that placozoans are the earliest diverging member of the eumetazoan clade , which includes placozoans cnidarians and bilaterians [27 , 44 , 48] . In contrast to our findings in cnidarians and placozoans , we were unable to identify ankyrin genes that contained ankyrin repeats with the ZU5-ZU5-UPA cassette in nine sponge species ( one genome and eight transcriptome assemblies , [28 , 49] ) , six ctenophores species ( 2 genomes and 6 transcriptome assemblies , [25 , 27] ) or two choanoflagelletes ( 2 genome assemblies , [29 , 50] ) ( S1 Table ) . Thus canonical ankyrins may have evolved in a common ancestor of eumetazoans after separation from the sponge and ctenophore lineages . This finding contrasts with a previous report that canonical ankyrins are bilaterian-specific and cnidarian ankyrins contained only the 24 ankyrin repeats and the first ZU5 domain [11] . However , it is consistent with the observation that many neuronal signaling proteins such as voltage-gated ion channel gene families also diversified at this time [24 , 26 , 31 , 44] . As giant ankyrins are important for ankyrin function at the AIS , we examined ankyrin loci and gene predictions from invertebrate bilaterians , cnidarians and placozoans for evidence of giant isoforms . Vertebrate giant ankyrins are generated by inclusion of a long exon after the UPA domain . We therefore searched for potential long exons ( defined here as > 3 Kb ) anywhere downstream of the UPA domain in invertebrate genomes and corresponding giant isoforms in transcriptome databases . Cnidarian and placozoan ankyrins do not appear to have giant isoforms encoded by long exons based on two lines of evidence . First , we found no giant isoforms in transcriptome data from three cnidarian species: Nematostella vectensis , Orbicella faveolata and Acropora millepora . Second , there were no ORFs > 1 Kb in genome sequence between the UPA domain exons of the ankyrin gene and the adjacent downstream gene in genome drafts for the cnidarians Nematostella vectensis , Orbicella faveolata and Exaptasia pallida and the placozoan Trichoplax adhaerens . These results suggest that while canonical ankyrins were present in a eumetazoan ancestor of cnidarians , placozoans and bilaterians , giant ankyrins are likely to be bilaterian-specific . To understand when in bilaterian evolution giant ankyrins appeared , and to determine whether insertion of long exons was likely to have occurred as a single event or multiple times , we analyzed ankyrin genes from key bilaterian species . We began with the lone ankyrin gene from the tunicate Ciona intestinalis because the absence of a gene prediction for a giant ankyrin with a long exon has been cited as evidence that giant ankyrins appeared only after the genome-wide duplications in vertebrates [5 , 14] . We mapped a predicted Ciona intestinalis ankyrin ortholog ( XM_009864015 ) containing the complete Ankyrin-ZU5-ZU5-UPA-DD structure ( but no long exon ) to the Ciona intestinalis genome draft . The gene prediction comprises 48 exons spanning ~33 Kb , with ZU5-ZU5-UPA cassette encoded in exons 27–36 and the DD encoded in exons 39 and 40 ( Fig 2A ) . Interestingly , we found a 14 Kb gap separating exons 38 and 39 which contained an unbroken 13 . 4 Kb ORF . A small EST cluster ( clones cien56932 , cijv010l23 , cijv033b07 , cidg851d03 , cilv031p13 and 031ZG09 ) mapping within this ORF is expressed in the nervous system ( Fig 2B , [51] ) , indicating that the ORF could encode a long exon for a giant ankyrin isoform . We therefore used PCR on cDNA from whole larvae to verify that this long ORF contains exonic sequence contiguous with the UPA domain on the 5’ end and the DD on the 3’ end; The intron/exon junctions we verified by PCR suggest a 13 , 344 bp exon lies within the long ORF . The giant ankyrin isoform predicted for inclusion of this exon is provided in S2 Table . To determine whether insertion of such a long ORF is a frequent event , we analyzed ORF length in the Ciona genome draft . We found 81 , 915 ORFs > 300 bp in the Ciona genome assembly with an average size of 436 ± 308 ( S . D . ) bp . The predicted long exon in the ankyrin gene is ~42 standard deviations larger than this average and is the 2nd largest ORF present in the entire genome assembly ( Fig 2C and 2D , S3 Table ) . Similar analyses show that the mouse AnkG long exon is in the 23 longest ORF in the mouse genome; the Drosophila Ank2 XL and L exons correspond to the largest and 84th largest ORFs in the Drosophila genome , and the C . elegans ankyrin ( unc-44 ) long exon sits within the longest ORF in the C . elegans genome ( Fig 2E–2G , S3 Table ) . Because the long ORFs in C . intestinalis , Drosophila , C . elegans and mouse ankyrin genes are all at the upper end of ORF size in their respective genomes , we considered that such uncommonly large ORFs might not be inserted as independent evolutionary events . Based on this initial finding of an unusually large ORF in the C . intestinalis ankyrin , we analyzed ankyrin genes in representative species across the Bilateria . We found evidence for giant ankyrins in genome and/or transcriptome data in all phylogenetic groups of bilaterians we examined including tunicates , cephalochordates , echinoderms , mollusks , annelids and diverse arthropods . Amino acid sequences and accession numbers of these ankyrins are provided in S2 Table , and long exon lengths and position ( relative to the DD ) are provided in S4 Table . 23/36 bilaterian ankyrins we examined have long exons and can encode giant isoforms , and an additional ankyrin ( from the giant earthworm Glossocolex paulistus ) has a giant isoform in transcriptome data . 8 bilaterian ankyrin loci , including 5 vertebrate AnkR orthologs and two insect Ank orthologs , do not have a potential long exon in genome sequence and thus are not able to encode giant isoforms . We identified four additional short ankyrin isoforms exclusively in transcriptome data from tunicates , echinoderms and Glossoscolex . Because we had no corresponding genome data for these species , we were not able to determine whether these genes also encode alternate giant isoforms . While most ankyrin loci contained a single long exon , 8/23 of the ones we examined had two tandem long exons , including Drosophila ( S4 Table ) . Two long exons in ankyrin from the cephalochordate Branchiostoma floridae are homologous and thus likely the result of a recent duplication of a single ancestral exon ( S1A Fig ) . Multiple long exons in the AnkB ortholog in zebrafish , green sea urchin ankyrin ( Lytechnicus variegatus ) and two horshoe crab ( Limulus polyphemus ) ankryins appear to be the result of recent intron insertions that split a single ancestral exon ( S1B Fig ) . The origin of the two long exons in Drosophila Ank2 , its flour beetle ( Tribolium castaneum ) ortholog and Aplysia ankyrin are less clear because the exons are not homologous and there are no ancestral homology blocks split by an intron . The size of the long ankyrin exons was quite variable , with a mean of 8103 ± 4850 bp ( S . D . , n = 30 ) . The bulk of the Drosophila XL exon has a core array of 93 highly-conserved repeats encoding 76 amino acids which has been postulated to serve as a molecular spacer , although only a fraction of the array is required for Ank2 function in maintaining axon diameter [21] . Arrays of homologous repeats are not present in vertebrate AnkGs [14] , and we found repeat arrays in only a subset of the long exons ( S4 Table ) . Repeat numbers varied from 5 to > 160 , the size of repeats varied from 36 bp– 228 bp , and only repeats from very closely related species had recognizable homology . Thus repeats are not a feature of ankyrin long exons that is conserved over large evolutionary distances . We conclude that giant ankyrins that include sequences encoded by very long exons/ORFs are present in all major bilaterian groups . While the ORFs are always thousands of nucleotides in length , their size and repeat content varies considerably . To better understand how the current set of ankyrins evolved in bilaterians including the mammalian short AnkR , giant AnkG , AnkB , and the protostome giant ankyrins , we built a maximum likelihood phylogeny for the ankyrin family using the ankyrin repeats and the ZU5-ZU5-UPA cassette ( Fig 3 ) . If giant ankyrins evolved independently in vertebrates and protostomes ( like Drosophila ) as has been previously proposed [5 , 14] , then we would expect the ankyrin genes without long exons to connect to the base of the bilaterian ankyrin tree through a continuous series of nodes , and giant ankyrins to arise multiple times from separate nodes . Instead , it is the short ankyrins that arise from isolated nodes in vertebrates ( AnkR ) , insects ( Drosophila Ank ) and Limulus ( Fig 3 , red circles ) . In contrast , the giant ankyrins , which we found in all bilaterian phyla examined , all connect to the base of the bilaterian ankyrin branch through nodes that connect to other giant ankyrins ( Fig 3 , green stars ) . The most parsimonious interpretation of the phylogeny is that ankyrins acquired a long exon and the ability to encode giant isoforms prior to the radiation of extant bilaterian phyla . For example , multiple genomes reveal that echinoderms , cephalochordates and tunicates have a single ankyrin gene with at least one long exon , implying that the long exon is ancestral in deuterostomes . Even if we assume that short ankyrins we found exclusively in transcriptome data ( from echinoderms , tunicates , and annelids , Fig 3 , blue circles ) are encoded by ankyrin genes that lack long exons , the giant ankyrins remain connected across the phylogeny ( Fig 3 ) . As phylogenies based on single genes do not always recapitulate species phylogenies based on more comprehensive data , we mapped the information about presence of giant ankyrins onto the known evolutionary relationships of major animal groups ( Fig 4 ) . Based on the types of ankyrins present in the extant species , we made predictions about whether giant ankyrins existed at each evolutionary node ( Fig 4 ) . Using this summary , several arguments against a single origin for giant ankyrins can be addressed . One argument against a common , ancestral origin for giant ankyrins is that the position of the long exon relative to the DD is different in Drosophila and mammals [14] . In Figs 3 and 4 the two positions are shown as light green and dark green stars , and mixed stars indicate a single cephalochordate sequence where the DD could not be identified . Interestingly the Drosophila position , after DD , is found quite broadly throughout protostomes ( Fig 4 ) . Most significantly this position after the DD is also found in multiple echinoderm ankyrins . Echinoderms are deuterostomes and this suggests that the position of the long exon after DD position is ancestral in both protostomes and deuterostomes ( Fig 4 ) . The simplest explanation to account for this overall pattern of long exon position is a single switch relative to the DD within the deuterostomes after the divergence of echinoderms and before the divergence of tunicates from the vertebrate lineage ( Fig 4 ) . The alternate explanation of a separate evolutionary origin for vertebrate and protostome long exons based on position requires loss of one of the longest ORFs in the genome followed by de novo insertion of a similarly rare long ORF in a neighboring position , all while maintaining continuity of long exon-containing ankyrins within the phylogeny . In contrast , a switch in the position of an existing long exon would require only a single step of exon shuffling . Another line of argument against a common origin for vertebrate and Drosophila giant ankyrins is that they lack homology . We therefore examined the homology of long exon-encoded polypeptide between key bilaterian species to see if homology is shared only between long exons with the same insertion site , which would strengthen the argument for a separate origin . However , we surprisingly did not find significant homology between mouse AnkG and long exon polypeptides from the deuterostome invertebrates Ciona intestinalis and S . purpuratus , species which have a single ankyrin gene and bracket the switch in long exon position ( S2 Fig ) . Since it is almost certain that the single ankyrin gene in Ciona and AnkB and AnkG in vertebrates , all of which share the same long exon insertion site , have a common ancestor , we conclude that the long exon sequence has evolved too rapidly for homology to be a useful indicator for common evolutionary origin . In support of this , there was no detectable homology shared between long exon polypeptides from Drosophila , C . elegans , or S . purpuratus , which share a common insertion site downstream of the DD , but represent deeply diverging bilaterian phyla ( S2 Fig ) . Although sequence homology seems to be rapidly lost , we found that long exon polypeptides share in common a strong amino acid composition bias with just two amino acids , Ser and Glu , accounting for approximately 25% of the polypeptide and the top six amino acids ( Ser , Glu , Thr , Asp , Lys and Pro ) accounting for >55% , even though they represent only 32 . 8% of all possible codons ( Fig 5A ) . The coding bias was present in long exons with or without repeat arrays . The Ankyrin-ZU5-ZU5-UPA core has a very different composition bias centered on amino acids common in the ankyrin repeats ( Ala , Val , Leu ) , while the most abundant 6 long exon amino acids account for only 31 . 8% , similar to their codon frequency ( Fig 5B ) . The exact functional significance of such a coding bias is unclear but it does suggest some level of commonality for giant isoform function and it is the most conserved feature of ankyrin long exon sequence . Based on the presence of unusually long exons that encode polypeptides with very distinctive amino acid composition in all bilaterian groups , we conclude that giant ankyrins most likely evolved in a common ancestor to all bilaterians . We hypothesized that if giant ankyrins have a common origin early in bilaterian evolution , then perhaps their function in the organization of the proximal axon might also be widespread across bilaterians . To test this hypothesis , we chose Drosophila dendritic arborization ( da ) sensory neurons as a model system to test ankyrin function . These cells are multipolar sensory neurons that extend their dendrites over the surface of the larva [52] . They are responsible for sensing contractions of the body wall to mediate coordinated movement [53] , for nociception [54] and light avoidance [55] . These multipolar neurons lie directly under the cuticle and epidermal cells while the axons dive below the surface and extend to the central nervous system where they make synapses in the ventral ganglion . The dendrites , cell body and proximal axon are thus accessible for live imaging in intact larvae [39 , 56] . Moreover , the same cell can be reproducibly labeled and identified in every animal [52] and the organization of the microtubule cytoskeleton is similar in these peripheral neurons and Drosophila central neurons [39] . For most experiments we used the ddaE neuron , one of the two dorsal class I neurons , which are responsible for sensing contractions of the body wall to mediate coordinated movement [53] . We focused on examining the role of Ank2 , because it is the only Drosophila ankyrin capable of encoding giant isoforms , it is enriched in neurons [57] , and giant Ank2 isoforms have other demonstrated roles in axon function [15 , 16] . When a protein containing YFP fused to aa1-1159 ( Ank2S ) [15] , which includes only the ankyrin repeats and the first ZU5 domain , was expressed in class I neurons , fluorescence extended out from the cell body quite evenly into the dendrites and axon , gradually tapering in both compartments with increasing distance ( Fig 6A and S3 Fig ) . In contrast , YFP-Ank2L4 , YFP fused to an Ank2 fragment that includes the first half of the L exon ( aa1530-3005 of the Ank2-LP protein on Flybase ) including a microtubule binding domain [15] , was concentrated in the part of the axon immediately adjacent to the cell body and very little was present in dendrites ( Fig 6B and 6C , S3 Fig ) . The distinct distribution pattern of Ank2L4 in the proximal axon is clear when compared to the general membrane marker mCD8-RFP , which was used to outline the cells ( Fig 6C ) . This ability of part of the polypeptide encoded by the L exon to concentrate in the proximal axon suggested that Drosophila giant ankyrins may well localize to the proximal axon in addition to distal axonal regions as has been reported for motor neurons [15 , 21] . We expressed two voltage-gated K+ channels fused to GFP , Elk-GFP [40] and GFP-Shal [58] , in ddaE neurons in order to see whether ion channels can also accumulate within the proximal axon . These two K+ channels have previously been shown to accumulate within the proximal axon of mushroom body γ-neurons [40] . In mammalian neurons , Shal channels ( Kv4 ) regulate spike shape and patterning [59] while Elk channels regulate firing threshold [60] . Both Elk-GFP and GFP-Shal concentrated within the beginning of the axon of ddaE neurons ( Fig 6D and 6E , S3 Fig ) , overlapping the proximal side of the AnkL4-YFP peak . The proximal axon of ddaE neurons thus appears to have an ability to accumulate membrane proteins such as ion channels . We hypothesized that if giant ankyrins play a role in plasma membrane protein tethering in the proximal axon of ddaE , then there should be an ankyrin-dependent plasma membrane diffusion barrier similar to that found in vertebrates . We expressed mCD8-GFP or mCD8-RFP in ddaE neurons to examine diffusion in the plasma membrane . mCD8-GFP is diffusible membrane marker used extensively in Drosophila . In the cell body mCD8-RFP ( Fig 7 ) and mCD8-GFP ( Fig 8 ) localize to both the ER and plasma membrane , but in the axon they are primarily at the plasma membrane and occasional internal vesicles ( Figs 7A and 8A ) . In a previous study mCD8-GFP was used to identify a plasma membrane diffusion barrier half way out developing axons of cultured neurons , and behaved similarly to other membrane markers including ROBO-GFP and a GPI-anchored GFP [41] . Here , we looked for diffusion barriers in axons in living animals for the first time . We first used fluorescent recovery after photobleaching ( FRAP ) to compare the mobility of mCD8-RFP in the proximal axon and dendrites , as no diffusion barrier has ever been described at the base of the dendrite . FRAP experiments were performed in whole , living 3-day old larvae . To normalize the results , the fluorescence intensity before bleaching was set to 100% and after bleaching it was set to 0 . Bleaching occurring during imaging itself was accounted for by taking the ratio of the fluorescence intensity in the bleached area and the fluorescence intensity in an unbleached area of the soma . Occasional blips in fluorescence occurred when vesicles trafficked through the axon . At the dendrite base , mCD8-RFP fluorescence routinely recovered to greater than 70% initial intensity by 60s , and by 90s it was often difficult to tell where the bleach region had been ( Fig 7A ) . In contrast , fluorescence in the proximal axon recovered to only about one third of the original intensity , and the bleach region could still be clearly seen at the end of the time course ( Fig 7A ) . For a direct comparison of recovery in the axon and dendrite , we performed simultaneous bleaching in both regions and made a movie of the FRAP experiment ( Movie 1 ) . Fluorescence very obviously recovers into the dendrite bleach area , while recovery into the axonal bleach area is very limited . Comparisons of the recovery plateau revealed that fluorescence recovery was significantly suppressed in axons relative to dendrites ( Fig 7B ) . This limited diffusion was not a general property of the axon as FRAP assays using an ER membrane marker Rtnl1-GFP ( protein trap G00071 [61] ) showed rapid recovery in the proximal axon at the same time point ( S4 Fig ) . We conclude that diffusional mobility of plasma membrane proteins is reduced in the proximal axon relative to the proximal dendrite . In cultured mammalian neurons , the AIS diffusion barrier is established at 7–10 days , well after axons and dendrites become morphologically distinct and take on their distinct microtubule organization [3 , 10] , which occurs between 4–7 days [62] . In Drosophila dendritic arborization neurons , axons emerge first during embryogenesis , followed by dendrites , and dendrite microtubule polarity becomes mature by the second day of larval life [63] . To test whether the diffusion barrier was in place by the time these other aspects of polarity emerged , we performed FRAP analysis of mCD8-RFP in the proximal axon of two day old larvae . At this time point , recovery was similar to that in dendrites ( Fig 7 ) indicating that , as in cultured mammalian neurons , restricted diffusion in the proximal axon is a relatively late polarized feature to develop . To determine whether limited diffusion represents a proximal axon diffusion barrier or is simply a generalized property of ddaE axons , we next performed FRAP experiments bleaching mCD8-GFP in the proximal axon ( 0–15 μm from the cell body ) as in Fig 7 , 20–50 μm from the cell body ( roughly within the Ank2L4 peak ) , or 70–140 μm from the cell body ( in the distal axon shaft ) . Measurements at the two distal sites were conducted 24h after ablating the neighboring ddaD neuron as its axon bundles with the ddaE axon ~25 μm from the cell body . We found that the diffusion barrier was limited to the proximal site 0–15 μm from the cell body; recovery of mCD8-GFP fluorescence was not significantly reduced relative to dendrites at either the middle or distal bleaching sites ( Fig 8A–8C ) . ddaD was not laser-ablated for bleaching at the proximal axon site for data shown in Figs 7 and 8 , but we confirmed that the diffusion barrier seen in the proximal axon remained when ddaD was ablated ( S5 Fig ) . Having shown that the axon diffusion barrier is restricted to a region proximal to the peak of Ank2L4 localization ( Fig 6C ) , we wished to determine whether the barrier indeed depended on Ank2 long exon-containing isoforms ( Referred to here as Ank2L ) . To reduce neuronal levels of Ank2 in animals in which development was otherwise normal , we expressed large RNA hairpins specifically in class I neurons with the Gal4-UAS system . This has been demonstrated to be an effective way to perform cell-type specific RNAi in Drosophila [64] . Hairpins are several hundred nucleotides long and are processed into dozens of different short hairpins that trigger destruction of their target . As a control we expressed a hairpin targeting γ-tubulin37C , a maternal gene that is not expressed in somatic cells [65] . In addition to the hairpin RNAs , mCD8-GFP and dicer2 were expressed in class I neurons . When an Ank2L-directed RNA hairpin was expressed in ddaE , mCD8-GFP fluorescence recovered to a higher plateau into the proximal axon after photobleaching than with control hairpins or hairpins targeting Ank or CRMP ( Fig 9 ) . The Ank2 hairpin used contains a sequence at the 3’ end of the L exon and thus specifically targets transcripts containing the L exon . Because the RNA hairpins are expressed in only a few cells in the animal compared to Ank2 itself , it is difficult to determine level of knockdown using qPCR , however the increased recovery specifically when Ank2 was targeted suggested that its protein levels were reduced and that its function is important for limiting diffusion in the proximal axon ( Fig 9B and 9C ) . We could not make any conclusions about the role of CRMP or Ank in the absence of verification of knockdown because they did not show a phenotype . To more rigorously test a specific requirement for giant Ank2 , we used a previously characterized P element insertion that disrupts the L exon [15 , 16] . We generated Drosophila larvae in which both copies of the L exon contained the f02001 insertion and mCD8-RFP was expressed in class I neurons . Diffusion in the ddaE proximal dendrite was similar to control neurons ( Fig 10 ) , but mCD8-RFP recovered to a significantly higher level after photobleaching in proximal axons of Ank2f02001 mutant neurons than in control neurons ( Fig 10 ) . Axonal membrane diffusion in the Ank2f02001 did not fully recover to dendritic levels , which could be due to incomplete removal of relevant long ankyrin isoforms by the Ank2f02001 mutation [15] , leftover maternal contribution of Ank2 , or ankyrin-independent mechanisms . The significant increase in diffusion in the Ank2f02001 mutant neurons strongly suggests that giant isoform ( s ) of Ank2 containing the L exon are required to establish a diffusion barrier in the proximal axon . We next examined whether GFP-Shal accumulation within the proximal axon depends on Ank2L . We used a Ank2f02001/Ank2f00518 trans-heterozygous background to reduce Ank2L; Ank2f00518 contains a p-element insertion in the common ankyrin repeat region and blocks expression of all Ank2 isoforms [15] . The peak accumulation of GFP-Shal occuring between ~10–50 μM from the cell body ( Fig 11A and 11B ) was greatly reduced in Ank2f02001/Ank2f00518 larvae ( Fig 11A and 11C ) . In the mutant background , axonal GFP-Shal concentration is highest immediately adjacent to the cell body and simply decays in concentration with distance . The accumulation of GFP-shal 10–50 μM from the cell body therefore appears to depend on Ank2 expression .
To assess whether giant ankyrins exist and function in the proximal axon outside the vertebrate lineage we performed extensive analysis of metazoan ankyrin genes to determine where complete ankyrins were present , and then determined which of these contained long exons that could potentially encode giant ankyrins . In parallel to this phylogenetic approach , we performed functional analysis to determine whether Drosophila neurons have a diffusion barrier in the proximal axon , and whether this depends on the Drosophila giant ankyrin . In contrast to the prevailing view in the field , we found evidence to support a single origin for giant ankyrins early in bilaterian evolution . This finding was previously obscured by rapid sequence divergence in the long exons included in giant isoforms . With the increased availability of transcriptome and genome data from phylogenetically diverse eumetazoans , we could circumvent this sequence homology limitation to trace the evolutionary origins of the long exons by their presence or absence and position . This analysis allowed us to determine that long exon-encoded giant ankyrins are the rule , not the exception , in bilaterians . We therefore propose a revised view of ankyrin evolution in which giant ankyrins have a single common evolutionary origin in ancestral bilaterians prior to the radiation of extant bilaterian phyla . While we do not know exactly when these giant ankyrins gained the ability to organize the AIS , the fact that they are required for diffusion barrier formation in the proximal axon in both mammals ( deuterostome ) and Drosophila ( protosome ) suggests that this function may be an ancestral feature of giant ankyrins . Drosophila Ank2 is unusual ( though not unique ) in the it has two long exons ( L and XL ) and can thus encode diverse giant isoforms . In this study we focused on disruption of isoforms that include the L exon to show that these giant isoforms are required to set up an axonal diffusion barrier and to localize a K+ channel marker ( GFP-Shal ) to the proximal axon . Because loss of L exon-containing isoforms is known to also disrupt localization and expression of XL exon-containing isoforms [16 , 21] , it is not currently possible for us to determine whether L , XL or both L and XL isoforms are required for the diffusion barrier and GFP-Shal localization in the proximal axon . The diffusion barrier we found occurs proximal to the peak of the Ank2L4 marker accumulation we observed in the axon , but the barrier nevertheless depends on Ank2 . This appears at odds with the model for cultured mammalian neurons that it is the density of membrane proteins attached to AnkG that forms the diffusion barrier [3] . However , it is important to note that the Ank2L4-GFP marker probably does not reflect the complete distribution of ankyrins within the proximal axon , and it is therefore possible that native ankyrin isoforms are present in the diffusion barrier region between the observed Ank2L4-GFP peak and the cell body . AIS patterning in cultured mammalian neurons is entirely dependent on giant AnkG , but not all membrane proteins concentrated in the AIS bind directly to AnkG . If the AIS is similarly complex in ddaE neurons , it alternatively could be that the diffusion barrier is formed by a protein-dense membrane region separate from ( but held in place by ) a neighboring giant Ank2-rich region . In cultured mammalian neurons , diffusion is limited in the AnkG-rich region [3] . However , membrane diffusion in mammalian axons has not been examined proximal to the cell body equivalent to where we find the barrier in ddaE neurons [2 , 3] , so the extent of overlap between diffusion barrier and AnkG is not fully known and a situation similar to that in ddaE neurons cannot be ruled out . The AIS as defined in mammalian neurons includes not only a special submembrane cytoskeleton organized by giant ankyrins and a plasma membrane diffusion barrier , but also a concentration of ion channels that control action potential initiation . While we did not look directly at native ion channel localization in this study , we showed that two potassium channels , Elk and Shal , can accumulate in the proximal axon and that the accumulation of Shal is dependent on Ank2 . We think it is likely that native ion channels will be found to accumulate in the proximal axon of ddaE neurons , as it is the concentration of membrane proteins including ion channels and cell adhesion molecules orchestrated by giant ankyrins that is believed to be the primary contributor to limited plasma membrane diffusion [3 , 8 , 10] . In a study focusing on the subcellular organization of Drosophila mushroom body γ neurons , a type of kenyon cell interneuron in the brain , a variety of different markers , including channels and cytoskeletal proteins were found to have specific localization patterns within the axon . Several markers including ankyrin , Elk and Shal were enriched in the proximal axon , and others were present only on one side of a boundary region [40] . Localization of one of this latter class of proteins , Fas2 , was sensitive to actin depolymerization [40] , which disrupts the AIS in mammalian cells [2] . The presence of a diffusion barrier and role of giant ankyrins in setting up these localization patterns was not examined . One point to note is that the kenyon cells are unipolar , so the specialized region is not adjacent to the cell body as in typical mammalian neurons , but is instead found further out the neurite , past the point where dendrites branch out . The multipolar geometry of dendritic arborization neurons used here is more similar to that of the mammalian neurons in which the AIS has been studied , and in this setting the diffusion barrier is next to the cell body as it is in mammals . The question of which native ion channels are clustered in the proximal axon of Drosophila neurons is an important one , and the answer will shed light on how and when the proximal axon evolved as the site of action potential initiation . Insect Na+ channels do not have the AnkG binding domain found on vertebrate neuronal Na+ channels . This binding site interestingly evolved early in the deuterostome lineage before the appearance of AnkG [12] . Our results now show that giant ankyrins were also present at this time , raising the possibility that giant ankyrins were mediating axonal Na+ channel clustering long before the evolution of vertebrates . While it is clear that the particular ankyrin binding domain found on vertebrate Na+ and KCNQ K+ channels is not conserved across the bilateria , this should not be taken as evidence to rule out an ancient , conserved role for ankyrins in clustering ion channels at the AIS or other axonal locations . A crystal structure of the ankyrin repeat array of a vertebrate ankyrin suggests that binding ankyrin domains could be highly diverse and probably cannot be predicted from primary sequence [66] . Furthermore , some mammalian AIS proteins such as Kv1 channels do not directly bind to AnkG [67] , but nevertheless show AnkG-dependent AIS localization [68] , presumably due to the central role AnkG plays in initiating AIS patterning . Therefore , the absence of this one ankyrin binding motif does not provide definitive evidence against ankyrin-dependent localization to the proximal axon . One can envision a scenario in which ankyrin-based tethering of channels in the axon and/or diffusional exclusion of somatodendritic channels evolved in ancestral bilaterians to facilitate axonal action potential initiation , but needed further refinement in vertebrates to accommodate the spatial constraints imposed by myelin . These refinements may have altered the composition of AIS-clustered channels over time . It will likely be necessary to experimentally determine a comprehensive map of channel localization in the proximal axon of neurons from key bilaterian model species such as Drosophila before reaching conclusions on how AIS function evolved within the bilateria . However , if organization of the plasma membrane proteins at the AIS is exclusively dependent on giant ankyrins , then it is unlikely that a specialized ankyrin-based AIS predates the bilaterians as we find no evidence for giant ankyrins in cnidarians or placozoans . The results we present here suggest that giant ankyrins have a common origin in a bilaterian ancestor , and that they may play an evolutionarily conserved role in patterning of the axon initial segment in protostomes and deuterostomes . These data require revision of the current model of axon evolution which places the giant ankyrin-based AIS as a relatively recent vertebrate-specific innovation . Further characterization of molecular architecture of axons in invertebrate model organisms will be needed to determine which features of the vertebrate AIS are indeed new innovations and which features are ancient and widely shared among bilaterians .
Ankyrin genes in genomes , genome annotations and transcriptomes could be reliably identified with TBLASTN [69] using mouse AnkG and Drosophila Ank2; these baits yielded an identical set of ankyrin hits in all databases tested . We first used only the ZU5-ZU5-UPA cassette of mouse ANKG and Drosophila Ank2 as queries for the TBLASTN searches because this cassette is specific to true ankyrin genes , whereas ankyrin repeats can be found in a wide range of proteins including several classes of TRP family ion channels [70 , 71] . Hits were verified as ankyrins if 1 ) they had the canonical ankyrin repeat cassette upstream of the ZU5-ZU5-UPA cassette , and 2 ) they had reciprocal best matches to the three mouse ankyrin orthologs when used as TBLASTN queries against the mouse RefSeq database . Gene predictions were manually adjusted if they had gaps in highly conserved regions and we were able to find the corresponding missing exon in the genome draft . Giant isoform predictions and transcripts were mapped to the corresponding genome draft to verify the presence of a long exon ( s ) and determine its position within the gene and boundaries . When necessary ( Ciona intestinalis for example ) , the presence or absence of potential long exons was determined by manually searching the ankyrin gene locus for ORFs between the UPA domain and the adjacent gene . 5’ and 3’ junctions of the Ciona intestinalis Ankyrin large exon were verified by PCR of cDNA from whole larvae . A PCR product spanning the 5’ long exon junction was amplified in two rounds using the following primers for the first round: 5’-TTAACATTCTACAAACTTCCACTGG-3’ ( sense ) and 5’-GAGGCCTCTTCTTTAATTATCACTTC-3’ ( antisense ) . A faint ~2 . 7 Kb band from this reaction was gel purified and amplified with the same sense oligo and a nested antisense oligo ( 5’- ATGGTTTGGAGAATCAGGTGAG-3’ ) to obtain an ~ 1 . 6 Kb band sufficient for cloning . An ~ 1Kb band spanning the 3’ junction was obtained using the same strategy; First round primers were 5’-CTGTTTCACCTGGTTTATCCCGTAGC-3’ ( sense ) and 5’-GCACAAATCTCCTCGACCAATACTAT-3’ ( antisense ) , and in the second round the sense primer was replaced with 5’- ACTTCACGGTCATGCAACTGCTCCTT-3’ . Both products were cloned and sequence verified to determine the Ciona ankyrin giant isoform sequence given in S2 Table . ORF sizes were calculated to determine rank of ankyrin long exon size for the following genome drafts: Mus musculus , GRCm38/mm10 , ( ftp://hgdownload . cse . ucsc . edu/goldenPath/currentGenomes/Mus_musculus/bigZips/ ) ; Ciona intestinalis , Mar . 2005 freeze , ( ftp://hgdownload . cse . ucsc . edu/goldenPath/currentGenomes/Ciona_intestinalis/bigZips/ ) ; Drosophila melanogaster , Aug . 2014 ( BDGP Release 6 + ISO1 MT/dm6 ) ( ftp://hgdownload . cse . ucsc . edu/goldenPath/currentGenomes/Drosophila_melanogaster/bigZips/ ) ; and Caenorhabditis elegans , UCSC version ce6 ( ftp://hgdownload . cse . ucsc . edu/goldenPath/currentGenomes/Caenorhabditis_elegans/bigZips/ ) . ORFs were identified using a custom Python script , which assumed a minimum ORF size of 100 amino acids free of stop codons , and also filtered out regions that would be interrupted by long stretches of simple repeats . The script is publicly available at https://github . com/bishoyh/ORFeome-calculator . Statistics were calculated using Mathematica version 10 . 3 ( Wolfram , Champain , IL ) . Fluorescence recovery after photobleaching ( FRAP ) was performed using the scanning confocal microscopes listed above . For each set of experiments shown in a figure the same microscope and conditions were used for controls and experimental samples . Quantification of all movies was done in ImageJ using a plug-in designed to analyze fluorescence intensity in manually picked regions . Bleaching of the whole cell due to imaging over time was accounted for by dividing the bleached region by the unbleached region , and then by dividing the resulting ratio by the initial prebleach ratio to correct for any differences . The results were then normalized to obtain the percentage of fluorescence recovery after bleaching . For the experiments testing the location of the diffusion barrier along the axon , 5-micron segments were photo-bleached in regions 0–15 μm , 20–50 μm or 70–140 μm from the cell body . Quantification of these FRAP movies was done in ImageJ using a FRAP Calculator macro developed by Dr . Robert Bagnell . One issue for the two regions further from the cell body was that axons from the ddaE neuron that we used for our experiments were bundled together with axons from the ddaD in these areas . To get around any interference from the ddaD axon , this cell was killed 24h before the bleached experiments were performed . Cell killing was accomplished by aiming a MicroPoint pulsed UV laser ( Andor ) at the ddaD nucleus .
|
The axon initial segment ( AIS ) is currently thought to be a distinguishing feature of vertebrate neurons that adapts them for rapid , precise signaling . It serves as a hub for the regulation of neuronal excitability as the site of action potential initiation and also acts as the boundary between the highly-specialized axon and the rest of the cell . Here we show that the giant ankyrins that structurally organize the AIS , and were thought to be vertebrate-specific , instead have an ancient origin in a bilaterian ancestor . We further show the presence of a giant ankyrin-dependent AIS-like plasma membrane boundary between the axon and soma in a Drosophila sensory neuron . These results suggest that the cytoskeletal backbone for the AIS is not unique to vertebrates , but instead may be an evolutionarily conserved feature of bilaterian neurons .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"invertebrates",
"vertebrates",
"neuroscience",
"animals",
"invertebrate",
"genomics",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"genome",
"analysis",
"nerve",
"fibers",
"neuronal",
"dendrites",
"drosophila",
"research",
"and",
"analysis",
"methods",
"genomics",
"animal",
"cells",
"proteins",
"axons",
"insects",
"animal",
"genomics",
"arthropoda",
"biochemistry",
"cytoskeletal",
"proteins",
"cellular",
"neuroscience",
"cell",
"biology",
"ankyrins",
"neurons",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"computational",
"biology",
"gene",
"prediction",
"organisms"
] |
2016
|
Bilaterian Giant Ankyrins Have a Common Evolutionary Origin and Play a Conserved Role in Patterning the Axon Initial Segment
|
Accurate prediction of active sites is an important tool in bioinformatics . Here we present an improved structure based technique to expose active sites that is based on large changes of solvent accessibility accompanying normal mode dynamics . The technique which detects EXPOsure of active SITes through normal modEs is named EXPOSITE . The technique is trained using a small 133 enzyme dataset and tested using a large 845 enzyme dataset , both with known active site residues . EXPOSITE is also tested in a benchmark protein ligand dataset ( PLD ) comprising 48 proteins with and without bound ligands . EXPOSITE is shown to successfully locate the active site in most instances , and is found to be more accurate than other structure-based techniques . Interestingly , in several instances , the active site does not correspond to the largest pocket . EXPOSITE is advantageous due to its high precision and paves the way for structure based prediction of active site in enzymes .
Prediction of functional sites in proteins is essential for a range of bioinformatics applications such as molecular docking , and structure based drug design . Traditional methods for predicting functional sites include three approaches: 1 ) . The first approach uses sequence homology to find evolutionary conserved residues with functional activity . 2 ) . The second approach utilizes structural homology with other proteins of known function to locate functional regions . 3 ) . The third and last approach uses geometry and physico-chemical attributes of the protein structure and sequence to identify areas with functional activity . Over the years , several techniques based on the third approach have been developed . These techniques include LIGSITE [1] , POCKET [2] , POCKET-FINDER [3] , SURFNET [4] , CAST [5] , PASS [6] , Cavity Search [7] , VOIDOO [8] , APROPOS [9] , LigandFit [10] , 3DLigandSite [11] , MSPocket [12] , Fpocket [13] , McVol [14] , Ghecom [15] , PocketDepth [16] , PocketPicker [17] , VICE [18] , as well as consensus techniques which use a combination thereof such as MetaPocket [19] . Other methods analyze the protein surface for pockets [20 , 21] , cavities [22–24] , and channels [25] using pure geometric characteristics , and do not require any prior knowledge of the ligand or of sequence homology . Other computational techniques use geometric characteristics in combination with physico-chemical traits . Such methods include FOD [26] , and Elcock [27] that analyze the hydrophobicity distribution under the assertion that functionally important residues are often in electrostatically unfavorable positions . Similarly , THEMATICS [28] uses geometric characteristics in combination with theoretical microscopic titration analysis , while the methods of Goodford [29] and Rupert et al . [30] , and SiteHound [31] identify ligand binding sites based on analyses of the binding energies of probes placed on a grid around the protein . Another purely geometric method , EnSite , uses the proximity of catalytic residues to the molecular centroid to accurately detect the active sites of enzymes with high accuracy [32] . When used in combination with sequence and structure homology , geometric techniques are enhanced and prediction is improved . Some techniques use a vast combination of parameters ranging from conservation , residue type , accessibility , 2D structure propensity , cleft depth , B-factors , etc . to predict active site residues . Using such parameters , Gutteridge et al . predicted the location of active sites in enzymes using a neural network and spatial clustering [33] . Similarly Petrova et al . used Support Vector Machine with selected protein sequence and structural properties to predict catalytic residues [34] . In both cases , about 90% of the actual catalytic residues were correctly predicted . From these data it is clear , that one should rely on sequence and structure homology when possible , and over the past decade , multiple methods to detect binding sites and functional pockets based on geometric , structural , and genetic data were developed [35–39] . Several webservers of ligand binding sites have also been constructed and may be used to infer unknown ligand binding sites based on homology and other attributes such as Pocketome [40] , FunFold [41] , scPDB [42] , IBIS [43] , Multibind [44] , fPop [45] , and FINDSITE [46] . To date however , no comprehensive study comparing geometry based techniques has been performed . Normal-mode analysis is one of the standard techniques for studying long time dynamics and , in particular , low-frequency motions . In contrast to molecular dynamics , normal-mode analysis provides a very detailed description of the dynamics around a local energy minimum . Even with its limitations , such as the neglect of the solvent effect , the use of harmonic approximation of the potential energy function , and the lack of information about energy barriers and crossing events , normal modes have provided much useful insight into protein dynamics . Over the past years , several techniques have been described to calculate large-scale motions using simplified normal-mode analysis [47–51] . Based on these techniques , several executable programs to calculate normal modes have been released , such as ElNemo [52] , GROMACS [53] , and STAND [49] . Recently , several studies have drawn attention to the allosteric effect of ligand binding on normal modes dynamics [54] . From these studies , a clear correlation between binding in the native site and perturbation of normal modes was identified . The same allosteric effect of ligand binding on molecular dynamics was also pointed out by Bhinge [55] and Ming [56] which proceeded to use molecular dynamics simulations in predicting ligand binding sites . It is based on these recent advances , that we became aware of the capacity of normal modes in predicting active sites . In this paper we present a novel structure based technique using normal modes to predict the location of active sites in enzymes . The technique exploits the normal mode opening and closing motion of enzymes and the accompanied change of solvent accessibility and highlights residues of the active site . The idea behind the presented technique is that active sites pockets become exposed in normal mode dynamics ( Fig 1 ) . The hypothesis that active sites are surrounded by a shell of flexibility is not new and has been proposed in the dynamic lock-and-key model for biomolecular interactions . The shell of flexibility allows the enzyme to adapt to its ligand through an induced fit . The hypothesis was demonstrated in several studies notably by Weng et al . in a recent study on the flexibility of enzyme active sites [57] , and less recently by Babor et al [58] . The technique which detects EXPOsure of active SITes through normal modEs is named EXPOSITE . The technique may also be used in association with other methods to rank geometrically calculated pockets according to their solvent exposure . First , the prediction strength of EXPOSITE is trained extensively in a dataset containing 133 enzymes with known active sites from the Catalytic Site Atlas ( CSA ) database [59] . Then , EXPOSITE is tested in a dataset containing 845 enzymes and found to be more robust than other structure-based techniques . EXPOSITE’s high success rate is valuable for structure-based identification of active sites and clearly shows the added value of using normal modes for finding active sites . The technique does not attempt to withdraw from the importance of using genetic data , and clearly , a combination of both structural and genetic data would be more useful for predicting active sites than any of them on their own .
To assemble a training dataset containing 133 enzymes with known active sites , enzymes were selected from the CSA database [59] , version 2 . 2 . 1 . The dataset enzymes were selected according to the following two criteria: 1 ) . The enzyme active site is known from the literature ( LIT ) , and not derived by homology . 2 ) . The biologically active enzyme is composed of a single polypeptide chain and a single oligomerization state . To assemble a test dataset containing 845 enzymes , enzymes were selected from the CSA database [59] , version 2 . 2 . 1 . The test dataset was compiled by extracting chain A of all LIT entries that were not included in the 133 training dataset . These two datasets were used for training and testing EXPOSITEs prediction consistency respectively . To calculate normal modes of the dataset enzymes , two programs were utilized namely STAND [49] and ElNemo [52] and were run locally . For STAND , both real normal modes ( REA ) and Tirion modes ( TIR ) were calculated . For speed , the STAND option of coarse graining , 1 point ( 1 pt ) , which accelerates the calculations yet does not flaw the results was used , and defaults values of deformation amplitude were used . For ElNemo , default values of DQMIN -100 and DQMAX 100 were utilized . The DQMIN and DQMAX parameters correspond to the deformation amplitude in the direction of a single normal mode . For both STAND and ElNemo , deformation amplitudes were not scaled , and the same amplitude produces smaller deformation for larger molecules . For both STAND and ElNemo , only the 10 non-trivial lowest frequency modes were calculated . For each of these 10 modes , 40 PDB files were generated by STAND and 10 PDB files were generated by ElNemo all distorted along the particular mode . The two methods are very different in that STAND ( REA ) minimizes the structure and then calculates modes in φ and ψ torsion angle space whereas STAND ( TIR ) and ElNemo avoid minimization by using Tirion modes [50] and then calculate modes in Cartesian coordinate space . For STAND , the opposite extremes of the harmonic motion were empirically chosen as the 1st and 14th structure out of 40 respectively . At these extremes , the structures look fully “distorted” from each other . For ElNemo , the opposite extremes of the harmonic motion are the 1st and 10th structure out of 10 . To calculate the solvent accessible surface ( SAS ) area of amino acids in the generated PDB files , the DSSP program was used [60] . For each mode , SAS for each residue in the two structures at opposite extremes of the harmonic motion were calculated , and the absolute change of SAS between the extreme mode distortions , |ΔSAS| was used . To calculate pockets , LIGSITE [61] was run locally using default parameters . In each case , the 10 largest pockets were calculated and the pocket center as well as the pocket size were collected . The predicted active site was defined as the geometrical center ( centroid ) of the Cα coordinates of all residues with a solvent exposure |ΔSAS| , in the range 20-40Å2 The observed active site was defined as the geometrical center ( centroid ) of the Cα coordinates of the active site residues specified in the CSA database [59] . The predicted and observed active sites were represented each by a single coordinate in Cartesian space . The distance between these two coordinates was defined as the distance between the predicted and observed sites . The success of a prediction was based on the distance between the predicted and observed sites in the training and test datasets . If the distance between the predicted and observed sites was less than 12Å , then a prediction was considered successful . Conversely , if the distance was larger than 12Å , then a prediction was deemed incorrect . In the special case of the PLD dataset and for easy comparison with other techniques , a prediction was considered successful if any atom coordinate of the ligand was within 4Å of the predicted site . If no atom coordinate of the ligand was within 4Å of the predicted site , then the prediction was considered wrong . To compare EXPOSITE with other techniques , several software were run on all datasets namely , the training dataset of 133 enzymes , and the testing dataset of 845 enzymes , as well as a dataset containing 48 proteins derived from the PLD dataset [62] and engineered by Huang et al [61] . First , each of the following software was downloaded: LIGSITE , CAST , PASS , and SURFNET . For EnSite , no software was available , and the script was reconstructed based on the algorithm described in the original paper [32] . Then , each of the software was run locally on a PC running under Windows or Linux . In the case of the training and test datasets ( which lacks ligands ) , a prediction was considered successful if the predicted and observed active site were less than 12Å apart . In the case of the PLD dataset ( which contains ligands ) , a prediction was considered successful if the predicted active site was less than 4Å apart from any ligand atom .
To reliably assess the success rate of our technique in an sizeable ensemble , two datasets were assembled from the CSA database [59] . The CSA database contains 23 , 265 enzymes with known active sites . Of these , only 845 had an active site known from the literature ( LIT ) , and comprised the test dataset . Of these , only 133 were composed of a single chain that is biologically active as a monomer in a single oligomerization state , and comprised the training dataset . The PDB IDs of the 133 selected enzymes of the training dataset are listed in S1 Table . The PDB IDs of the 845 enzymes of the test dataset are listed in S2 Table . To test for homology within the datasets , the enzyme commission ( EC ) numbers were retrieved . Although , some homologues were found within a single dataset , no homologues were found between the training and test dataset . A number of programs were tested to calculate geometric pockets of biomolecular structures , i . e . POCKET [2] , LIGSITE [1] , POCKET-FINDER [3] , SURFNET [4] , CAST [5] , PASS [6] . The program LIGSITECSC [61] provides a list of pocket centers and size in a PDB format and was subsequently utilized in all our calculations . Surprisingly , there are significant differences between SAS of residues calculated by DSSP and other techniques such as ENCAD , CNS , and Accelrys . These differences arise from the different approaches used in calculating SAS . Nonetheless , when calculating the change of surface areas , ΔSAS , these differences cancel out and all programs produce comparable ΔSAS values . Biologically relevant modes are not always represented in the lowest frequency modes . Sampling more data , i . e . by calculating more modes could provide better results . Similarly , changing the |ΔSAS| thresholds could also lead to a higher success rate by allowing more exposure data to be included . To test this assertion and optimize the success rate of EXPOSITE the following parameters were varied: the threshold value of |ΔSAS| and the number of normal modes sampled . The number of modes sampled was varied from 0 to 10 and the |ΔSAS| minimum and maximum thresholds were changed from 0 to 60 Å2 . As seen in S4 Table , the optimal |ΔSAS| thresholds for ElNemo were around 20 and 40 Å2 respectively . Below the threshold of 10 Å2 , normal exposure fluctuations contribute little to EXPOSITE’s accuracy . Above the threshold of 40 Å2 , exposure changes arise from the normal mode tip effect ( bond breaking and exaggerated exposure ) and contribute little to the EXPOSITE accuracy . For STAND , the optimal |ΔSAS| threshold values were 20 and 40 Å2 respectively . This difference of |ΔSAS| thresholds between STAND and ElNemo is due to the fact that STAND uses coarse graining , inherently reducing the surface area , whereas ElNemo does not . STAND uses coarse graining and represents each amino acid with a single bead , while ElNemo uses a heavy atoms representation . In both cases , the maximum deformation amplitude were not chosen and default values were used . Also , the maximum deformation amplitude was not scaled in this study . The optimal number of mode sampling peaks to a plateau around modes 8 , 9 , and 10 for both STAND and ElNemo ( S5 Table ) . Below this sampling number important information is lost . Intriguingly , when using no threshold for |ΔSAS| , the accuracy of EXPOSITE is consistently 86% , no matter how many modes are sampled . EXPOSITE uses solvent accessibility changes in normal-modes to predict the location of active sites in enzymes . As seen in Fig 2 , residues experiencing large accessibility changes ( colored cyan and green ) are likely to be found in proximity to active site residue ( shown in text ) . In contrast , residues experiencing little exposure change ( colored blue ) are less likely to be found in vicinity of the active site . The proximity between residues experiencing large |ΔSAS| and the experimentally observed active site residues is an indicator of the precision of EXPOSITEs prediction . On average , the predicted and observed active sites in the training dataset are separated by 7 . 9 Å , and a standard deviation of 4 . 4 Å ( S1 Fig ) . The maximum success rate of EXPOSITE in the training dataset consisting of 133 enzymes was 92% . Curiously , in the training dataset , the binding pocket coincides mostly with the largest pocket ( 82% ) but not always ( 18% ) . This finding accounts for the pitfall of other techniques which rely on pocket size only for ranking . Also interesting is the fact that no active site was found in pockets with a size less than 7 Å3 . Such pockets are too small to accommodate ligands and validate our convention of discarding them as insignificant . Shown in Fig 3 is a histogram of distances between the predicted and observed active sites in the 845 enzyme test dataset . In this dataset , the predicted and observed catalytic sites are separated by an average of 9 . 2 Å , 11 . 5 Å , and 14 . 1 Å for EXPOSITE , ENSITE , and LIGSITE respectively ( Fig 3 ) . Significantly , if a successful prediction is arbitrarily defined by a distance cutoff of 4 Å , then the number of hits of EXPOSITE ( 16 . 1% ) is almost double that of ENSITE ( 8 . 7% ) . Similarly , if a successful prediction is arbitrarily defined by a distance cutoff of 3 Å , then the number of hits of EXPOSITE ( 10 . 4% ) is 2 . 4 times that of ENSITE ( 4 . 3% ) . To test the robustness of EXPOSITE , we tested its success rate in a dataset containing 845 enzymes ( S2 Table ) . Not surprisingly , the success rate is much lower than in the 133 enzyme dataset . Reliably however , EXPOSITE is better that EnSite in predicting the active site by >2% . The sharp decrease of prediction success rate in the 845 enzymes dataset is not surprising , as the dataset does not discriminate between real homomonomeric enzymes with high success rates , and homomultimeric enzyme assemblies with low success rates ( close to 0 ) . Even if statistically robust , the large 845 enzyme dataset does not reflect the real success-rate of prediction techniques , and the smaller 133 enzyme dataset should be regarded as a more representative alternative . The large 845 enzymes dataset is too diverse , and demonstrates the difficulty in assembling representative datasets . EXPOSITE highlights the binding site of proteins of the Protein Ligand Dataset ( PLD ) published elsewhere [62] . Shown in Fig 4 ( and in S2 Fig ) are a few examples of ligand binding site prediction . Residues experiencing large accessibility changes ( colored green ) are likely to be found in proximity to the ligand ( colored red ) , whereas residues experiencing little exposure change ( colored blue ) are further away . The proximity of residues with large accessibility changes and residues of the observed active site is a success indicator of EXPOSITEs predictions . On average , the predicted and observed centers in the protein PLD dataset are separated by 7 Å with a standard deviation of 3 . 3 Å . Intriguingly , the separation in the PLD dataset is smaller than that of the CSA dataset by almost 1 Å , and it is probably a flaw due to the handpicked nature of the PLD dataset . To accurately and robustly compare EXPOSITE with other techniques , all other software were run on all datasets namely the training dataset of 133 enzymes , the testing dataset of 845 enzymes . A prediction was considered accurate if the distance between the predicted and observed sites was less than 12Å . If the distance was larger than 12Å , then a prediction was considered inaccurate . The calculated prediction accuracies are listed in Table 1 . When compared to other geometric techniques EXPOSITE is advantageous due to its high success rate . As seen in Table 1 , EXPOSITE is only slightly better than EnSite at predicting active sites and EnSite is still superior to EXPOSITE in speed as it is ingenious in simplicity . Also note that prediction of binding sites in unbound proteins is less successful than that of ligand-bound proteins simply because the ligands occupy and expose the binding site through induced fit thereby easing its identification . To accurately and robustly compare EXPOSITE with other techniques , all other software were run on the bound and unbound PLD dataset [61] . A prediction was considered accurate if any ligand atom was within 4Å of the predicted site . If no ligand atom was within 4Å of the predicted site , then the prediction was considered inaccurate . The calculated prediction accuracies are listed in Table 2 . The data for EXPOSITE and Ensite is reported by us , the data for VICE was reported by Tripathi et al [18] , the data for Fpocket was reported by Le Guilloux et al . [13] , the data for PocketPicker was reported by Weisel et al . [17] , the data for LIGSITEcs , CAST , PASS and SURFNET were first reported by Huang et al . [63] . Please note that EXPOSITE is not always successful , such as in the case of PDB 1igj , 3gch , 3mth , and 2tmn as may be seen in Fig 5 . Intriguingly , the classically accepted metric for binding site prediction is 4Å , and we used this metric in the classical PLD dataset when comparing the classical performance of EXPOSITE , Ensite , VICE , Fpocket , PocketPicker , LIGSITEcs , CAST , PASS and SURFNET ( Table 2 ) . However , in the unclassical training and test datasets which were never tested before , we relied on an unclassical distance of 12Å . The training and test datasets contain 20 times more proteins than the hand-picked PLD dataset , and if the classical distance of 4Å was used , then the performance of all techniques sank drastically . To maintain good performances for all techniques in the training and test datasets , the classically accepted metric for binding site prediction was raised to an unclassical 12Å . Generally speaking , the success rate in the handpicked PLD dataset is higher than in the non-handpicked 845 test dataset . This discrepancy suggests that the PLD dataset was not randomly picked , and could artificially increase prediction success rates . EXPOSITE’s feature , of highlighting active sites is very useful for ranking pockets . Indeed , the technique is capable of ranking enzyme pockets according to their degree of exposure in normal mode dynamics . This ranking enables EXPOSITE to choose the correct binding pockets from a list of pockets calculated by LIGSITE . The assumption that the active site pockets is usually in the largest pocket [1 , 4 , 64] is being used by several pocket detection programs and the top site is generally the largest one . However , this assumption is not always true and in several instances , the active site corresponds to the second , third , or fourth largest pocket .
The rationale behind the success rate of EXPOSITE is fairly simple . For proper enzyme activity , protection from the surrounding water is often necessary as shown by normal modes closure of the active site . Proteins in general and enzymes in particular often act as environment protectors . They envelop substrates to catalyze chemical reaction that would otherwise not take place in aqueous solution . They conceal prosthetic groups to coordinate binding thus increasing affinity which is negligible in water . They act as small shielding cases displaying alternating motions of opening and closing to allow ligand entrance and protection respectively . Throughout this motion , protein residues located at various distances from the active site are exposed to the solvent to a different degree . Residues in proximity to the active site are exposed more than those faraway . This idea lays down the foundation for EXPOSITE suggesting the pocket closest to the maximum exposure center is the active site . The change in solvent accessibility between the X-ray structure and the largest deformation of either of the normal mode extremes could also have been used . However , the maximum effect of motion is observed between the two extremes which vibrate around the X-ray structure corresponding to a local minimum . EXPOSITE takes into account several parameters such as accessibility change in normal modes , centroid distance from pockets , as well as pocket size . Normal modes by their own virtue take into account more parameters such as contact network and distances . Together , these parameters resemble those used in neural network techniques [33 , 34] where they are analogous to accessibility , cleft depth , B-factors , etc… As much as these techniques seem different , the analogy between the parameters is astounding . The success rate is not affected by the different types of normal modes techniques , STAND and ElNemo . The success rate remains unchanged even when STAND and ElNemo are used in different combinations with accessibility calculators ( i . e . ElNemo with ENCAD accessibility calculator [65] . The success rate does not originate from the difference in the atomic representation used by ElNemo and STAND . In fact , when running STAND in full-atom representation the success rate remains unchanged . These data indicate that coarse graining which ignores the amino acid type and accessible surface does not influence the success rate of EXPOSITE . In fact , adding heavy atoms to the PDB files generated by STAND also does not decrease the success rate of EXPOSITE . We conclude that coarse-graining and accessibility calculation methods do not affect the success rate of EXPOSITE . Care should be taken when using our technique on structures composed of several domains . Practical interpretation of normal modes of multi-domain structures tend to be problematic in the sense that bending and twisting of one domain relative to another tend to overshadow modes with biological meaning . One way to circumvent this problem is to run normal modes of single domains to predict its active site . We excluded multi chain enzymes which are biologically active in oligomeric states from our CSA dataset . Similarly , care should be taken when using EXPSOITE on structures with elongated termini or exceedingly flexible loops . Such structures often present odd normal modes around these areas which tend to overshadow modes with biological meaning . Some strongly recommended ways to circumvent the problem of exaggerated motion is simply to clip out ( or edit out ) the stretches and rerun normal mode computation or to set an upper value for the cutoff of |ΔSAS| of 75 Å2 when calculating modes with ElNemo ( 40 Å2 for STAND ) . The cutoff should minimize the effect of loose and flexible termini with exaggerated exposure change . A complete list of success and failures is provided in S6 and S7 Tables . A distinction should be made between the concepts “binding site” and “active site” . Usually , an active site is found in a single copy in an enzyme , while binding sites may be present in multiple copies in proteins . Thus , prediction of active sites and ligand binding sites are very different , and whereas only one prediction is correct for enzymes , several predictions are correct for proteins . To complicate things further , some enzymes are composed of multiple chains , each equipped with a distinct active site , and so much care should be taken so as not to over interpret a prediction . As a rule of thumb structure based predictions ( EXPOSITE , EnSite , etc ) are more accurate in single chain polypeptide enzymes . In an attempt to correlate between pocket size and active site , the following parameters of active site were calculated in the PLD dataset: 1 ) . The number of Cα atoms of the active site was derived from the CSA database . 2 ) . The number of heavy atoms in the substrate was calculated from the PLD database . 3 ) . The number of residues of with high accessibility change was calculated from EXPOSITE . 4 ) . The size of the predicted pocket in Å3 was from LIGSITE . These parameters all reflect on the size of the active site yet there is no obvious correlation among them . There was no correlation ( R = 0 . 12 ) between pocket size and the number of active site residues . This is partially due to fractionation of active sites into adjacent pocket ( POK ) which decrease “real” active site size . This fractioning of active sites is a problem often encountered in pocket calculating programs . Adjoining sizes of vicinal pockets did not improve the correlation significantly .
Over the past years normal modes have enjoyed a revival . In this article , the biological relevance of normal modes is illustrated in a new technique . The presented technique exposes active sites of enzymes with high success rates . As pocket detection methodologies normal mode techniques improve so will our technique . In the future , EXPSOITE is expected to become publicly available as a basic tool ( website and/or program ) for predicting active sites of enzymes . The Perl code used in this study is freely available in the supplementary data . Note that DSSP , LIGSITE , ElNemo , and/or STAND must be obtained from third parties , and that the time bottleneck of the method is normal mode calculation .
|
In this paper , we present an improved technique to predict active sites in enzymes . Our technique is based on changes of solvent accessibility that accompany normal mode dynamics . We assert the technique strength using several enzyme datasets with known catalytic residues . We show the technique successfully locates the active site in most cases , and consistently surpasses the accuracy of other techniques . We show how the technique is advantageous and paves the way for high precision prediction of active sites .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Conclusion"
] |
[
"chemical",
"characterization",
"classical",
"mechanics",
"molecular",
"dynamics",
"enzymes",
"enzymology",
"protein",
"structure",
"prediction",
"damage",
"mechanics",
"protein",
"structure",
"research",
"and",
"analysis",
"methods",
"bioinformatics",
"proteins",
"deformation",
"chemistry",
"binding",
"analysis",
"molecular",
"biology",
"physics",
"biochemistry",
"enzyme",
"structure",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"chemistry",
"macromolecular",
"structure",
"analysis"
] |
2016
|
Normal Modes Expose Active Sites in Enzymes
|
Hepatitis E virus ( HEV ) is a positive-strand RNA virus encoding 3 open reading frames ( ORF ) . HEV ORF3 protein is a small , hitherto poorly characterized protein involved in viral particle secretion and possibly other functions . Here , we show that HEV ORF3 protein forms membrane-associated oligomers . Immunoblot analyses of ORF3 protein expressed in cell-free vs . cellular systems suggested a posttranslational modification . Further analyses revealed that HEV ORF3 protein is palmitoylated at cysteine residues in its N-terminal region , as corroborated by 3H-palmitate labeling , the investigation of cysteine-to-alanine substitution mutants and treatment with the palmitoylation inhibitor 2-bromopalmitate ( 2-BP ) . Abrogation of palmitoylation by site-directed mutagenesis or 2-BP treatment altered the subcellular localization of ORF3 protein , reduced the stability of the protein and strongly impaired the secretion of infectious particles . Moreover , selective membrane permeabilization coupled with immunofluorescence microscopy revealed that HEV ORF3 protein is entirely exposed to the cytosolic side of the membrane , allowing to propose a model for its membrane topology and interactions required in the viral life cycle . In conclusion , palmitoylation determines the subcellular localization , membrane topology and function of HEV ORF3 protein in the HEV life cycle .
Hepatitis E virus ( HEV ) infection is believed to be the most common cause of acute hepatitis and jaundice in the world [1 , 2] . It is a positive-strand RNA virus classified in the Hepeviridae family [3 , 4] . Most human pathogenic strains can be assigned to genotypes 1–4 within the species Orthohepevirus A [5] . HEV genotypes 1 and 2 are transmitted by the fecal-oral route and can cause large , primarily waterborne outbreaks in low-income countries with poor sanitation , infecting about 20 million people and claiming 70 , 000 lives every year [6] . On the other hand , infection with HEV genotypes 3 and 4 has been recognized as a porcine zoonosis in high-income countries , with much higher than anticipated seroprevalence rates , reaching 86 . 4% in the South of France [7] . HEV genotype 3 ( and , to a lesser extent , genotypes 4 as well as 7 ) can persist in immunocompromised patients and cause chronic hepatitis with potential rapid progression to cirrhosis [8 , 9] . In addition , HEV genotype 3 is a cause of diverse neurological manifestations , especially neuralgic amyotrophy ( also known as Parsonage-Turner syndrome ) [10 , 11] . Hence , HEV represents a growing global health concern [12–14] . HEV is a nonenveloped virus of 27–34 nm in diameter . The 7 . 2-kb RNA genome encodes three open reading frames ( ORF ) which are translated into ( i ) the ORF1 polyprotein , representing the viral replicase , ( ii ) the ORF2 protein , corresponding to the viral capsid , and ( iii ) the ORF3 protein , a small , hitherto poorly characterized protein [3] . Although HEV is found as a nonenveloped virus in bile and feces , it circulates in the bloodstream , similar to hepatitis A virus , as a 'quasi-enveloped' particle wrapped in cellular membranes likely derived from exosomes [15–17] . HEV ORF3 protein plays an essential role in virion secretion [18 , 19] but not in RNA replication or virion assembly [20] . More specifically , ORF3 protein supports HEV egress using the exosomal pathway [16 , 21] and has been shown to be associated with the quasi-enveloped viral particle [22] . Interestingly , a recent study reported that HEV ORF3 protein possesses ion channel activity required for the release of infectious virus [23] . ORF3 encodes , depending on the genotype , a 112- , 113- or 114-amino-acid ( aa ) protein with a predicted molecular weight ( MW ) of 12–13 kDa . It has been reported to be phosphorylated at Ser 71 in genotype 1 ( corresponding to Ser 70 in genotype 3 ) by the cellular mitogen-activated protein kinase [24] . A yeast two-hybrid study suggested oligomerization of ORF3 protein through a proline-rich C-terminal region [25] , and homotypic interaction of ORF3 protein was recently confirmed in mammalian cells [23] . Moreover , phosphorylated ORF3 protein was reported to interact with the ORF2 ( capsid ) protein [26] . Early reports indicated that ORF3 protein associates , via an N-terminal hydrophobic domain , with the cytoskeleton and , more specifically , with microtubules [24 , 27] . In addition , the protein has also been observed at early and recycling endosomes [28] as well as multivesicular bodies ( MVBs ) [16] . Notably , a conserved PSAP motif may be necessary for its interaction with tumor susceptibility gene 101 ( Tsg101 ) , a component of the endosomal sorting complexes required for transport ( ESCRT ) pathway [19 , 21 , 29] . In the present study , we describe the oligomerization as well as the association of HEV ORF3 protein with intracellular and plasma membranes . We identified a determinant for oligomerization at the N-terminal end of ORF3 protein . Furthermore , we demonstrate palmitoylation of ORF3 protein on cysteine residues in its N-terminal segment . Alanine substitution of the cysteine residues or inhibition of palmitoylation by 2-bromopalmitate ( 2-BP ) lowered the MW of the viral protein and altered ORF3 protein subcellular localization . Importantly , serine substitution of 5 cysteine residues in a full-length HEV clone strongly impaired the secretion of infectious particles . Finally , selective membrane permeabilization coupled with immunofluorescence analyses allowed to determine the membrane topology of HEV ORF3 protein .
The N-terminal region of ORF3 protein has previously been reported to harbor two hydrophobic segments of which the second has been predicted as a transmembrane segment [23] . Sequence analysis of HEV ORF3 from all 8 genotypes revealed that the protein sequence is highly conserved especially in the N-terminal part ( aa 1–28 ) ( Fig 1A ) . Of note , the predicted transmembrane segment , boxed in grey in Fig 1A ( segment aa 30–53 for gt 3 ) , shows more variability in primary sequence , especially in gt 4 as compared to the other genotypes , and , intriguingly , is not predicted to form an α-helix . HEV ORF3 protein has been reported to oligomerize in a yeast two-hybrid study as well as in mammalian cells [23 , 25] . However , the determinants for oligomerization need to be further delineated . As shown in Fig 1B , we confirmed oligomerization of ORF3 protein by coimmunoprecipitation of HA- and FLAG-tagged ORF3 constructs in transiently transfected U-2 OS cells . On this basis , we developed a fluorescence resonance energy transfer ( FRET ) assay to further explore oligomerization of ORF3 . A CFP-YFP fusion protein and cotransfection of CFP and YFP served as positive and negative controls , respectively . As shown in Fig 1C , coexpression of CFP- and YFP-tagged full-length ORF3 protein ( ORF31-113 ) constructs yielded a significant FRET signal as compared to the negative and positive controls . Moreover , analyses of C- and N-terminal deletion constructs revealed that the N-terminal 28 aa are essential for the oligomerization of HEV ORF3 protein . ORF3 expressed as a GFP fusion protein or as an untagged protein either individually or in cells replicating full-length infectious HEV is consistently observed at intracellular membrane compartments , likely corresponding to the previously described localization in endosomes and in MVBs [16 , 28] , but also at the plasma membrane ( Fig 2A ) . Membrane extraction with 1 M NaCl , sodium carbonate pH 11 . 5 , or 1% Triton X-100 , followed by flotation assay confirmed that ORF3 protein is tightly associated with cellular membranes ( Fig 2B ) , as indicated by a similar behaviour of ORF3 protein than as compared to the integral endoplasmic reticulum membrane protein CLIMP63 . To determine the domain responsible for membrane association of HEV ORF3 protein , we analyzed GFP-fused deletion constructs by membrane flotation assay . The endoplasmic reticulum-resident integral membrane protein CLIMP63 and GFP alone were used as reference for membrane-associated and soluble proteins , respectively ( Fig 3A ) . First , we confirmed that ORF31-113-GFP is associated with membranes , similarly to ORF3 protein alone ( cf . Fig 2B ) . As shown in Fig 3A , the N-terminal 28 aa ( ORF31-28-GFP ) were sufficient to mediate membrane association of ORF3 protein . Deletion of the N-terminal 27 aa ( ORF328-113-GFP ) or 52 aa ( ORF353-113-GFP ) completely abolished membrane association . Hence , the N-terminal 28 aa are required for proper membrane association of HEV ORF3 protein . At the subcellular level , fluorescence microscopy showed that ORF31-28-GFP had a very similar localization than ORF31-113-GFP ( Fig 3B ) while ORF328-113-GFP and ORF353-113-GFP were displayed predominantly a diffuse pattern . These results show that in addition to the oligomerization of ORF3 protein , the N-terminal region and more particularly aa 1–28 determine the membrane association and subcellular localization of ORF3 protein . The 113-aa ORF3 protein of HEV genotype 3 has a theoretical molecular weight ( MW ) of 11 . 3 kDa . However , separation of lysates from pCMVORF3-transfected mammalian cells by 17% SDS-PAGE followed by immunoblot analysis revealed an apparent MW of about 15 kDa ( Fig 4A ) . Interestingly , ORF3 protein produced in wheat germ-based cell-free expression system migrated at the expected theoritical MW of about 11–12 kDa ( Fig 4A ) . This observation suggests that ORF3 may undergo posttranslational modification when expressed in cells . As it has been postulated that ORF3 may be phosphorylated , we prepared an alanine substitution mutant ( S70A ) to determine whether phosphorylation reported at this particular serine residue was involved in the observed MW change . However , as the pattern obtained for this mutant was similar to that of wild-type ( wt ) ORF3 protein , phosphorylation on Ser 70 does not explain the higher apparent MW observed in our experimental settings ( Fig 4A ) . Bands other than the one at 15 kDa were observed at lower MWs , including one with a similar migration as compared to cell-free expressed ORF3 protein , i . e . likely the non-modified form . Hence , we validated our observations in lysates from cells replicating infectious HEV from two different molecular clones , i . e 83–2 and p6 . As shown in Fig 4B , a single band for ORF3 protein similar to the higher MW signal obtained with the single ORF3 expression was observed under these conditions . Altogether these data show that HEV ORF3 protein has a higher apparent MW than the theoretical , which is likely the consequence of a posttranslational modification . Because an ORF3 protein construct deleted from the first 18 aa displayed a MW which is close to the expected theoretical MW ( S1B Fig ) , we focused our attention on the N-terminal region of ORF3 . Interestingly , the first N-terminal 21 aa of the ORF3 protein comprise 8 cysteine residues at conserved positions in all HEV genotypes with the exception of HEV genotypes 4 and 6 which have only 7 cysteines ( Fig 1A ) . Hence , we next substituted the cysteine residues by alanines and analyzed the resulting constructs by immunoblot . Single alanine substitutions resulted in a slight MW shift for all cysteine residues ( S1E Fig ) . Mutants with grouped alanine substitutions , i . e . C1-4 , C5-8 and C1-8 , showed a more pronounced effect on the MW of ORF3 protein ( Fig 4C ) . Indeed , mutants C1-4 and C5-8 showed a lower apparent MW and also a reduced protein amount for the latter . Strikingly , alanine substitution of all 8 cysteine residues ( mutant C1-8 ) destabilized the protein which could no longer be detected by immunoblot ( Fig 4C ) . Stability of these mutant proteins could be increased by fusion with GFP , as shown in S1F Fig . As shown in Fig 5D , the subcellular localization of ORF3 protein changed upon alanine substitution of the cysteines , i . e . mutants C1-4 , C5-8 or C1-8 ( Fig 4D ) , and showed a pattern similar to that of ORF328-113-GFP when all cysteines are mutated ( Fig 3B ) . Furthermore , this observation is correlated with a loss of membrane association of the mutants as assessed by membrane flotation ( Fig 4E ) . Together , these findings indicate that the cysteine residues of ORF3 protein are likely engaged in a posttranslational modification which plays an important role for the subcellular localization , membrane association and stability of the protein . Among common posttranslational modifications involving cysteine residues , palmitoylation results in the covalent link of a palmitate , i . e . a C16 acyl chain . A consequence of the addition of palmitate to cysteine residues is the association with cellular membranes . This modification is mediated by host palmitoyltransferases . To assess whether ORF3 protein is palmitoylated , cells replicating the infectious HEV clones p6 and 83–2 as well as cells transfected with wt , C1-4 and C5-8 ORF3 expression constructs were incubated for 3 h with 3H-palmitate before harvesting and immunoprecipitation with anti-ORF3 pAb . Immunoblot analysis revealed that comparable amounts of ORF3 protein were specifically immunoprecipitated in the different conditions ( Fig 5 , top panel ) . Furthermore , the MW observed for the C1-4 and C5-8 mutants was lower as compared to the wt ORF3 protein , as expected and previously observed ( Fig 4C ) . 3H-palmitate incorporation was then assessed by separation of protein lysates under the same conditions , followed by autoradiography . As shown in Fig 5 , ORF3 protein incorporated palmitate , as revealed by a strong signal for cells transfected with pCMVORF3 from both gt1 and gt3 . Importantly , palmitate incorporation into ORF3 protein is also observed in lysates from cells harboring the infectious HEV clones , albeit at a much lower intensity , reflecting a low activity of translation of ORF3 during the limited 3-h period of 3H-palmitate incubation ( Fig 5 , lower panel ) . Of note , a radioactive signal is also recorded for the for C1-4 and C5-8 mutants , indicating that , although harboring 4 alanine substitutions each , they still incorporate palmitate , likely at the 4 remaining cysteine residues . Altogether , these observations demonstrate that ORF3 protein is palmitoylated when expressed as a single protein as well as , more importantly , in the context of the replication of full-length infectious HEV . We next treated ORF3-expressing cells with increasing concentrations of the palmitoylation inhibitor 2-BP and analyzed the protein lysates for ORF3 protein by immunoblot ( S3 Fig ) . Protein analysis of cells transfected with the pCMVORF3 construct and cultured with increasing concentrations of 2-BP ( 0 . 5 , 5 , 25 and 50 μM ) revealed a dose-dependent accumulation of the lower MW signal corresponding to the non-modified ORF3 protein while the higher MW signal was decreasing in parallel ( S3 Fig ) . To determine the role of ORF3 palmitoylation in its subcellular localization , S10-3 cells transfected with GFP fusion constructs were treated with 25 μM 2-BP or DMSO as control for 24 h . As shown in Fig 6 , treatment with the palmitoylation inhibitor induced a relocalization of ORF3-GFP within the cytoplasm as a diffuse signal ( Fig 6 ) . Furthermore , cells expressing the minimal membrane-associated construct ORF31-28-GFP showed a similar diffuse pattern of the GFP signal within the cytoplasm ( Fig 6 ) . To investigate the role of palmitoylation in the function of HEV ORF3 protein , 5 cysteine residues were replaced by serine in the full-length HEV p6 infectious clone , yielding mutant p6_C5S ( Fig 7A ) . Because ORF2 and ORF3 are overlapping , the cysteine-rich coding region tolerates minimal sequence changes . Therefore , aa at positions 11 , 13 , 16 , 18 and 21 were substituted by a serine residue . Similarly to what was observed for the C1-8 mutant , where the 8 cysteine residues have been substituted in a heterologous expression setting , the ORF3_C5S protein was not detectable by immunoblot ( Fig 7B ) and the corresponding GFP fusion construct showed a diffuse subcellular distribution ( Fig 7C ) . Transfection of S10-3 cells with either p6 or p6_C5S RNA resulted in the immunofluorescence detection of similar amounts of ORF2 protein while ORF3 protein was detectable only for the wt ( Fig 7D ) . This was also confirmed by immunoblot analysis ( S5B Fig ) . Functional consequences were assessed by the determination of the infectivity in the intracellular and extracellular compartments 5 d post-transfection for the p6 wt and p6_C5S constructs . While a similar infectivity was observed intracellularly for both constructs , infectious virus secretion was strongly impaired in the p6_C5S mutant ( Fig 7E ) . These results indicate that the cysteine residues of HEV ORF3 protein engaged in palmitoylation are essential for the secretion of infectious particles . Palmitoylation occurs on the cytosolic side of the membrane at different subcellular sites depending on the palmitoyltransferase ( s ) involved . Since the cysteine residues of HEV ORF3 protein are present in the N-terminal segment of the protein , we hypothesized that the N terminus is exposed to the cytosol . To further validate this observation and to position the C-terminal end , we transfected a construct tagged at the N-terminal end with a FLAG epitope and at the C-terminal end with an HA epitope . The tags were detected by immunofluorescence by double staining in the absence of cell permeabilization vs . after treatment with 0 . 5% saponin . As shown in Fig 8A , both epitopes were detectable only after permeabilization of the plasma membrane with saponin . Hence , as suggested by the presence of the palmitoylated cysteines , the N-terminal region of ORF3 protein is exposed to the cytosolic side . Furthermore , these results also suggest that the C-terminal end of the protein is present on the cytosolic side and not exposed to the extracellular space , as it would be the case if HEV ORF3 possessed a transmembrane passage . Given that the addition of epitope tags may perturb the membrane association of a small protein like HEV ORF3 , a similar experiment was performed with the expression of an untagged ORF3 construct ( Fig 8B ) . The detection of the cell surface-expressed tetraspanin CD151 and of cytosolic mitochondrial antiviral-signaling protein ( MAVS ) have been employed as controls in both conditions , i . e non permeabilized and saponin 0 . 5% treated cells . Indeed , the cytosolic protein , i . e . MAVS , could be detected only in permeabilized conditions while the surface exposed protein , i . e . CD151 , is always accessible to antibody . Thus , we used monoclonal antibody ( mAb ) MRB198 recognizing an epitope of ORF3 protein in the C-terminal region aa 62–113 to investigate the topology of the viral protein . The ORF3 protein was well detectable at intracellular membranes and , to some extend , at the plasma membrane , by the anti-ORF3 mAb following permeabilization . However , HEV ORF3 protein was not detectable in the absence of permeabilization ( Fig 8B ) . These results further demonstrate that the N- and C-terminal regions of ORF3 protein are both present on the cytosolic side of the plasma membrane . To corroborate our findings , we also employed selective permeabilization combined with immunofluorescence detection of HEV ORF3 using the specific mAb described above . As described earlier [30] , transfection of plasmids allowing the expression of the membrane-associated hepatitis C virus ( HCV ) proteins core and E1 followed by immunofluorescence detection was used to assess the quality of the selective permeabilization . Indeed , differential digitonin-based permeabilization of cells allows to discriminate between epitopes exposed to the cytosolic vs . luminal side of intracellular membranes . Accordingly , we observed the detection of the cytosolically-oriented HCV core protein under total and selective permeabilization conditions , i . e . digitonin 0 . 2% and 0 . 01% , respectively ( Fig 8C ) . However , the luminal HCV E1 glycoprotein was detectable only following total permeabilization . Detection of HEV ORF3 protein was carried out with recombinant mAb MRB198 under both permeabilization conditions with a similar pattern of subcellular localization , i . e . at the plasma and intracellular membranes ( Fig 8C ) . The lack of difference between the detection of ORF3 protein under the two permeabilization conditions indicates that the epitope recognized by the mAb is oriented toward the cytosol . Importantly , given that these data were obtained under ORF3 single expression settings , we confirmed our findings using the same approach in S10-3 cells replicating the full-length HEV infectious clone p6 for which fluorescence intensity was measured and plotted for the different conditions ( Fig 8C ) . Altogether , our results indicate that HEV ORF3 protein is exposed to the cytosolic side of the plasma membrane as well as of the intracellular membranes . Thus , our data indicate that HEV ORF3 protein is membrane-associated via palmitoylation of the N-terminal segment and does not traverse the membrane .
In the present study , we identified the N-terminal region of HEV ORF3 protein as the determinant for membrane association and oligomerization . More precisely , we found that ORF3 protein is palmitoylated at the N-terminal cysteine-rich segment . Palmitoylation of HEV ORF3 protein determines its membrane association , subcellular localization and likely stabilizes the viral protein . Moreover , the mutation of cysteine residues of ORF3 protein engaged in palmitoylation strongly impairs the secretion of infectious HEV . Additionally , investigation of the membrane topology of HEV ORF3 protein by selective permeabilization coupled with immunofluorescence detection supports a model where the entire ORF3 protein is oriented toward the cytosolic side of the membrane . We have found that ORF3 protein is membrane-associated and localized to intracellular membranes as well as the plasma membrane . While the latter localization is not predominant in HEV-replicating S10-3 cells , ORF3 protein accumulates more importantly at this compartment when expressed alone in U-2 OS or Hep293TT ( S1D Fig ) . Inhibition by 2-BP revealed that subcellular localization of ORF3 protein is determined by palmitoylation . Similarly , it has been shown that the NB glycoprotein of influenza B virus is palmitoylated and that this posttranslational modification determines the subcellular localization of the viral protein to the plasma membrane [31] . Emerson and colleagues reported an accumulation of ORF3 protein at the apical membrane of C25j cells , a subclone of the human intestinal cell line Caco-2 which was associated with a preferential release of infectious virus at this apical membrane [19] . Furthermore , it is interesting to note that ORF3 protein is present at bile canaliculi of humanized liver chimeric mice infected with HEV [32 , 33] . Therefore , these observations are in agreement with a plasma membrane and bile canalicular localization of the ORF3 protein . Oligomerization of HEV ORF3 protein had been previously investigated in yeast 2-hybrid system and revealed that determinants for self-interaction reside in aa segment 71–113 [25] . A recent report showed oligomerization of ORF3 protein in mammalian cells by co-immunoprecipitation [23] . In our study , we confirm oligomerization of ORF3 protein by co-immunoprecipitation as well as FRET and we map the interacting domain to the region spanning aa 1 to 53 . However , our results cannot exclude a possible interaction domain in segment aa 71–113 as suggested by Tyagi and colleagues [25] . The N-terminal segment comprises the palmitoylated cysteine-rich domain as well as the following hydrophobic domain aa 28–53 . Palmitoylation is known to regulate protein-protein interactions [34] . Therefore , it is likely that the C16 acyl chains linked to the cysteine residues are important contributors of ORF3 oligomerization , possibly together with the C-terminal region of the protein . Indeed , similarly to observations made with the oligomeric HCV nonstructural protein 4B [35] , protein oligomerization may depend on several interacting domains . While ORF3 is often referred to as a 13-kDa protein , reports in the literature showed ORF3 protein with higher apparent MW at about 15 kDa in infected cells [21 , 22 , 36] as well as in vivo in liver chimeric mice infected with HEV [32] . Discrimination of the MW of HEV ORF3 protein required optimal separation by high percentage SDS-PAGE . Of note , we have included the ORF3S70A mutant in our analyses to examine the potential contribution of the phosphorylation described previously [24] . However , our data shows that alanine substitution at aa position 70 does not change the migration pattern of ORF3 protein in SDS-PAGE . Thus , our analyses did not allow to state whether HEV ORF3 protein is phosphorylated in addition to its palmitoylation . While the HEV ORF3 sequence of Orthohepeviruses A , including HEV gt 1 and gt 3 , is not well conserved as compared to the one of Orthohepeviruses B or C , i . e . avian or rat HEV , the presence of cysteine residues in the N-terminal region is conserved . Interestingly , avian HEV ORF3 protein , which possesses 10 cysteine residues among its N-terminal 30 aa ( S4A Fig ) , displays a similar subcellular localization as compared to HEV genotype 3 ORF3 protein ( S4B Fig ) . Moreover , avian HEV ORF3 protein , which displays an apparent molecular weight above the theoretical one ( 15 kDa vs . 11–12 kDa ) , can incorporate 3H-palmitate ( S4C Fig ) when expressed in human cells . These observations strongly suggest that palmitoylation of the HEV ORF3 protein is a conserved feature among the Orthohepeviruses . A recent report by Ding et al . suggested that ORF3 protein possesses an ion channel activity required for virus secretion [23] . Such function implies that ORF3 protein has a transmembrane segment to form a pore upon oligomerization . As shown here , the protein sequence of HEV ORF3 is highly conserved in the N-terminal region and more specifically in region aa 1–28 . A transmembrane segment is predicted by some algorithms ( i . e . PHDHTM , TMpred , TMHMM ) in region aa 31–52 , however with less consistency for HEV ORF3 protein of genotype 4 . In addition , secondary structure prediction for an alpha-helix is poor ( Fig 1A ) . Furthermore , we experimentally observed that ORF3 expression constructs where the predicted transmembrane segment ( aa 28–53 ) alone or the C-terminal part ( aa 28–113 ) is fused to GFP were found to be mainly soluble in transfected cells ( Fig 3 ) . As it is difficult to prove the existence of an internal transmembrane segment when expressed out of its natural protein context , we employed an antibody accessibility assay using selective membrane permeabilization and immunofluorescence with a mAb against the C-terminal part ( epitope aa 62–113 ) of ORF3 protein . Our data demonstrate that the epitope present in the C-terminal part of HEV ORF3 protein is detectable upon selective permeabilization of the plasma membrane , indicating its presence on the cytosolic side of cellular membranes . Given the palmitoylation of ORF3 protein at N-terminal cysteine-rich domain ( aa 5–21 ) , our data reveal that ORF3 protein is entirely oriented to the cytosolic side while associated with the membrane at least through palmitoylation . Supporting our findings , the well described PSAP motif of HEV ORF3 protein ( aa 95–99 for genotype 3 ) required for the interaction with Tsg101 [19 , 21 , 37] needs to be oriented toward the cytosol to allow association with the ESCRT-I protein . Furthermore , ORF3 protein is not only present in infected cells but is also believed to be part of the quasi-enveloped virus secreted into the bloodstream or in cell culture [15 , 19 , 22] . These reports demonstrated that ORF3 protein , more precisely epitope aa 90–113 , is detectable only when quasi-enveloped virus is treated with a detergent to permeabilize the membrane . Hence , consistent with our findings , these reports indicate that ORF3 protein is present inside the exosomes constituting the quasi-enveloped HEV particle [16] . Thus , the membrane topology suggested by Ding and colleagues , where ORF3 protein would possess a transmembrane segment comprising aa 30–52 , is not fully supported by our observations as well as by others . However , viral proteins are often multifunctional and may adopt different membrane topologies to exert different functions , as it is the case for the hepatitis B virus large surface protein [38] , the fusion protein of Newcastle disease virus [39] or HCV nonstructural protein 4B [30] . Therefore , it is conceivable that a subset of ORF3 protein at a particular localization or at a defined stage of the infection forms pores to exert ion channel activity . Previous reports showed that HEV ORF3 protein is crucial for infectious virus secretion [20] and is incorporated into the quasi-enveloped virion [15 , 22] . Given the membrane association as well as the palmitoylation of ORF3 protein described here , one may hypothesize that the posttranslational modification of ORF3 protein regulates its function in virus secretion . Indeed , we demonstrated that the cysteine residues of HEV ORF3 protein , engaged in palmitoylation , are essential for the secretion of infectious particles . Interestingly , a recent report demonstrated that palmitoylation of the small transframe protein of Sindbis virus , another positive-strand RNA virus , determines its localization to the plasma membrane and regulates its incorporation into the particle and thereby virus secretion [40] . Hence , future investigations shall aim at identifying the host palmitoyltransferase ( s ) responsible for ORF3 protein modification to further explore the functional consequences of this posttranslational modification on the virus life cycle but also the pathogenesis of HEV . To conclude , our results point toward an important host-dependent posttranslational modification of HEV ORF3 protein , its palmitoylation , which has important functional implications for HEV biology .
U-2 OS human osteosarcoma ( obtained from the American Type Culture Collection , Rockville , MD ) and S10-3 human hepatocellular carcinoma cell lines ( kindly provided by Suzanne U . Emerson , National Institutes of Health , Bethesda , MD ) were maintained in Dulbecco’s modified Eagle medium supplemented with 10% fetal bovine serum . The Hep293TT human hepatoblastoma cell line ( kindly provided by Gail E . Tomlinson , San Antonio , TX ) was maintained in Roswell Park Memorial Institute medium containing HEPES buffer , L-glutamine and 10% fetal bovine serum . Plasmids encoding the full-length HEV infectious clones p6 ( Kernow-C1 strain ) [41] and pSK-HEV-2 ( Sar55 strain ) [42] as well as the subgenomic replicon p6-luc were kindly provided by Suzanne U . Emerson . The polymerase-deficient replicon construct p6-luc-GAD was prepared as described previously [1 , 43] . Plasmid encoding the full-length HEV infectious clone 83–2 [44] was kindly provided by Koji Ishii as well as Takaji Wakita ( Department of Virology II , National Institute of Infectious Diseases , Japan ) . Monoclonal antibodies ( mAb ) anti-FLAG M2 and anti-β-actin were from Sigma-Aldrich ( Saint-Louis , MI ) . Polyclonal antibody ( pAb ) against HA tag ( Y-11 ) and mouse mAbs 11G5a against CD151 were from Santa Cruz Biotechnology ( Dallas , TX ) . Rabbit pAb against MAVS was from Enzo Life Sciences ( Farmingdale , NY ) . MAb JL8 against GFP was from Takara Bio ( Mountain View , CA ) , mAb G1/296 against CLIMP-63 from Enzo Life Sciences and mAb JG1 against heat shock protein 70 ( Hsp70 ) from Affinity Bioreagents ( Golden , CO ) . Secondary antibodies were HRP-conjugated anti-mouse ( GE Healthcare; Chicago , IL ) and anti-rabbit ( Agilent Technologies; Santa Clara , CA ) . Rabbit anti-ORF3 and swine anti-ORF2 pAbs were kind gifts from Suzanne U . Emerson . Rabbit antiserum against HEV ORF2 was a kind gift from Rainer G . Ulrich ( Friedrich-Loeffler-Institute , Greifswald-Insel Riems , Germany ) [45] . Recombinant mouse mAb against genotype 3 HEV ORF3 protein was produced as described [46] by the Geneva Antibody Facility ( https://web . expasy . org/abcd/ABCD_RB198 ) . MAbs C7-50 against HCV core [47] and A4 against HCV E1 [48] ( kindly provided by Jean Dubuisson , University of Lille , France ) have been described . Primers used in this study are listed in S1 Table . Unless specified otherwise , the constructs described here were derived from the HEV genotype 3 p6 infectious clone ( Kernow-C1 strain , GenBank accession number JQ679013 ) as template and verified by sequencing . ORF3 expression construct pCMVORF3 and pCMVORF3Δ19 were prepared by PCR amplification with primers ORF3-1-Hind-fd or ORF3-19-Hind-fd , respectively , and ORF3-113st-Bam-rv , followed by cloning into the HindIII/BamHI sites of pcDNA3 . 1 ( + ) ( Thermo Fischer Scientific , Waltham , MA ) . The phosphorylation-deficient mutant ORF3 construct pCMVORF3S70A was prepared by site-directed mutagenesis using primer pair HEVKc1_S70A-fd and HEVKc1_S70A-rv as well as pCMVORF3 as template . Similarly , pCMVORF3_gt1 , allowing the expression of genotype 1 HEV ORF3 protein , was prepared by PCR amplification with primers ORF3-gt1-1Hind-fd and ORF3-gt1-115Bam-rv , using pSK-HEV-2 ( Sar55 strain , GenBank accession number AF444002 ) as template , followed by cloning into the HindIII/BamHI sites of pcDNA3 . 1 ( + ) . Constructs used for co-immunoprecipitation , i . e . pCMVORF3-FLAG and pCMVORF3-HA , were prepared by PCR amplification with forward primer ORF3-1-Hind-fd and reverse primer ORF3-113-Bam-rv , followed by HindIII/BamHI digestion and cloning into previously described plasmids pCMV-X-FLAG and pCMVNS4B-HA [35] . The double-tagged ORF3 construct pCMVFLAG-ORF3-HA used for selective permeabilization was prepared by PCR amplification with forward primer ORF3noATG-1-Bsp-fd and reverse primer BGH-rv , followed by BspEI/BamHI digestion and cloning into pCMVFLAG-JFH4B-HA [35] . GFP fusion constructs pCMVORF3-GFP , pCMVORF31-53-GFP , pCMVORF31-28-GFP , pCMVORF328-113-GFP , pCMVORF353-113-GFP and pCMVORF328-53-GFP were prepared by PCR amplification with forward primers ORF3-1-Hind-fd , ORF3-28-Hind-fd or ORF3-53-Hind-fd and reverse primer ORF3-113-Bam-rv , ORF3-53-Bam-rv or ORF3-28-Bam-rv , followed by cloning into the HindIII/BamHI sites of pCMVKEB-GFP [49] . FRET constructs , i . e . pCMVORF3-CFP , pCMVORF3-YFP , pCMVORF31-93-YFP , pCMVORF31-93-CFP , pCMVORF31-70-CFP , pCMVORF31-70-YFP , pCMVORF31-53-CFP , pCMVORF31-53-YFP , pCMVORF328-113-CFP , pCMVORF328-113-YFP , pCMVORF353-113-CFP and pCMVORF353-113-YFP , were prepared by PCR amplification with forward primers ORF3-1-Hind-fd , ORF3-28-Hind-fd or ORF3-53-Hind-fd and reverse primer ORF3-113-Bam-rv , ORF3-93-Bam-rv , ORF3-70-Bam-rv or ORF3-53-Bam-rv , followed by cloning into the HindIII/BamHI sites of pCMVNS4B-YFP or pCMVNS4B-CFP [35] . Alanine substitution mutants , i . e . ORF3C1-4 , ORF3C5-8 and ORF3C1-8 , were synthesized by GenScript ( Piscataway , NJ ) . Constructs pCMVORF3C1-4 , pCMVORF3C5-8 and pCMVORF3C1-8 were prepared by cassette exchange by cloning into HindIII-BamHI sites of pCMVORF3 . GFP fusion constructs harboring the mutations , i . e . pCMVORF3C1-4-GFP , pCMVORF3C5-8-GFP and pCMVORF3C1-8-GFP , were prepared by PCR amplification from the corresponding pCMVORF3 constructs with forward primer CMV-fd and reverse primer ORF3-113-Bam-rv , followed by HindIII/BamHI digestion and cloning into pCMVKEB-GFP . The construct allowing expression of ORF3 in a wheat germ-based cell free expression system , pEU-ORF3 , was prepared by cassette exchange of , the ORF3 coding sequence by digestion of pCMVORF3 with PmeI-BamHI and cloning into the EcoRV-BamHI sites of pEU-MCS-E01 ( CellFree Sciences; Matsuyama , Japan ) . A 977-bp DNA fragment of the HEV p6 clone bordered by AflII and PmlI restriction sites and harboring a serine substitution of 5 N-terminal cysteine residues , i . e . p6_C5S , was synthesized by GenScript . Construct p6_C5S was prepared by cloning into the AflII-PmlI sites of the p6 wt plasmid . ORF3 expression constructs harboring the C5S mutation , i . e . pCMVORF3C5S and pCMVORF3C5S-GFP , were prepared by PCR amplification from the p6_C5S construct with forward primer ORF3-1-Hind-fd and reverse primer ORF3-113st-Bam-rv or ORF3-113-Bam-rv , respectively , followed by HindIII/BamHI digestion and cloning into pcDNA3 . 1 ( + ) or pCMVKEB-GFP . Protein lysates were prepared and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-PAGE ) , followed by immunoblot analysis as described previously [50] . Capped full-length HEV RNA was produced by in vitro transcription using the mMACHINE mMESSAGE Kit from Ambion ( Thermo Fischer Scientific ) , followed by electroporation into Hep293TT cells or transfection of S10-3 cells using TransIT-mRNA transfection kit ( Mirus Bio LLC , Madison , WI ) [43] . Cells were cultured for 5 to 8 days before harvesting to prepare protein lysates , to perform immunofluorescence or to collect intra- and extracellular compartments for infectivity determination . FRET analysis has been performed as previously described [35] . Briefly , U-2 OS cells cultured on glass coverslips were transfected with constructs expressing CFP- and YFP-tagged proteins and fixed at 24 h posttransfection with 2% paraformaldehyde for 5 min and the coverslips were mounted on glass slides with SlowFade reagent ( Thermo Fischer Scientific ) . Acceptor photobleaching FRET was performed using an SP5 AOBS confocal laser scanning microscope ( Leica; Wetzlar , Germany ) . To detect palmitoylation , U-2 OS cells were transfected with pCMVORF3 , pCMVORF3C1-4 or pCMVORF3C5-8 plasmids and Hep293TT were electroporated with full-length HEV p6 or 83–2 RNAs prior to incubation with radiolabelled palmitate 24 h or 6 days post-transfection , respectively . The cells were washed twice and incubated at 37°C for 1 h in the respective medium without FCS then followed by 3 h incubation in the same medium supplemented with 200 μCi/mL 3H-palmitate ( 9 , 10-3H ( N ) ) ( American Radiolabeled Chemicals , St Louis , MO ) . After three washes with cold PBS , cells were lysed in immunoprecipitation buffer ( 50 mM Tris-HCl pH 7 . 4; 1 mM EDTA; 150 mM NaCl; 1% Triton X-100 ) supplemented with 1X cOmplete protease inhibitor cocktail ( Roche , Basel , Switzerland ) . Immunoprecipitation was carried out by the incubation of each protein lysates overnight at 4°C with 50 μL of Dynabeads Protein G ( Thermo Fischer Scientific ) pre-adsorbed with 1 μl of rabbit anti-ORF3 pAb , gift from Suzanne Emerson ( NIH , Bethesda , MD ) , or unrelated rabbit serum as control , following manufacturers’s recommendations . After three washes with 0 . 02% Tween-20 in PBS using a magnetic device , elution was performed by incubation of the beads in 30 μL of 100 mM glycine pH 3 for 5 min at RT . Samples were then incubated for 5 min at 90°C in Laemmli buffer and separated onto a 17% SDS-PAGE . Gel was either incubated 30 min at 20°C in a fixative solution ( 25% isopropanol , 65% H2O , 10% acetic acid ) , followed by a 30 min incubation with signal enhancer Amplify NAMP100 ( GE Healthcare ) and subjected to autoradiography for 40 days or undergone immunoblot analysis with pAb anti-ORF3 . S10-3 cells were grown onto 22-mm diameter glass coverslips before being assayed . Briefly , cells were fixed 10 min with 2% paraformaldehyde at 20°C . Fixed cells were washed with PBS and permeabilized with 0 . 5% saponin in PBS for 5 min , followed by a 20-min incubation in 10% goat serum blocking solution . Cells were successively incubated for 1 h with the first antibody and then with Alexa Fluor 488- and 594-conjugated anti-mouse and anti-rabbit IgG antibodies ( Thermo Fisher Scientific ) . Additional incubation with DAPI ( Thermo Fisher Scientific ) allowed staining of the nuclei . Slides were prepared using antifade mounting medium ProLong ( Thermo Fisher Scientific ) . Cells were examined under a Zeiss LSM 710 quasar laser scanning fluorescence confocal microscope and images were treated with ImageJ software . S10-3 cells were seeded onto glass coverslips and transfected 24 h later with ORF3 expression vectors or co-transfected with pUHD15-1 [51] and pUHD-Cp7 plasmids allowing the expression of HCV core-p7 region , as described previously [30] . Forty-eight h post-transfection cells were fixed with 2% paraformaldehyde ( Sigma-Aldrich ) for 10 min and then permeabilized , totally , with either 0 . 5% saponin or 0 . 2% digitonin ( Sigma-Aldrich ) or selectively with 0 . 01% digitonin . Cells were then washed and incubated 15 min at 20°C in blocking buffer containing 3% bovine serum albumin in PBS . Indirect immunofluorescence was then performed by one-hour incubation at 20°C with primary antibody , followed by 3 PBS washes and incubation with the secondary antibody as described above . Significance values were calculated by using the unpaired t test with the GraphPad Prism 6 software package ( GraphPad Software ) .
|
Hepatitis E virus ( HEV ) infection is believed to be the most common cause of acute hepatitis and jaundice in the world . HEV is a positive-strand RNA virus found as a non-enveloped virion in bile and feces or as a quasi-enveloped virion in blood and in cell culture . The HEV ORF3 protein is involved in viral particle secretion likely through the exosomal pathway . Here , we provide evidence for palmitoylation of ORF3 protein at its N-terminal cysteine-rich domain . Palmitoylation of ORF3 protein determines its subcellular localization and function in particle secretion . In addition , our data indicate a membrane topology where HEV ORF3 protein is entirely exposed to the cytosol , providing important insight into its interactions in the viral life cycle .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"molecular",
"probe",
"techniques",
"pathogens",
"microbiology",
"organic",
"compounds",
"membrane",
"proteins",
"viruses",
"amino",
"acids",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"cysteine",
"research",
"and",
"analysis",
"methods",
"immunoblot",
"analysis",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"chemistry",
"hepatitis",
"viruses",
"subcellular",
"localization",
"molecular",
"biology",
"cell",
"membranes",
"sulfur",
"containing",
"amino",
"acids",
"biochemistry",
"palmitoylation",
"cell",
"biology",
"post-translational",
"modification",
"organic",
"chemistry",
"integral",
"membrane",
"proteins",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms",
"hepatitis",
"e",
"virus"
] |
2018
|
Palmitoylation mediates membrane association of hepatitis E virus ORF3 protein and is required for infectious particle secretion
|
Eukaryotic mRNA transcription and turnover is controlled by an enzymatic machinery that includes RNA polymerase II and the 3′ to 5′ exosome . The activity of these protein complexes is modulated by additional factors , such as the nuclear RNA polymerase II-associated factor 1 ( Paf1c ) and the cytoplasmic Superkiller ( SKI ) complex , respectively . Their components are conserved across uni- as well as multi-cellular organisms , including yeast , Arabidopsis , and humans . Among them , SKI8 displays multiple facets on top of its cytoplasmic role in the SKI complex . For instance , nuclear yeast ScSKI8 has an additional function in meiotic recombination , whereas nuclear human hSKI8 ( unlike ScSKI8 ) associates with Paf1c . The Arabidopsis SKI8 homolog VERNALIZATION INDEPENDENT 3 ( VIP3 ) has been found in Paf1c as well; however , whether it also has a role in the SKI complex remains obscure so far . We found that transgenic VIP3-GFP , which complements a novel vip3 mutant allele , localizes to both nucleus and cytoplasm . Consistently , biochemical analyses suggest that VIP3–GFP associates with the SKI complex . A role of VIP3 in the turnover of nuclear encoded mRNAs is supported by random-primed RNA sequencing of wild-type and vip3 seedlings , which indicates mRNA stabilization in vip3 . Another SKI subunit homolog mutant , ski2 , displays a dwarf phenotype similar to vip3 . However , unlike vip3 , it displays neither early flowering nor flower development phenotypes , suggesting that the latter reflect VIP3's role in Paf1c . Surprisingly then , transgenic ScSKI8 rescued all aspects of the vip3 phenotype , suggesting that the dual role of SKI8 depends on species-specific cellular context .
Production and turnover of eukaryotic mRNAs are highly conserved processes , which are mainly driven by RNA polymerase II ( RNAPolII ) and the 3′ to 5′ exosome ( exosome ) , respectively [1] , [2] . Regulation of transcription initiation by RNAPolII through promoter sequence-specific transcription factors is a major topic in developmental biology , since it is considered the prime mechanism for differential , cell and organ type-specific gene expression [3] . However , generic accessory factors , which are typically heteromultimeric protein complexes , exist as well . Compared to the RNAPolII machinery , they are less conserved but have been found in all uni- and multicellular eukaryotes investigated so far . In line with their lower conservation , these factors are generally not essential . However , loss of function mutations in their subunits typically result in pleiotropic phenotypes with varying degrees of severity . An example is the Mediator complex , which typically comprises more than 15 subunits and interacts with the C-terminal domain of the largest RNAPolII subunit [4] , [5] . In yeast ( S . cerevisiae ) , Mediator is associated with constitutively transcribed genes [6] and yeast Mediator mutants are typically viable but display impaired growth [4] . In multicellular organisms , the composition of Mediator is even more complex and individual subunit loss of function can lead to rather specific phenotypes . For instance , in the model plant Arabidopsis ( A . thaliana ) , in which several additional Mediator subunits have been identified [7] , respective mutants display such diverse phenotypes as increased cell proliferation , shifts in embryonic patterning or early flowering [7] , [8] , [9] . Screens for early flowering mutants also identified Arabidopsis subunit homologs of another conserved multimeric regulator of transcription , the RNAPolII-associated factor 1 complex ( Paf1c ) [10] , [11] , [12] . In yeast , Paf1c consists of five subunits [13] , whose Arabidopsis homologs are VERNALIZATION INDEPENDENCE ( VIP ) 4 , VIP5 , EARLY FLOWERING ( ELF ) 7 , VIP6/ELF8 and PLANT HOMOLOGOUS TO PARAFIBROMIN ( PHP ) [10] , [14] , [15] , [16] . Among the respective loss of function mutants , php mutants only flower early , whereas vip4 , vip5 , elf7 and vip6/elf8 mutants all display additional pleiotropic growth defects and aberrant flower development ( e . g . , variable floral organ number ) . The early flowering phenotype of vip/elf mutants has been linked to down-regulation of the central flowering time regulator , FLOWERING LOCUS C ( FLC ) , via an epigenetic mechanism , consistent with a role of Paf1c in chromatin modification through changing histone methylation patterns [10] , [11] , [14] . The latter could also explain the phenotypes in flower development , which can be altered by mutation in epigenetic regulators [17] . Another mutant with dwarf , early flowering and aberrant flower development phenotypes is vip3 . VIP3 encodes a WD40 repeat protein , which is the putative Arabidopsis homolog of the yeast Superkiller ( Ski ) 8 gene [12] . SKI8 is part of the cytosolic SKI complex , which is thought to positively regulate exosome activity [1] , [18] , [19] . The SKI complex consists of a SKI8 dimer and the SKI2 RNA helicase , which are connected by their mutual interaction with the scaffold protein SKI3 [20] . Interestingly , human hSki8 as well as VIP3 also associate with Paf1c [11] , [21] , which is not the case for yeast ScSki8 [21] , [22] . Rather , ScSki8 has a SKI complex-independent nuclear function in meiotic recombination [23] . This feature is not conserved in VIP3 [24] , suggesting that Ski8 activity in plants might be functionally closer to mammals than unicellular eukaryotes . Compared to its well documented role in Paf1c , the potential role of VIP3 in the SKI complex has not been characterized . Notably , although VIP3 is the top hit in a homology search of the Arabidopsis proteome using ScSki8 as a query , the next best hits are nearly equally significant with better overall coverage and represent structurally similar WD40 repeat proteins . Conversely , if the yeast proteome is queried with VIP3 , more than two dozen hits score markedly better than ScSki8 . Interestingly , the top hits , like PRP4 or TUP1 , have been described as modifiers of pre-mRNA processing or chromatin modifications , respectively [25] , [26] , which would also be consistent with existing experimental data on VIP3 activity . However , reciprocal BLAST searches with higher eukaryotes clearly identify the respective SKI8 homologs as best hits . Still , experimental evidence for VIP3 involvement in the Arabidopsis SKI complex and the facets of the vip3 phenotype that could be attributed to this role is missing . In this study , we investigate this question by a combination of biochemical , genetic and high throughput techniques .
Analysis of natural genetic variation has become a common tool for isolation of allelic variants in Arabidopsis , facilitated by availability of collections of wild strains , so-called accessions . In the Slavice-0 ( Sav-0 ) accession , we found largely infertile dwarf plants segregating at low frequency when grown in permanent light conditions and low humidity ( ∼40% ) ( Figure 1A ) . Moreover , careful investigation of the segregating population revealed a substantial fraction of non-viable , seedling lethal individuals . At higher humidity ( ∼60% ) and long day conditions , the fraction of seedling lethals decreased considerably , whereas the ratio of dwarfs increased to near Mendelian ( typically>20% ) proportion , suggesting that the two classes represent the phenotypic spectrum of the same underlying genetic cause . The infertility of the dwarf plants , which rarely produced seeds and if so , very little ( Figure 1E–1G ) , could be overcome to some degree by out-crossing with pollen from wild type looking plants . This allowed us to generate a segregating F2 population derived from a cross to the standard lab accession , Columbia-0 ( Col-0 ) . Genetic mapping revealed that the dwarf and infertility phenotypes segregated as a recessive single Mendelian locus on the lower arm of chromosome 4 , which we named ZWERGERL ( ZWG , Bavarian for “dwarf” ) . A detailed analysis of the zwg phenotype revealed various floral defects . These included a low penetrance aberrant floral organ number phenotype ( Figure 1B–1C ) and shorter sepals ( Figure 1D ) . Shorter anthers with few viable pollen accounted for the decreased fertility . The floral phenotypes were accompanied by early flowering , which was evident both in terms of age and rosette leaf number ( Figure 1H ) . By following the development of individual seedlings from germination in tissue culture onwards , we could also detect reduced root elongation in zwg plants ( Figure 1I ) . Radial growth of all organs was affected as well , as exemplified by the dramatically reduced diameter of the main inflorescence stem ( Figure 1J–1M ) . Since the relative decrease in cell number ( ∼75% of wild type ) ( Figure 1J ) was not as strong as the overall decrease in diameter ( ∼50% of wild type ) ( Figure 1K–1M ) , the reduced organ size in zwg mutants likely represents a combination of impaired cell proliferation and expansion . Complementing the morphological characterization , we also analyzed the zwg transcriptome by hybridization of CATMA microarrays [27] with cDNA prepared from aerial tissues . Based on four replicate hybridizations , statistically solid expression changes ( ≥2-fold; p≤0 . 05 ) were found for 173 genes that were up-regulated and 425 genes that were down-regulated in zwg as compared to wild type ( Table S1 ) . These gene lists did not point to any specific defect in zwg mutants , such as mis-regulation of a particular hormone pathway . Rather , the genes represented an overall balanced sample across functional categories as illustrated by their gene ontology analysis ( Figure 1N–1O ) . The only consistently over-represented category in both the up- and down-regulated sets was response to stress . In summary , our morphological as well as molecular characterization suggests that a general growth defect is responsible for the panoply of zwg mutant phenotypes . To identify the molecular cause of the zwg mutation , we sequenced the genomes of Sav-0 wild type and zwg individuals with short reads [28] . Mapping of the reads onto the Col-0 reference genome revealed an extended region of heterozygosity on the lower arm of chromosome 4 in Sav-0 that encompassed the ZWG locus . The sequence information was exploited to generate polymorphic molecular markers that allowed mapping of the zwg mutation in the zwg x Col-0 population ( Figure 2A ) . Within the zero recombination mapping interval , the sequence reads indicated the presence of a homozygous 7 bp deletion in the coding sequence of At4g29830 , previously described as VIP3 , in zwg but not in wild type ( Figure 2B ) , which was confirmed by Sanger sequencing of respective PCR fragments . Analysis of a cross between zwg and a vip3 null mutant ( SALK_083364 ) [29] indicated non-complementation , confirming that zwg is indeed a new vip3 allele , which we thus named vip3zwg . The deletion in vip3zwg encompasses nucleotides 861–867 of the open reading frame of the mRNA , which is expressed at similar levels in vip3zwg and wild type . Conceptual translation predicts that the deletion causes a frameshift to produce a 36 kDa instead of a 32 kDa protein with a modified and extended C-terminus , thereby disrupting the last of the five WD40 repeats of VIP3 ( Figure 2C ) . Because of its phenotypic resemblance with the knock out allele , including the down-regulation of FLC expression ( Table S1 ) , and its recessive behavior , vip3zwg can be considered a null allele . To clarify whether VIP3 is indeed the functional Arabidopsis SKI8 homolog , we sought to determine its subcellular localization . To this end , we created a binary construct for expression of a GFP-VIP3 fusion under control of the constitutive 35S promoter . This transgene was introduced into Sav-0 wild type-looking plants that were heterozygous for the zwg mutation as determined by genotyping of the 7 bp deletion on high resolution agarose gels . Western analysis revealed variable expression of GFP-VIP3 fusion protein of the expected size in several independent lines ( Figure 2D ) , within one order of magnitude of the level of endogenous VIP3 as judged from qPCR . In the progeny of these plants , the zwg phenotype segregated in a proportion close to 1/16th rather than 1/4th and was significantly different from the segregation in the parallel grown non-transgenic mother line ( Chi-square = 11 . 54 for df = 1 , significant at p<0 . 001 ) . None of the plants with a zwg phenotype carried the transgene as determined by genotyping . All other plants appeared wild type , suggesting that the fusion protein is functional and rescues all zwg phenotypes . Confocal microscopy showed both cytoplasmic and nuclear ( but not nucleolar ) localization of GFP-VIP3 , in differentiated as well as proliferating cells , in both root and shoot tissues ( Figure 2E–2G ) . Matching the dual subcellular localization , analysis of protein extracts by gel filtration detected the presence of GFP-VIP3 in at least two peaks , one in the ∼690 kDa and another in the 300 kDa range ( Figure 2H ) . Moreover , substantial amounts were observed in smaller ( 100–200 kDa ) fractions . To determine whether any of these fractions could represent the SKI complex or its sub-components , we collected three distinct sets of fractions after gel filtration and performed immunoprecipitations with anti-GFP antibody . Subsequent MALDI-TOF identified peptides of the Arabidopsis SKI3 homolog ( At1g76630 ) in the pool of the smaller fractions ( Figure 2I ) . Notably , protein homology searches unequivocally identify At1g76630 and ScSki3 as unique reciprocal and highly significant hits , suggesting that At1g76630 represents indeed the Arabidopsis SKI3 homolog . No other SKI complex or Paf1c components were identified , which might have resulted from our stringent conditions combined with previous gel filtration . Direct immunoprecipitation from total protein extract using the same conditions indeed not only identified AtSKI3 , but also AtSKI2 ( At3g46960 , see below ) and the Paf1c component PHP ( Figure 2J ) . Thus , our analyses suggest that VIP3 is not only part of Paf1c in the nucleus , but also of the cytoplasmic SKI complex and likely represents the true SKI8 homolog . To corroborate the consequent notion that VIP3 should have a role in mRNA turnover , we applied a high throughput sequencing strategy to RNA samples isolated from vip3zwg and wild type . Because we aimed to sequence both full length mRNAs and mRNAs undergoing ( 3′ to 5′ ) degradation , cDNA from these samples was produced by random-primed rather than poly-T-primed synthesis . Prior to this , mRNA was enriched by removing the bulk of ribosomal RNA with the help of capture columns . The cDNA was then size-fractionated and the 200 bp fraction was used for preparation of the library , which was sequenced to produce single reads of 75 bp ( 21 . 3 mio . for wild type; 25 . 2 mio . for vip3zwg ) . The reads were mapped onto the Col-0 reference transcriptome , including the 5′ and 3′ UTRs , with relaxed stringency to accommodate nucleotide polymorphisms between Sav-0/vip3zwg and Col-0 [28] . Parallel mapping onto the Col-0 reference genome placed the large majority of reads in exons ( 80 . 2% in wild type; 84 . 5% in vip3zwg ) , confirming that our sequence data represent RNA molecules and that genomic contamination , if any , is negligible ( Table S2 ) . For the follow up analyses , we concentrated on the reads that mapped onto mRNA ( 27 . 0% in wild type; 13 . 8% in vip3zwg ) , and in particular on the nuclear encoded genes ( 16 . 7% of reads in wild type; 6 . 6% in vip3zwg ) . In total , of 26’598 transcripts interrogated , 14’228 were covered by at least one read in both the wild type and vip3zwg sample . In order to obtain a parameter that would allow us to estimate the steady state abundance of full length versus degrading mRNAs , we calculated the ratio between the number of reads mapping onto the 5′-most 20% of a transcript versus those mapping onto the 3′-most 20% . After removal of nonsense values ( i . e . 0 or ∞ because of absent coverage of one end ) and outliers with extreme values ( resulting from excess read abundance combined with obvious mis-mapping , e . g . reads covering the flanking region of a gypsy-like retrotransposon [At4g06477] ) , the distribution of this 5′ to 3′ coverage index was skewed towards values >1 . To some degree this likely represents a technical bias [30] , but could also reflect a dominant role of 3′ to 5′ degradation in mRNA turnover . Interestingly , the 5′ to 3′ coverage index was generally higher in the wild type than in the vip3zwg sample ( Figure 3A ) . To verify that this was not a technical artifact , we compared the relative proportion of the accumulated reads in 1% bins along the 10% most highly expressed nuclear encoded transcripts of wild type . The respective profiles for the wild type and vip3zwg sample were similar ( Figure 3B ) , suggesting that the RNA sequencing data from the two samples are comparable . In order to remove statistically doubtful 5′ to 3′ coverage index values that were due to low transcript abundance , we only considered the 6’500 transcripts for which at least 50% of sequence was covered in both the wild type and vip3zwg samples in follow up analyses . Remaining outliers with index values ≥10 or ≤0 . 1 were removed as well . From this set , we extracted the group of 5’617 nuclear encoded transcripts that were not strongly affected by depletion of exosome activity [31] , as well as 34 chloroplast-encoded transcripts and 68 nuclear-encoded transcripts that were significantly stabilized upon exosome depletion and can be considered prime exosome targets [“the hidden transcriptome”; 31] ( Table S3 ) . The 5′ to 3′ coverage index value distribution in the nuclear control group confirmed the earlier picture of higher overall values in wild type as compared to vip3zwg , which is for instance also evident in the ratios between the respective averages or medians ( Figure 3C ) . While no significant difference was found in the chloroplast transcripts , this trend was amplified in the prime exosome targets , which displayed higher average and median index values in wild type and lower ones in vip3zwg as compared to the nuclear control group ( Figure 3C ) . To evaluate the robustness of the difference between the nuclear control group and the exosome targets , we determined the index value distribution for 1’000 random sets of 68 genes extracted from the nuclear control group . These analyses confirmed the trend towards higher values in wild type , underlined by the finding that the wild type to vip3zwg ratio of averages and medians was nearly always >1 ( Figure 3D ) . Notably , even the maximum ratios observed within the 1’000 sets did not or barely reach the values observed in the exosome target set ( 1 . 50 versus 1 . 53 for the average , 1 . 40 versus 1 . 39 for the median ) . Conversely , within 59 random sets of 10 transcripts extracted from the exosome targets , the trend towards higher values in wild type and lower ones in vip3zwg including the ratios was always evident ( Figure 3D ) . We confirmed this finding by an independent method with independent , triplicate RNA preparations for a set of five randomly chosen genes . For each gene , oligonucleotide pairs for qPCR detection of the respective 5′ and 3′ mRNA ends were designed . The reverse primers for each fragment were used to prime separate cDNA synthesis reactions . Subsequent qPCR allowed quantification of the 5′ and 3′ end abundance for each gene in the replicate samples of the two genotypes . With one exception , the ratio between the 5′ and 3′ end abundance was always higher in wild type than in vip3zwg mutants , as expressed by the ratio between those ratios being greater than 1 ( Figure 3E ) . In summary , these analyses suggest that nuclear encoded mRNAs in general and prime exosome targets in particular are stabilized in vip3zwg mutants . To determine which aspects of the phenotype spectrum of vip3 mutants are due to its involvement in Paf1c or the SKI complex , respectively , we sought to characterize mutants in other SKI subunit homologs of Arabidopsis . Whereas knock out mutants in the SKI3 homolog were not available in reverse genetic collections [29] , a line segregating a T-DNA insertion in exon 9 out of 23 of At3g46960 ( SALK_118579 ) was available . Similar to At1g76630 and ScSki3 , reciprocal homology searches between Arabidopsis and yeast using ScSki2 or At3g46960 as a query identified each other as the uncontested top hits , suggesting that At3g46960 represents the unique ScSki2 homolog in Arabidopsis ( AtSKI2 ) . This notion is also supported by a phylogenetic analysis ( Figure 4A; Text S1 ) . Analysis of the SALK_118579 line revealed that it segregates up to ∼25% of dwarf plants ( Figure 4B–4C ) . This phenotype co-segregated perfectly with homozygosity of the T-DNA insert and absence of full length AtSKI2 mRNA . With the caveat that residual RNA production 3′ from the T-DNA insertion site has been reported previously [32] , the SALK_118579 line therefore might represent the Atski2 null mutant phenotype . Contrary to the vip3 mutants however , the dwarf phenotype was neither accompanied by a flower development nor an early flowering phenotype ( Figure 4D–4E ) . In line with the latter observation , FLC expression was strongly diminished in vip3zwg , but not in Atski2 mutants ( Figure 3F ) . In summary , these observations suggest that the flower development and early flowering phenotype of vip3 mutants could reflect VIP3's role in Paf1c rather than the SKI complex . Considering that ScSki8 has not been found to associate with Paf1c , we sought to corroborate this notion by testing whether transgenic ScSki8 could rescue the dwarf phenotype of vip3zwg mutants . To this end , the ScSki8 open reading frame was cloned into a binary construct for constitutive expression under control of the 35S promoter . Again , the transgene was introduced into Sav-0 wild type plants that were heterozygous for the vip3zwg mutation . Genotyping of the 7 bp deletion and the transgene in the segregating progeny identified several homozygous vip3zwg mutants carrying the 35S::ScSki8 transgene . These plants developed either as dwarf or as wild type , and this was correlated with transgene expression ( Figure 4F–4G ) . Thus , transgenic expression of ScSki8 could rescue the dwarf phenotype of vip3zwg mutants . Moreover , surprisingly both the early flowering and flower development phenotypes were also rescued ( Figure 4H ) . Therefore , our data suggest that once introduced into Arabidopsis , ScSki8 can fulfill all functions of VIP3 , including those not normally encountered in yeast itself .
In this study , we present experiments that lead to four main conclusions: First , we show that VIP3 is the bona fide SKI8 homolog of Arabidopsis; second , we demonstrate that next generation sequencing of random-primed RNA samples with short reads can be used to estimate the turnover of mRNA transcripts; third , we show that the phenotypic aspects of VIP3 function in Paf1c and the SKI complex can be separated; and fourth , we provide evidence that the dual role of SKI8 homologs in Paf1c and the SKI complex appears to depend on the species-specific cellular context . Our interest in VIP3 originates from the discovery of the zwg mutant that segregated in the Arabidopsis Sav-0 accession . It seems unlikely that the 7 bp deletion in vip3zwg represents indeed an allelic variant recovered from a natural environment because of its detrimental phenotypic consequences . A haplo-insufficient beneficial effect of vip3zwg could explain maintenance of the allele by balancing selection , however , we did not observe any obvious phenotypes in the heterozygous plants that would support this idea . Rather , it appears likely that vip3zwg is a spontaneous allele that has arisen during the propagation of the Sav-0 accession in stock centers starting in the 1960s over several decades [33] , [34] . At the outset of our study , it was still unclear whether VIP3 is indeed the Arabidopsis SKI8 homolog . While its role in epigenetic regulation of FLC transcription through association with the Paf1c complex had been well documented [11] , [14] , its potential role in the SKI complex had not been characterized . Because of the evolutionary distance between higher plants , yeast and mammals this could not be considered a given , in particular as VIP3 and SKI8 fall into an abundant class of structurally similar WD40 repeat proteins . This was underlined by the finding that ScSki8 is by far not the closest VIP3 homolog in yeast . For instance , position-specific iterated BLAST identifies more than two dozen yeast proteins that are more homologous to VIP3 than ScSki8 ( e . g . , a 93 . 2 score , 69% coverage and 5×10−24 e-value for PRP4 as compared to 48 . 9 score , 40% coverage and 5×10−8 e-value for ScSki8 ) . It is only our functional analyses that suggest that VIP3 is indeed the bona fide ScSki8 homolog . Consistent with a potential role in both Paf1c and the SKI complex , we found that VIP3 is present in both the nucleus and cytoplasm , and in at least two protein complexes of distinct size . The larger peak fractions around 690 kDA could represent Paf1c , whereas the peak around 300 kDa could represent the SKI complex [21] . A third peak around even smaller ( 100–200 kDa ) size fractions could represent partial components of these complexes or VIP3 dimers , which might accumulate in excess as the GFP-VIP3 transgenes were typically expressed at higher levels than endogenous VIP3 . Interestingly , immunoprecipitation of GFP-VIP3 after gel filtration identified association with the Arabidopsis SKI3 homolog , but not the SKI2 homolog . This might mean that the SKI complex dissociates into sub-components during gel filtration and/or that SKI2 is lost during immunoprecipitation washes . Alternatively , it could reflect the fact that SKI8 interaction with SKI3 is direct , while interaction with SKI2 is indirect [20] . However , when directly immunoprecipitated from total protein extract , AtSKI3 as well as AtSKI2 was pulled down in our stringent conditions , underlining that VIP3 is indeed part of the SKI complex . The notion that VIP3 is a functional subunit of the SKI complex is supported by our genome-wide analysis of mRNA stability in vip3zwg mutants . To estimate mRNA turnover was foremost a technical challenge , because it meant that standard cDNA synthesis using poly-T oligonucleotides directed against the 3′ poly-A tail of mRNAs could not be applied . This also abolished the inherent selection of the mRNA fraction for sequencing from the much larger amount of ribosomal or transfer RNAs . Instead , to also capture mRNAs undergoing 3′ to 5′ degradation , cDNA was synthesized with random-priming , and the mRNA fraction was enriched by removing ribosomal RNAs through capture columns . High throughput sequencing of the cDNA samples and subsequent read mapping onto the reference transcriptome revealed that our method efficiently enriched the mRNA fraction , which generally represents 1–2% in total RNA samples , about 5 to 10-fold . The relative read abundance along transcripts is to some degree determined by technical biases , such as the directionality of cDNA synthesis [30] . However , it should also reflect the steady state equilibrium between mRNA synthesis and breakdown considering that primers were not limiting in cDNA synthesis and that poly-A tails provide priming sites but are not included in the sequence analysis . Generally , the coverage profiles displayed a decrease from 5′ to 3′ , suggesting that exosome-mediated 3′ to 5′ degradation is the main driver of mRNA breakdown [18] , [35] . To quantify the stability of individual transcripts , we defined a 5′ to 3′ coverage index , which was generally >1 , consistent with the overall profile . The comparison of the 5′-most 20% of a transcript versus its 3′-most 20% was designed to avoid skewed values in the case of poorly covered transcripts , and indeed comparatively few outliers were observed . In some cases , these reflected obvious mismappings because of repetitive or redundant sequences ( e . g . retrotransposon borders ) , while in others mismapping might have occurred because of the relaxed stringency that was required to map mRNA sequences from a divergent accession onto the reference transcriptome [28] . Overall , the patterns as well as the quantitative difference between the wild type and vip3zwg samples were robust , even if more selective criteria were applied or if other indexes were considered , such as linear fitting of read coverage . Thus , the index values suggest that in the vip3zwg sample the relative abundance of intact 3′ ends as compared to 5′ ends is higher , pointing to a shifted steady state equilibrium between mRNA transcription and degradation . This finding is consistent with the generic role of the SKI complex in exosome activation [18] and was particularly evident in the group of the most prominent exosome targets , termed the “hidden transcriptome” [31] . In summary , our data support the idea that VIP3 is a SKI complex component that affects mRNA stability and that random-primed RNA-Seq is a valid approach to estimate mRNA turnover . The implication of VIP3 in the SKI complex suggests that the vip3 phenotype should reflect the combination of VIP3 function in both Paf1c and the SKI complex . The availability of a mutant in the AtSKI2 gene , which can be unequivocally identified by homology searches , enabled us to disentangle the two activities . Interestingly , Atski2 plants displayed dwarfism , but neither early flowering nor aberrant flower development . Thus , the latter aspects of the vip3 phenotype should primarily result from impaired Paf1c function . It is noteworthy however that the Atski2 dwarf phenotype is not as severe as in vip3 , and that growth defects have also been observed in mutants of other Paf1c components . It thus appears likely that the SKI complex-related growth defects in vip3 are aggravated by the additionally impaired Paf1c activity . To clarify more directly which portions of the vip3 phenotype are attributable to impaired Paf1c or SKI complex function , we sought to exploit the fact that ScSki8 does not associate with Paf1c in yeast [21] , [22] and presumably also not in Arabidopsis . However , to our surprise ScSki8 was able to fully rescue all aspects of the vip3 phenotype . Thus , it appears that in the cellular context of Arabidopsis , ScSKI8 can fulfill VIP3's role in Paf1c . This could mean that other factors determine whether SKI8 is recruited to Paf1c or not , and that in this sense Arabidopsis is closer to mammals than yeast . Indeed we also tried to complement vip3zwg by constitutive expression of the mouse SKI8 homolog , WDR61 . However , for unknown reasons , we never managed to recover transgenic plants in repeated transformation attempts , which could mean that WDR61 expression is poisonous for Arabidopsis . Thus , while cellular context must play an important role , SKI8 function might to some degree also depend on inherent features . Future experiments to determine the interaction patterns of different SKI8 homologs and derivative point mutants of interest are a promising avenue to clarify this issue in detail .
The Sav-0 accession used in this study has been described previously [28] . T-DNA insertion lines were obtained from the Nottingham Arabidopsis Stock Centre and the insertions in lines SALK_083364 [knock out of VIP3 ( At4g29830 ) ] and SALK_118579 [knock out of AtSKI2 ( At3g46960 ) ] were confirmed by PCR analysis . For propagation and analysis of lines , seeds were germinated on half-strength Murashige & Skoog media in tissue culture and transferred to soil at 10–12 days after germination . Plants were then grown in either permanent light and ∼40% humidity , or 16 hr light–8 hr dark cycles and ∼60% humidity at 22°C . The latter conditions were used for characterization of the phenotypes displayed in the figures . For determination of flowering time , seeds were germinated directly on soil and the number of days or of rosette leaves was scored on the first day when the inflorescence meristem became visible . For root growth measurements , seedlings grown vertically in tissue culture were scored at 9 days after germination using ImageJ software and then transferred onto soil to determine wild type or mutant phenotype in the adult shoot . For transverse sections , stem segments encompassing the first internode were cut with a razor blade and embedded in 6% agarose . From these samples , 85 µm sections were obtained using a Leica-VT 1000S vibratom and photographed using a Leica Diaplan 3 microscope . Subcellular localization of GPF-VIP3 was determined in shoots and roots of 6 day old seedling using a Zeiss LSM 510 confocal microscope . For the propidium iodide staining the roots were incubated for 2–5 min in a 50 µg/ml solution . For expression of VIP3 or ScSki8 under control of the 35S promoter , the respective open reading frames amplified from cDNA samples were cloned into vectors pMDC43 or pMD32 [38] , respectively . Construct integrity was verified by Sanger sequencing before transfer into Agrobacterium and transformation of Arabidopsis plants by the floral dip method . Constructs were introduced into Sav-0 wild type looking plants that were heterozygous for vip3zwg as determined by genotyping . For genotyping , PCR was performed on genomic DNA using oligonucleotides GAG CTG CGA TTC AGA CAA TGA G and GCC CGG ACA CCG GTT CCA C . The PCR products of 87 bp from the wild type or 80 bp from the vip3zwg allele were resolved on 4% agarose gels . Transformants were selected on hygromycin and homozygous vip3zwg plants among the transformants were selected by genotyping . Transgenic GFP-VIP3 or ScSki8 expression levels were determined by quantitative real time or semi-quantitative RT-PCR , respectively , and normalized compared to the EF1 gene as described [36] . For microarray analysis , a mix of equivalent amounts of aerial tissues ( rosette leaves , stems , cauline leaves , inflorescences ) from 4 week old adult plants was collected and frozen in liquid nitrogen before total RNA was prepped using the QIAGEN RNeasy Plant Mini Kit . cDNA synthesis , labeling , hybridization onto CATMA microarrays [27] and data analysis was then performed as described previously [36] . For RNA-Seq , rosette leaves were harvested from 25 day old phenotypically wild type or vip3zwg plants . Genomic DNA was isolated from one part of each sample to verify genotypes , whereas total RNA was prepped from the remaining tissue using a QIAGEN RNeasy Plant Mini Kit . Ribosomal RNA was subsequently largely removed by treating 10 µg of total RNA with Invitrogen RiboMinus Plant Kits following the manufacturer's instructions . The enriched mRNA samples were then subjected to random-primed cDNA synthesis , amplification , size selection and high throughput sequencing with 75 bp single reads on an Illumina instrument as described [28] . Read mapping onto the Col-0 reference genome or transcriptome ( main gene models , TAIR 9 . 0 release ) was performed using the BWA program [39] with a seed length of 50 bp and up to 5 mismatches or gaps allowed . Total protein was extracted from 35S::GFP-VIP3 or 35S::GFP transgenic plants as described before fractionation by gel filtration using an Amersham Superdex 200 10/300 GL FPLC column with a buffer flow rate of 0 . 5 ml/min [40] . Consecutive 0 . 5 ml fractions were collected , concentrated and subjected to 10% SDS–PAGE followed by protein immunoblot analysis . Fusion protein was detected using an anti-GFP antibody ( dilution 1∶3000 ) ( Living colors , Clontech ) . For co-immunoprecipitation of GFP-VIP3 , three consecutive sets of gel filtration fractions were pooled and incubated for 90 min . at 4°C with 50 µl of μ-magnetic beads conjugated to anti-GFP antibody ( μMACS anti-GFP MACS , Miltenyi Biotec ) . The slurry was passed through a magnetic column , washed 5 times with protein extraction buffer before elution of proteins with hot protein loading buffer . Samples were analyzed by immunoblot ( anti-GFP ) and silver staining prior to MALDI-TOF analysis . For MALDI-TOF , samples were migrated on a 12% mini polyacrylamide gel for about 2 . 0 cm , and rapidly stained with Coomassie blue . Entire gel lanes were excised into 5 equal regions from top to bottom and digested with trypsin ( Promega ) as described [41] , [42] . Data-dependent LC-MS/MS analysis of extracted peptide mixtures after digestion with trypsin was carried out on a hybrid linear trap LTQ-Orbitrap XL mass spectrometer ( Thermo Fisher Scientific ) interfaced to a nanocapillary HPLC equipped with a C18 reversed-phase column ( Agilent Technologies ) . Collections of tandem mass spectra for database searching were generated from raw data with Mascot Distiller 2 . 3 . 2 , and searched using Mascot 2 . 3 ( Matrix Science ) against the 2011_03 release of the UNIPROT database ( SWISSPROT+TrEMBL , www . uniprot . org ) , restricted to Arabidopsis thaliana taxonomy ( 50’756 sequences after taxonomy filter ) . Mascot was searched with a fragment ion mass tolerance of 0 . 50 Da and a parent ion tolerance of 10 ppm . The digestion enzyme trypsin was specified with one missed cleavage . Iodoacetamide derivative of cysteine was specified as a fixed modification . N-terminal acetylation of protein , deamidation of asparagine and glutamine , and oxidation of methionine were specified as variable modifications . The software Scaffold ( version Scaffold_3 . 0 . 9 , Proteome Software Inc . ) was used to validate MS/MS based peptide ( minimum 90% probability [43]] and protein [min 95% probability [44] ) identifications , perform dataset alignment as well as parsimony analysis to discriminate homologous hits . To determine the abundance of 5′ and 3′ ends of selected mRNAs , total RNA was prepared from three independent wild type and vip3zwg replicate samples . Separate cDNA syntheses were performed for each individual gene fragment , followed by qPCRs that were performed as described [36] to detect the respective 5′ and 3′ ends of the transcripts . The following oligonucleotides were used: AT1G01010: GAC AGC TCA ACA CTT TTC CAC TTC and CTT TTA TCC TAA ACA AGA CCC GTA AAG ( 5′ end ) ; GAA CGA AGC ATG TTT GAT TTA TCA TTG and TTG TTG GTG GTT CAT TGG AGT ACA ( 3′ end ) ; AT1G24706: GTT CCT CTC CCT TTT CAT CTT ATC G and CAT CTT AAA CCC CTT TCG TGT GTA T ( 5′ end ) ; GAT TTG CAG ATC CTT TGG TTT GTT C and GCT ATG AAT ATA TCT GAA GTC TGG CAA G ( 3′ end ) ; AT1G64570: CCA TTT ATC GAT TCT TCA CAG ACA CG and GAT TTC ATG ACT CAA ATT AGG GTT CCA ( 5′ end ) ; GAT GCT GAG GAT GAG TAA GTT CCT TC and GCT AGT AAT CTG CAT TCA AAC AGC ACT A ( 3′ end ) ; AT3G02830: CAC TAC CTC TCA CCT CTC TGT TTA CAC and CCA TAG ACG TGA AGA GGA AGA ATG ( 5′ end ) ; GAA GAA ACA AAG GAA GAA GAA GAA GAG and CCA TAG ACG TGA AGA GGA AGA ATG T ( 3′ end ) ; AT5G56860: ATT GAT GAG ATA AAC AAA TGA AGA CAC AAA G and CCA TGT GTG TTT GGC TCG TGT C ( 5′ end ) ; GTT GAT CAG ATC ATC ACA ATA TCC TCA TTA C and GCT ATT AAT TAT CAT ATT AAA CTC TCA CAC ACT CT ( 3′ end ) ; For detection of FLC expression in relation to the EF1 housekeeping gene , qPCR was performed as described [36] using the same oligonucleotide pairs .
|
The production and turnover of messenger RNAs ( mRNAs ) are conserved processes in eukaryotes , from single-cell organisms to plants and mammals . To some degree , this is also true for modulators of these processes , such as the Paf1 and SKI complexes . One particular protein , SKI8 , has been described to have a role in the SKI complex , which influences mRNA stability , both in yeast and in mammals . Moreover , in yeast SKI8 has an additional role in meiotic recombination , whereas in humans it influences mRNA production through association with the Paf1 complex . This functional divergence is commonly thought to arise from differences in protein sequence between the yeast and mammalian SKI8 homologs . Here we show that the conserved SKI8 homolog of the model plant Arabidopsis acts in the SKI complex as well as the Paf1 complex , similar to human . However , using an Arabidopsis ski8 mutant as a tool , we show that yeast SKI8 can fulfill all roles of Arabidopsis SKI8 if introduced into Arabidopsis cells . Thus , it appears that the functional divergence of SKI8 homologs might a priori be related to species-specific cellular context rather than divergence in protein sequence .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology",
"model",
"organisms",
"plant",
"and",
"algal",
"models",
"molecular",
"cell",
"biology",
"plant",
"biology",
"gene",
"expression",
"genetics",
"yeast",
"and",
"fungal",
"models",
"biology",
"genomics",
"evolutionary",
"biology",
"microbiology",
"molecular",
"biology",
"genetics",
"and",
"genomics"
] |
2012
|
Context-Dependent Dual Role of SKI8 Homologs in mRNA Synthesis and Turnover
|
In bacteria the concurrence of DNA replication and transcription leads to potentially deleterious encounters between the two machineries , which can occur in either the head-on ( lagging strand genes ) or co-directional ( leading strand genes ) orientations . These conflicts lead to replication fork stalling and can destabilize the genome . Both eukaryotic and prokaryotic cells possess resolution factors that reduce the severity of these encounters . Though Escherichia coli accessory helicases have been implicated in the mitigation of head-on conflicts , direct evidence of these proteins mitigating co-directional conflicts is lacking . Furthermore , the endogenous chromosomal regions where these helicases act , and the mechanism of recruitment , have not been identified . We show that the essential Bacillus subtilis accessory helicase PcrA aids replication progression through protein coding genes of both head-on and co-directional orientations , as well as rRNA and tRNA genes . ChIP-Seq experiments show that co-directional conflicts at highly transcribed rRNA , tRNA , and head-on protein coding genes are major targets of PcrA activity on the chromosome . Partial depletion of PcrA renders cells extremely sensitive to head-on conflicts , linking the essential function of PcrA to conflict resolution . Furthermore , ablating PcrA’s ATPase/helicase activity simultaneously increases its association with conflict regions , while incapacitating its ability to mitigate conflicts , and leads to cell death . In contrast , disruption of PcrA’s C-terminal RNA polymerase interaction domain does not impact its ability to mitigate conflicts between replication and transcription , its association with conflict regions , or cell survival . Altogether , this work establishes PcrA as an essential factor involved in mitigating transcription-replication conflicts and identifies chromosomal regions where it routinely acts . As both conflicts and accessory helicases are found in all domains of life , these results are broadly relevant .
Transcription is a major impediment to DNA replication . Head-on conflicts arise when a gene is encoded on the lagging strand , prompting transcription in the direction opposite to the movement of the replisome . Conversely , transcription of genes encoded on the leading strand causes co-directional conflicts which occur between replication and transcription complexes moving in the same direction . The significantly faster rate of replication , relative to transcription , leads to the meeting of the two complexes co-directionally . Though the deleterious effects of head-on replication-transcription conflicts have been appreciated for some time , the impact of the less severe , but more common , co-directional conflicts has only recently been established [1–6] . The majority of genes in bacterial genomes are co-oriented with replication [7–11] . This genome co-orientation bias is thought to be a strategy to avoid the more deleterious head-on replication-transcription conflicts . Although the co-orientation bias of bacterial genomes reduces the prevalence of head-on conflicts , by definition , it increases the prevalence of co-directional genes . Importantly , most highly transcribed and essential genes , including rRNA and tRNA operons , are co-oriented with replication [7–9 , 11–15] . Previously identified consequences of co-directional conflicts at rRNA genes include replication stalling and restart in Bacillus subtilis [5] . In Escherichia coli , co-directional conflicts caused by permanently arrested RNA polymerases have been shown to cause double-strand breaks [6] . Cells possess mechanisms that promote replication progression through conflict regions [16] . One strategy is the use of accessory helicases such as E . coli UvrD , Rep , and DinG [17] . Previous work has shed light on the beneficial effects of accessory helicases on replisome progression . There is strong evidence showing that in E . coli these proteins can promote replisome progression through artificially inverted ( head-on ) rRNA genes and that their combined activities contribute to cell survival under these conditions [17] . Additional work showed that mutations in RNA polymerase that impact the expression of rRNA genes ( rpoB* mutations ) can rescue the viability of strains lacking Rep and UvrD , implying that accessory helicases may also function at endogenous co-directionally oriented rRNA genes [18] . However , direct evidence that accessory helicases resolve co-directional conflicts is lacking . The bulk of previous reports have focused on the rRNA genes , except for one instance where inversion of a large region of the chromosome containing several protein-coding genes of both head-on and co-directional orientations also led to growth defects in a rep uvrD double mutant [17] . Though these data implied that protein-coding genes may also produce physiologically relevant replication-transcription conflicts , they did not dissect the contribution , if any , that co-directional conflicts played in the associated growth defect . Because co-directional genes represent at least 50% of most bacterial genomes , and are known to cause replisome stalling in B . subtilis , the additive impact of many co-directional conflicts , especially at rRNA or highly expressed protein-coding genes , may be quite significant . Therefore , clarifying the relative impact of accessory helicases in otherwise identical head-on versus co-directional conflicts should yield insight into their impact on replication . Moreover , identifying the regions where accessory helicases act within the genome should clarify their predominant function ( s ) in the cell . Homologues of the E . coli accessory helicases UvrD and Rep exist in all domains of life . B . subtilis , which diverged from E . coli more than 1 billion years ago [19] , harbors one such homologue , PcrA [20 , 21] . Whereas ΔuvrD and Δrep are only lethal in E . coli in combination , deletion of pcrA alone is lethal in B . subtilis . Currently , the reason PcrA is essential remains unclear , however the lethality of both ΔuvrD Δrep and ΔpcrA strains can be rescued by inactivation of the RecFOR pathway . This complex facilitates the loading of RecA onto single stranded DNA gaps [22 , 23] . PcrA and UvrD can remove RecA from DNA [24] and PcrA depletion strains are hyper-recombinogenic [21] . These findings suggest that the essential nature of these helicases is related to excessive RecA activity . It is unclear whether the RecA removal activity of UvrD is important in the context of conflicts . Additionally , it is unclear whether the conflict resolution activity of UvrD is conserved in PcrA . Several studies have shown that both E . coli UvrD and B . subtilis PcrA interact with RNA polymerase [25 , 26] . However , the physiological significance of these interactions remains unclear , except for a role recently identified for UvrD in transcription-coupled repair [27] . Furthermore , although there is an abundance of in vitro studies on UvrD and PcrA’s helicase and ATPase activities , which are coupled , the physiological relevance of these functions in vivo are poorly understood [24 , 28 , 29] . The physiological significance of PcrA’s RNA polymerase interaction or helicase/ATPase activities is unclear . Additionally , whether these features of accessory helicases are important in conflict resolution is unknown . Here we show that PcrA associates with both head-on and co-directional genes and reduces transcription-dependent replisome stalling at these regions . Using chromatin immunoprecipitations ( ChIPs ) of the replicative helicase DnaC and 2D gel analyses , we were able to detect increased replisome stalling at a single conflict in both the head-on and co-directional orientations when PcrA is depleted . Accordingly , partial depletion of PcrA , which is normally sub-lethal , causes a severe survival defect when a single head-on gene is highly transcribed . Using ChIP-Seq of DnaC and PcrA we identified chromosomal regions where PcrA predominantly associates and impacts replisome stalling . These regions include the heavily transcribed rRNA , tRNA , and other co-directionally and head-on oriented protein-coding genes . Additionally , we found that the helicase/ATPase activity of PcrA , but not its C-terminal RNA polymerase interaction domain , is required for survival in general . Furthermore , although its recruitment is not ablated , a helicase/ATPase mutant of PcrA cannot mitigate conflicts and shows dominant negative effects on replisome stalling at specific transcription units . Altogether , these results identify PcrA as an essential conflict mitigation factor , provide direct evidence for its activity in replication progression through co-directional genes , map the endogenous regions of the chromosome where PcrA routinely resolves conflicts , and establish a correlation between PcrA’s essential function and its helicase/ATPase activity in resolution of conflicts .
To investigate whether PcrA might mitigate replication-transcription conflicts in B . subtilis , we generated a conditional mutant by developing a PcrA degron strain as previously described [30] . In this strain , the C-terminal end of the endogenous pcrA gene is translationally fused to an ssrA degradation tag . At a second locus , we integrated an IPTG-inducible gene encoding the SspB adaptor protein . PcrA is depleted when IPTG is added to the culture , inducing the expression of SspB , which then binds the ssrA tag and delivers PcrA to the ClpXP protease . After treatment of cells with 100 μM IPTG for 15 minutes we observe a 60–90% depletion of PcrA ( Fig 1A ) . Under these conditions cell survival is completely ablated ( Fig 1B ) , indicating that our conditional depletion system functions as expected . To further validate that our PcrA depletion mimics a complete pcrA knockout , we also tested the ability of recF deletion to rescue viability defects of PcrA depleted cells . We found that PcrA depletion in the absence of recF no longer causes viability defects , consistent with previous studies ( Fig 1B , and [21] ) . If PcrA mitigates conflicts , then replisome progression should be hindered in the absence of PcrA . We previously showed that ChIP of the replicative helicase DnaC is sensitive enough to identify replisome stalling and restart at both head-on and co-directional genes [5] . Because the majority of the genes within the genome are co-directional , and we are most interested in the physiologically relevant and naturally occurring conflicts , we again chose to use this technique for our studies . To test the potential role of PcrA in conflict mitigation , we used DnaC ChIPs to measure replisome stalling at two inducible genes: hisC under the lower strength , IPTG-inducible Pspank ( hy ) promoter and lacZ under the strongly expressed Pxis promoter , which is constitutively active or repressed , depending upon the strain ( Fig 2A and 2B ) . These constructs were integrated into the chromosome in either the head-on ( HO ) or co-directional ( CD ) orientation relative to replication . To account for the possibility that local context might affect our experiments , we used two different integration loci: Pxis-lacZ was integrated at thrC ( left chromosomal arm ) and Pspank ( hy ) -hisC was integrated at amyE ( right chromosomal arm ) . These constructs each allow for the direct comparison of otherwise identical co-directional and head-on conflicts . The two loci , and distinct coding genes , also control for potential chromosomal location-dependent and gene sequence-dependent effects . Using ChIP-qPCR we measured the degree of DnaC association with the regions expressing either lacZ or hisC in the two orientations , when transcription is activated or repressed , and in the presence or absence of PcrA . We observed a transcription-dependent enrichment of DnaC with both the head-on and co-directional lacZ and hisC genes relative to a previously established control region , yhaX ( [5 , 31–33] and several other control loci around the chromosome give similar results to yhaX [34] ) ( Fig 2C and 2D ) . Normalization of ChIP data from a region of interest compared to a control locus ( in this case yhaX ) generally provides the most consistent results between experiments . However , we also analyzed non-normalized , raw IP/input values to rule out any potential artifacts of normalization in these experiments . Induction of transcription leads to a significant increase in DnaC association with the head-on lacZ construct regardless of normalization to yhaX , though the absolute degree of enrichment varies between experiments ( S1A Fig ) . Consistent with the higher transcriptional level of lacZ , DnaC association at head-on lacZ was approximately 4-fold higher than at head-on hisC ( Fig 2B , 2C and 2D ) . To measure the impact of PcrA on replisome stalling we carried out DnaC ChIPs in strains harboring the PcrA degron system . After depletion of PcrA , DnaC association increased significantly relative to the conditions where PcrA was present at wild-type levels , with both of the head-on reporters ( Fig 2C and 2D , ( and see S1B Fig ) for non-normalized IP/Input values in the degron experiment ) . Again , the effect of PcrA depletion on replisome stalling was less severe at hisC relative to the lacZ gene , where DnaC association tripled , reaching 55-fold over the control region . PcrA depletion did not affect DnaC association with co-directional hisC , however it caused a small increase in DnaC association—from 2 . 4 to 4 . 0 ( p < 0 . 01 ) with co-directional lacZ . To determine if the increased stalling in the absence of PcrA is due to transcription , after we induced the conflict by de-repressing transcription , we treated cells with 300 μg/ml rifampicin for 3 minutes to inhibit transcription initiation . We observed that DnaC association both before and after PcrA depletion decreased to baseline after rifampicin treatment , indicating that the increased replisome stalling at these conflict regions , without PcrA , is due to replication-transcription conflicts ( Fig 2C and 2D ) . Previous reports have demonstrated that certain proteins may artificially stick to transcription units and produce ChIP artifacts [35 , 36] . To control for the possibility that DnaC could bind non-specifically during ChIPs , we shut off replication for 15 minutes through the addition of HPUra . HPUra is a specific nucleotide analogue that inhibits PolC by inserting into its active site [37] . Under these conditions DnaC association with lacZ drops by approximately 90% , demonstrating that this signal is replication-dependent ( Fig 2D ) . In contrast , a 15 minute HPUra treatment does not reduce the association of RNA polymerase beta subunit ( RpoB ) with this reporter region , as determined with RpoB ChIPs ( S2 Fig ) . Therefore DnaC ChIP signal is unlikely to be an artifact of transcription . To further confirm the results of our ChIP assay we analyzed replication fork stalling at head-on oriented Pxis-lacZ using 2D gels ( Fig 3 ) . 2D gels allow for the direct visualization and analysis of replication intermediates [38] . Restriction digestion ( Fig 3A ) of replicating chromosomes releases branched fragments that generate Y-arcs on 2D gels ( Fig 3B diagram ) . The 2D gels with lacZ fail to reveal any replication intermediates in the absence of transcription ( Fig 3B , top panels ) —with or without PcrA—because replication through the transcriptionally silent region of the chromosome is extremely fast . However , after transcriptional induction an arc of replication intermediates forms , consistent with impaired replication fork movement approaching and entering the lacZ gene . The comparison between the +PcrA and—PcrA gels reveals that the signal intensity is not uniform across the Y arcs , with areas of pausing indicated by locally darkened regions ( Fig 3B lower panels ) . Quantification of the Y-arc demonstrates that the increase in replication fork stalling at the 3`end of lacZ is roughly 1 . 7 ± 0 . 25 following PcrA depletion ( Fig 3C and 3D ) . These results indicate that replication is indeed slowed by the head-on transcription of lacZ , and that depletion of PcrA exacerbates this effect . We also quantified the apparent decrease in signal within the region of the larger Y-intermediates 5`of the initial stall site: Signal in this region drops as low as ~ 0 . 64 ± 0 . 13 when PcrA is depleted ( PcrA-/PcrA+ ) . We note that a digestion intermediate is present on the EagI/ApaLI gels , partly obscuring the lacZ region . These undigested DNAs produced a second Y-arc that is not obscured and showed the same trends displayed in 3B on the right hand side , and 3D ( S3 Fig ) . These data suggest that , absent PcrA , replication forks are highly compromised in their ability to proceed past the initial point of contact with head-on RNA polymerases . They also serve as confirmation of our ChIP experiments which indicated that that DnaC accumulates at the 3`end of head-on genes after PcrA depletion ( Fig 2 ) . It is conceivable that the effects of PcrA depletion on replisome stalling is indirect . If PcrA acts directly at our reporters , it should be physically present there . To address this possibility , we tested the association of an N-terminally Myc-tagged PcrA with our reporters by ChIP-qPCR . After confirming that Myc-tagged PcrA compliments the cell death phenotype observed in the degron ( S4 Fig ) , we performed ChIP in our reporter strains using a monoclonal antibody specific for the Myc peptide . We observed that PcrA associates with both hisC and lacZ reporters after transcription induction ( Fig 4A and 4B ) . Similar to DnaC , PcrA also associates more with both head-on reporters compared to their co-directional counterparts . To further confirm that PcrA association with the conflict region depends on transcription , we treated cells with rifampicin for 3 minutes . Because rifampicin inhibits transcription initiation , the RNA polymerase occupancy within the gene significantly decreases after treatment , thereby preventing the occurrence of replication-transcription conflicts . We found that addition of rifampicin reduced PcrA association with the hisC and lacZ loci to background levels ( Fig 4A ) . These findings suggest that PcrA association with a given conflict region depends on transcription . We set out to identify endogenous chromosomal regions where PcrA promotes replication progression . To globally examine all chromosomal regions , we conducted DnaC ChIP-Seq experiments in the PcrA degron strain ( HM448 ) . After deep sequencing both the total ( input ) DNA used for the IPs , and the DNA recovered from the DnaC ChIPs , we normalized the ChIP sequence reads to the input signal for both the +PcrA ( -IPTG ) and the—PcrA ( +IPTG ) samples by subtracting input from IP signal . We then subtracted the normalized-IPTG signal from the normalized +IPTG signal , identifying the regions where DnaC association increases after PcrA depletion . We found that following PcrA depletion , DnaC association increased significantly at several regions including the rRNA and tRNA genes , and many protein-coding genes of both orientations ( Fig 5A and S5 Fig ) . For comparison , non-normalized DnaC ChIP-Seq data are displayed in S6 Fig . In addition to the rRNA and tRNA genes , the most prominent peaks included the head-on operons dltABCDE , amtB/glnk , ykaA-pit , the head-on ribosomal protein gene rpsD , and the co-directional ribosomal protein genes encoded between rrnW and rrnI ( from 8°-13° ) ( Fig 5A and S5 Fig ) . As expected , activation of the degron system also led to a significant increase in DnaC association downstream of the thrC locus where Pspank-sspB produces an artificial head-on conflict ( Fig 5A , peak H5 ) . Consistent with the observation that PcrA promotes replication across transcription units , we frequently observed an increase in ChIP signal throughout whole genes . Therefore , rather than calling peaks , we found it most appropriate to simply identify genes affected by PcrA depletion . To do this , we calculated the ChIP signal in terms of the maximum signal , or area under the curve within each gene on the chromosome . To account for peaks present in intergenic regions , we performed the same analysis with all intergenic regions greater than 5 nucleotides in length . Genes containing ChIP signal with a maximum height of more than 5-fold over background were considered peak containing regions . A list containing the 50 regions that met these criteria is presented in S1 Table . However , a comprehensive list of all genes and intergenic regions is also included in S2 Table . Interestingly , though replisome stalling was most prominent within genes , we also observed peaks within promoter regions , including the promoters for nagP and qdoI which contain transcriptional repressor binding sites . These data are consistent with reports that PcrA removes DNA binding proteins in addition to RNA polymerase [28] Among the gene regions affected by PcrA activity , approximately 16% ( 8 genes ) are head-on , and 84% ( 42 genes ) were co-directional . These data demonstrate that replisome progression slows within genes of both head-on and co-directional orientation , consistent with DnaC association measurements at the lacZ reporter genes . Previous reports have suggested that ChIP may produce inaccurate data due to non-specific association of target proteins with the rRNA genes . To assess the accuracy of our DnaC ChIP signal at endogenous rRNA genes , we analyzed the formation of replication intermediates at these regions using 2D gels ( S8 Fig ) . When PcrA is present we did not observe any replication intermediates within rRNA genes , despite our use of digest conditions that allowed us to collectively probe for all 10 rDNA repeats simultaneously ( S8 Fig , left side ) . However , after PcrA depletion we clearly observed the formation of replication intermediates within these regions ( S8 Fig , right side ) . This result is consistent with our DnaC ChIP-Seq data and demonstrates that when PcrA is absent , replication slows when it passes through the co-directionally oriented rRNA genes . This strongly suggests that the DnaC ChIP-Seq data is accurate and not an artifact caused by the non-specific adhesion of DnaC proteins to rDNA or other genes . We also considered the possibility that the activity of PcrA at conflict regions could be related to the removal of RecFOR-loaded RecA—a function of PcrA that has been well-characterized in vitro . To test this possibility we carried out the DnaC ChIPs-qPCRs , with and without PcrA , in a ΔrecF background and measured the levels of DnaC association at the most common conflict region we identified—rRNA genes . We found that even when RecF is not present , depletion of PcrA leads to increased DnaC association with rRNA loci ( S9 Fig ) . This result suggests that any effects of RecF related to PcrA activity in conflicts occur either downstream of replisome stalling or are independent of conflicts . In keeping with the results of our reporter assays , we anticipated that on the chromosome PcrA should associate with the regions where replisome stalling increases following PcrA depletion . To identify these regions we conducted ChIP-Seq of Myc-PcrA in strain HM224 ( Fig 5B ) . This strain differs from the PcrA degron strain ( HM448 ) used in Fig 5A in that it does not possess the sspB gene encoded at thrC . To reduce background signal due to non-specific interaction of the Myc antibody with endogenous proteins , we normalized Myc-PcrA ChIP signal to both input DNA and a mock IP with the anti-Myc antibody . We first normalized both the PcrA ChIP and mock IP to their respective input ( total ) DNA data sets by subtracting the input signal from the IP signal at each nucleotide position . Furthermore , because this normalization still produced nonspecific signal ( peaks present in both the experimental and mock IPs ) we also subtracted the mock IP-total signal from the PcrA ChIP-total signal . For comparison , non-normalized and normalized data are shown together in S7 Fig . The resulting data set indicated that PcrA associates predominantly with the rRNA and tRNA genes ( Fig 5A ) . We quantified the data as with the DnaC ChIP-Seq data set , and present a list of peak-containing regions in S3 Table . A comprehensive list of all gene regions can be found in S4 Table . The absence of detectible signal at protein-coding genes identified in the DnaC ChIP-Seq following PcrA depletion suggests that either the association of PcrA with these loci is simply below our detection limit or that its activity at these regions is transient . We set out to determine if PcrA association and activity at the endogenous loci we identified is transcription-dependent . To address this question and to confirm our ChIP-Seq data we analyzed DnaC and PcrA association with different candidate loci relative to the control locus yhaX using ChIP-qPCRs ( Fig 6 ) . The candidate loci included rRNA , tRNA and protein-coding genes of both orientations . Specifically , for genes in the co-directional orientation , we examined the ribosomal RNA gene rrn23S , the Val1-Thr1 tRNA pair , trnSL-Ser1 , and the protein coding gene rplGB . Though we did not detect PcrA association with rplGB in our PcrA ChIP-Seq experiment , DnaC levels increased at this locus following PcrA depletion . Therefore , rplGB was anticipated to represent the lower end of our detection range for PcrA association . For genes in the head-on orientation , we examined the ribosomal protein gene rpsD and two genes from the dltA-E operon , dltA and dltB . ( We also confirmed DnaC association , with and without PcrA , with the genes pit , cotC , and yoaM—see S10 Fig ) . As a negative control , we also investigated the head-on gene yutJ which showed no detectable PcrA or DnaC association . To determine if the association of DnaC and PcrA with the candidate loci was transcription-dependent we used rifampicin to shut off transcription initiation: we treated cells for 3 minutes then analyzed PcrA and DnaC association with these regions . Consistent with the ChIP-Seq experiments , we found that PcrA association with some coding genes was low in ChIP-qPCR experiments ( Fig 6A ) . However , regardless of the degree of association , transcription shut-off reduced association of PcrA with all examined loci to some degree , except at the negative control locus , yutJ ( Fig 6A ) . Also in agreement with our ChIP-Seq data , in the ChIP-qPCR experiments , DnaC association increased after PcrA depletion at all chromosomal loci examined , with the exceptions of the single tRNA gene trn-SL-ser1 , dnaK and yutJ ( Fig 6B , S10 Fig and see S11 Fig for non-normalized IP/Input values for DnaC association with the rRNA genes ) . Presumably , a single tRNA gene may simply be so short that local RNA polymerase occupancy remains limited , thereby minimizing the impact on replication . Regardless , we do observe stalling at a locus encoding multiple tRNA genes ( Val1-Thr1 ) . As with the PcrA ChIPs , after rifampicin treatment and in the absence of PcrA , DnaC association with the other loci was ablated ( Fig 6B ) . These data confirm that PcrA’s association and activity at conflict regions requires active transcription as initially indicated by our reporter data . To determine if the association and activity of PcrA was correlated with transcription , we took a second approach: we measured RNA polymerase occupancy at the loci identified in the ChIP-Seq experiments ( as a measure of transcription level ) by conducting ChIP-qPCRs of a GFP-fusion allele of the beta’ subunit of RNA polymerase , RpoC , using an anti-GFP polyclonal antibody . We found that RpoC associates at predictably varying degrees with all but the negative control candidate regions tested ( Fig 6C ) . As anticipated , rifampicin treatment reduced this association at all examined loci by 80% or more ( Fig 6C ) . To control for potential artifacts of the GFP-fusion we carried out ChIP-qPCRs of the beta subunit of RNA polymerase , RpoB , using a native antibody . Although the absolute degree of association of RpoB compared to the RpoC ChIPs was different , the relative association patterns with conflict regions were equivalent ( S12 Fig ) . RpoC ChIP-qPCRs allowed us to compare RNA polymerase occupancy with PcrA association and conflict severity ( DnaC association ) . RpoC , DnaC and PcrA associations closely correlated with all co-directional genes examined ( Pearson coefficient 0 . 7 for DnaC ( -PcrA ) vs . RpoC , and 0 . 9 for PcrA vs . RpoC ) . Among the head-on genes replisome stalling in the absence of PcrA correlates with PcrA occupancy ( Pearson coefficient 0 . 89 for DnaC ChIP—PcrA vs . +PcrA ChIP ) and is highest at the gene with the highest RNA polymerase occupancy , rpsD . This correlation suggests that conflict severity is related to transcription levels for head-on genes . We also find comparison of the co-directional and head-on genes to be informative: replisome stalling is similar between dltA , dltB and the co-directional rplGB gene despite the significantly lower transcription levels for dltA and dltB ( roughly 3 fold lower RpoC association compared to rplGB ) . Similarly , replisome stalling at rpsD ( head-on ) is equivalent to levels at the Val1-Thr1 tRNA genes ( co-directional ) despite an approximately 3 fold lower RpoC association with rpsD . Therefore , these data are consistent with our reporter experiments where we observed that a head-on conflict causes significantly more replisome stalling than an equivalent co-directional conflict ( Fig 2C and 2D ) . Furthermore , the results of our genome-wide analyses , together with the results of the engineered conflict experiments , underscore the potent effects of head-on relative to co-directional conflicts . Although the signals observed in these experiments are relatively small , altogether , the consistency between the data from the various experiments shows that PcrA resolves both head-on and co-directional conflicts genome-wide . Highly expressed co-directional genes are common in the genome . Therefore , expression of an additional co-directional gene should not have a major effect on cell viability . However , since the number of highly expressed head-on genes is limited during fast growth , we wondered if the addition of a highly expressed head-on gene would increase the sensitivity of cells to PcrA depletion . To test this hypothesis , we measured viability before and after partial PcrA depletion in cells harboring the lacZ reporters in both orientations . In the presence of PcrA , transcription of lacZ had no effect on cell survival regardless of its orientation ( Fig 7A , no IPTG ) . However , following partial PcrA depletion with 2 μM IPTG we observed slow growth in cells with the co-directional lacZ gene and cells harboring the repressed head-on lacZ reporter ( Fig 7A , 2 μM IPTG , and 7B ) . Expression of co-directional lacZ did not cause a decrease in plating efficiency , suggesting that a single additional co-directional conflict does not have a major effect on replication or growth rate . However , expression of head-on lacZ caused a severe decrease in plating efficiency after an otherwise non-lethal degree of PcrA depletion ( Fig 7B ) . Strains harboring the hisC reporters ( which are incorporated at different regions on the chromosome and are expressed under a different promoter ) showed a similar asymmetric effect on plating efficiency ( S13 Fig ) . These results suggest that mitigation of severe conflicts by PcrA is essential for viability . The mechanism ( s ) allowing accessory helicases to be recruited to conflict regions have not been defined . PcrA could be recruited to either the replication fork or to RNA polymerase during a conflict . Previous work has established that PcrA interacts with RNA polymerase [25 , 26] . Whether this interaction is important for its role in conflict mitigation is unknown . To address this question , we produced a mutant allele of myc-pcrA shown to dramatically reduce PcrA’s RNA polymerase association in vitro , in the closely related species , Geobacillus stearothermophilus: myc-pcrA-ΔC [25] . This mutant lacks the final 71 amino acids of PcrA’s C-terminal domain . To avoid the potential problem of lethality , or accumulation of suppressor mutations , we expressed this mutant conditionally by placing it under the control of an IPTG-inducible promoter . When expressed in cells already harboring the PcrA degron system , the addition of IPTG triggers the simultaneous depletion of PcrA-ssrA and induction of myc-pcrA-ΔC . We found that the PcrA-ΔC protein had wild-type level activity in preventing replisome stalling and association with conflict regions of both orientations ( Fig 8A and 8B ) . Furthermore , this mutant completely rescued the viability of PcrA degron strains , indicating that RNA polymerase interaction through the C-terminal domain of PcrA is not essential for its conflict mitigation activity ( Fig 8C ) . It is possible that this mutant does not completely ablate PcrA’s interaction with RNA polymerase , as an N-terminal interaction between PcrA and RNA polymerase has also been detected [39] . However , based on previous work , we expect this disruption to at least partially reduce PcrA’s association with RNA polymerase . The complete lack of a phenotype in conflict mitigation and survival in strains harboring this mutant suggests that its RNA polymerase interaction is not required for PcrA’s association or activity at conflict regions . In vitro studies of PcrA have demonstrated that its helicase and ATPase activities are required for its ability to separate DNA strands , but are dispensable for RecA removal . To determine if PcrA’s helicase/ATPase activities are required for its conflict mitigation functions and viability , we constructed a previously characterized separation of function allele , myc-pcrA K37A Q254A ( PcrA H- ) which is defective in helicase/ATPase activity [24] . By expressing this allele in a strain already harboring the conditional degron system ( as discussed above ) we were able to use ChIP-qPCR to determine whether this mutant is capable of mitigating head-on and co-directional conflicts , and measure its association with conflict sites ( Fig 8 ) . We found that the PcrA H- allele failed to resolve both head-on ( dltB ) and co-directional ( rrn23S and rplGB ) replication-transcription conflicts and that conflict severity at these regions was exaggerated in the presence of this mutant ( Fig 8A ) . The inability of PcrA H- to resolve conflicts does not reflect an inability to associate with conflict sites , as we actually observed increased association of this mutant with rrn23S and rplGB ( Fig 8B ) . This increase could potentially reflect a reduced ability to release from the DNA due to a defect in ATP hydrolysis [40] . Though we did not observe a significant association of either wild-type or the PcrA H- mutant protein at dltB , this result is not entirely surprising given the low PcrA ChIP signal we previously observed at these loci and the further decreased overall ChIP signal in these experiments . Nevertheless , the effect of PcrA H- on replisome stalling at dltB indicates that it is active at this location . To determine if the helicase/ATPase activities of PcrA are important for viability , we carried out plating efficiency assays . Here we observed that cells depleted of the wild-type PcrA and expressing the PcrA H- allele are inviable ( Fig 8C ) . Because this mutant is capable of removing RecA and preventing RecA-dependent strand exchange in the closely related species S . aureus [24] , the loss of viability in the strain harboring the PcrA H- protein may not be due to inability to remove RecA .
There are at least two models that could potentially explain how PcrA associates with the site of a conflict . Based on data from E . coli regarding the interaction and movement of Rep with the replication fork , it is conceivable that PcrA is also recruited to conflict regions via interactions with the replication fork [41] . On the other hand , reports also indicate that PcrA interacts with RNA polymerase , suggesting that PcrA may be recruited to the conflict site through this association . In our system , removing the C-terminal domain of PcrA , which prevents detectable association with RNA polymerase in vitro , did not impact its function in conflicts [25] . Because the data suggest that PcrA is recruited to conflict regions independent of its interaction with RNA polymerase , recruitment via an interaction with the replisome seems more plausible . There are at least two models for PcrA activity at conflict regions . PcrA may directly remove either RNA polymerases or RecA bound to single stranded DNA ahead of the replication fork . These two possibilities are not mutually exclusive . We found that the helicase/ATPase activities of PcrA , which facilitate strand separation , are required for conflict mitigation . This result was not necessarily expected given that mutants lacking these activities retain the ability to efficiently remove RecA from DNA in related species [24] . ( As the potential RecA removal activity of PcrA H- has not yet been demonstrated in B . subtilis or in vivo , the following analysis is predicated on the assumption that this in vitro activity of S . aureus PcrA is relevant to living B . subtilis cells . ) If PcrA’s role in replisome progression through transcription units stems from direct removal of RecA , then a mutant defective in helicase/ATPase activity should have , at least partially , retained the ability to mitigate conflicts . However , PcrA H- is defective in both conflict mitigation and survival . Therefore , we propose that PcrA may not mitigate conflict severity via the direct removal of RecA . In addition , the inability of this mutant to support life suggests that the rescue of ΔpcrA strains by inactivation of the RecFOR pathway stems from an activity of PcrA that is upstream of RecA recruitment ( i . e . PcrA indirectly prevents excessive RecA recruitment ) . Given the correlation between decreased viability and increased conflict severity , we suspect that the essentiality of PcrA is due specifically to its activity in conflict mitigation . We prefer a model in which PcrA clears RNA polymerases ahead of the replisome , and thereby prevents excess single stranded DNA formation and subsequent RecA binding at conflict regions . Previous studies comparing the two types of conflicts have shown that head-on conflicts are far worse than co-directional conflicts . Furthermore , a number of studies have suggested that accessory helicases reduce the deleterious impact of transcription on replisome progression in vivo and in vitro . In vitro work had previously suggested that B . subtilis PcrA , similar to UvrD and Rep , promotes replisome progression past a single protein block [41] . Furthermore , PcrA can complement the survival deficiencies of UvrD and Rep mutants [20 , 41] . Although these studies suggested that PcrA acts similarly to UvrD and Rep , the role of PcrA in conflict resolution was not directly shown prior to this study . Furthermore , the endogenous chromosomal regions where UvrD , Rep or PcrA act have not been reported . Here , we provide direct evidence for PcrA activity in conflict mitigation and identify for the first time the natural targets of PcrA . These include , as we and others suggested , naturally occurring co-directionally oriented rDNA . Interestingly , we also identified specific head-on and co-directional genes that were not necessarily predictable . Previous studies investigating the role of accessory helicases in conflict mitigation took advantage of severe conflicts caused by artificially inverted ( head-on ) rDNA , which causes major survival defects . However , the interpretations of these studies regarding conflicts may be complicated by the unique properties of rRNA genes such as GC richness , RNA polymerase stabilizing anti-termination proteins , and secondary DNA structures . Our experiments using lacZ , hisC , and several endogenous coding genes , circumvent these potential complications . Also , by detecting effects on replisome stalling at an otherwise identical gene ( i . e . hisC or lacZ ) in the two orientations , we can estimate the relative impact of gene expression and orientation on replisome progression: when PcrA is present , transcription-dependent replication stalling increases roughly by 6-fold at the head-on lacZ construct compared to its co-directional counterpart . However , in the absence of PcrA , this differential reaches more than 20-fold . Unfortunately , because these results are gathered from ensemble assays , it is difficult to further quantify the severity or frequency of replisome stalling in single cells or during a round of replication . Future experiments examining the impact of conflicts on replisome progression in single cell studies could potentially answer these questions . Bacterial genomes are generally organized such that highly transcribed and essential genes are oriented co-directionally with respect to DNA replication . In the case of rRNA genes , co-orientation is essentially universal . Though co-orientation reduces conflict severity , replication-transcription conflicts still occur in these regions to some degree . Our observations highlight the importance of co-directional conflicts in vivo , despite the apparent lack of effect on replisome progression in vitro [42] . Though any one co-directional conflict may not severely inhibit replication , the abundance of co-directional conflicts suggests that collectively , they can significantly slow replication . This consideration is especially important given that co-orientation appears to be a major strategy cells use to reduce conflict severity; in doing so , cells increase the impact of co-directional conflicts . Although similar conflict mitigation mechanisms seem to exist in both B . subtilis and E . coli , there is a significant difference in how the two species tolerate head-on oriented rRNA genes [3 , 43 , 44] . Furthermore , the genome co-orientation bias in the two organisms is significantly different: there are many more head-on genes in E . coli compared to B . subtilis ( 45% vs . 26% , respectively ) . Together , the rDNA inversion experiments and the genome co-orientation biases from these two different species suggest that E . coli cells are much more tolerant of conflicts than B . subtilis . In keeping with these differences , studies in E . coli were unable to detect replication intermediates in rRNA genes , even after accessory helicase deletion , or rDNA inversion ( replication intermediates only formed in inverted rDNAs after accessory helicase deletion ) [17] . However , we observed replication intermediates in naturally occurring , co-directionally oriented rRNA genes following PcrA deletion in B . subtilis . What is the reason for the higher conflict tolerance of E . coli ? Since multiple E . coli accessory helicases can be deleted without detectibly slowing replication through the rDNA , it seems likely that either E . coli possesses additional , as yet unidentified , conflict mitigation mechanisms , or that the replication machinery is inherently less susceptible to stalling at transcription units in E . coli .
For all experiments , B . subtilis 168 cells were plated on LB supplemented with the corresponding antibiotics . Single colonies from plates were used to inoculate cultures of liquid rich medium ( Luria-Bertain ( LB ) ) . Liquid cultures were grown to mid-log at 30°C , shaking at 260 r . p . m . , then diluted back to OD 0 . 05 , and grown again to OD 0 . 3–0 . 35 before harvesting . For Rifampicin treatments , 30 mg/ml Rifampicin in DMSO was added to a final concentration of 0 . 3 mg/ml , for 3 minutes . PcrA Degron cultures were grown as above , with the exception that cells were split at OD 0 . 2 into two cultures , with or without IPTG at 100 μM final concentration . At OD 0 . 3–0 . 35 , cells were harvested . B . subtilis cultures were grown to OD 0 . 3 , then treated with 0 . 2% NaAzide to arrest growth . 20 mg of cells were then suspended in low-melt agarose plugs ( 0 . 5% ) as previously described [45] . Lysis was performed in 2 mg/ml lysozyme for 16 hours at 37°C . Protein was removed via incubation with 5 mg/ml proteinase K , 5% sarkosyl , 0 . 5 M EDTA for 4 hours at 37°C . Proteinase K was then removed by 8 successive 4 hour washes in TE at 4°C . DNA was digested overnight in plugs equilibrated in 1x CutSmart buffer plus a 0 . 5 μl of each of the indicated enzymes ( NEB ) . DNA was subjected to 2-dimensional electrophoresis and Southern blotting as previously [46] . Probes for Southern blots were generated via random priming of gel-extracted PCR products corresponding to the lacZ region or the rrn16S-23S rRNA regions . Probes were radioactively labelled using α-32P-dATP . Hybridization images were generated on X-ray film or with phosphor screens ( GE Healthcare ) . Y-arcs were quantified by conforming a single lane to the shape of the entire arc in ImageQuant software ( GE Healthcare ) . This yielded a histogram consisting of approximately 450 data points along the arc . Strains used in this publication are listed in Table 1 . The ImmR protein , which is coded for by the mobile genetic element ICEBs1 , tightly represses Pxis [47] . We introduced the Pxis-lacZ constructs into strains that either harbored ( Trx- ) or were cured of ( Trx+ ) ICEBs1 . To produce the PcrA degron strains , genomic DNA from strain HM448 was transformed into strains harboring the Pxis-lacZ constructs prior to selection on MLS for the Pspank-sspB construct . The pcrA-ssrA allele from HM448 was then transformed into the resulting strains as a second transformation . Single colonies were grown in liquid LB and grown to OD 0 . 5 , then serially diluted at a 1:10 ratio prior to plating . Polyclonal rabbit anti-DnaC antibodies were used for ChIP of native DnaC [5 , 48] . Mouse monoclonal anti-Myc antibody purchased from Invitrogen was used for anti-Myc-PcrA ChIP experiments ( Product Number 13–2500 ) . Polyclonal rabbit anti-GFP antibodies were used for RpoC-GFP ChIP experiments . DNA samples for ChIP were prepared essentially as previously described [5 , 48]: Bacillus subtilis cells were grown in LB medium as described . Cells were crosslinked with formaldehyde at a final concentration of 1% v/v . Following 20 minutes of incubation at room-temperature , reactions were quenched with glycine , and cells were pelleted , washed once in 1x PBS , pelleted again , then frozen at -80C . Pellets were re-suspended in 1 . 5 ml solution A ( 20% sucrose , 50 mM NaCl , 10 mM EDTA , 10 mM Tris pH 8 . 0 ) , plus 1 mg/ml lysozyme , and 1 mM AEBSF . Following a 30 minute incubation at 37°C , lysates were sonicated on a Fisher sonic dismembrator ( Fisher FB120 ) for 40 seconds ( 10 seconds on , 10 seconds off ) , at 30% amplitude . Lysates were spun at 8k rpm for 15 min at 4°C in microcentrifuge tubes , and the supernatant cell extract was transferred to fresh tubes and frozen at -80C . ChIP was performed by adding 12 μl anti-Myc or 1 μl anti-DnaC antibody to 1 ml aliquots of extract , then incubating over-night at 4°C in an end-over-end nutator . Antibody-bound protein:DNA complexes were precipitated using protein A sepharose beads ( GE 45000143 ) , decrosslinked over-night at 65°C , then purified by phenol:chloroform extraction and ethanol precipitation . qPCR analysis was performed on a Bio-Rad CFX connect ( Product Number 1855201 ) using Sso Advanced SYBR green master mix ( product number 1725262 ) . Primer pairs include HM192 ( 5`-CCGTCTGACCCGATCTTTTA-3` ) and HM193 ( 5`-GTCATGCTGAATGTCGTGCT-3` ) which amplify the low conflict region yhaX , HM80 ( 5`-AGGATAGGGTAAGCGCGGTATT ) and HM81 ( 5`-TTCTCTCGATCACCTTAGGATTC-3` ) which amplify the rrn23S repeat of rRNA gene repeats , HM766 ( 5`-GCT GGG AGA GCA TCT GCC TT-3` ) , HM767 ( 5`-CCAACCTACTGATTACAAGTCAGTTGCTCTA-3` ) which amplify between Threonine and Valine tRNA genes at 4 repeats , HM892 ( 5`-CATGAAAAAGCTCGGCAAAG-3` ) and HM893 ( 5`-TGGAATCTTACGCAAAAACAAA-3` ) which amplify within the rplGB gene , HM803 ( 5`-TGTTTTGCGGAGAGGTTCTT-3` ) and HM804 ( 5`-CGGGCCGTACGTATTAAAAA-3` ) which amplify within the dltA gene , and HM902 ( 5`-CGGGGTCAGCTACATTATGG-3` ) , and HM903 ( 5`-AGACATATGCCAGCGATTCC-3` ) which amplify within the dltB gene . HM770 ( 5`-TCTCCAGCTGTGATAAACGGTA-3` ) and HM771 ( 5`-AAAACGGCATTGATTTGTCA-3` ) which amplify within the dnaK gene , HM952 ( 5`-GGTGTAAACGAACGTCAATTCCGCAC-3` ) and HM953 ( 5`-AGCTTGTACACAACGTTATCAAGACGAGAATC-3` ) which amplify within the rpsD gene , and HM954 ( 5`-GAAGAAAAAGTGAATGAGCTGCTGAAGGAA-3` ) and HM955 ( 5`-AATGTCTTCGCTCTCAAAAAACTCAATCAAACG-3` ) which amplify within the yutJ gene . qPCR analysis was conducted for both yhaX and test DNA species in both input and ChIP samples . Final fold enrichment was calculated as ( Test DNA in ChIP Sample/Test DNA in Input Sample ) / ( yhaX in ChIP sample/yhaX in Input sample ) . ChIP DNA samples were analyzed first using qPCR to validate that a given sample is representative . DNA samples were then processed and sequenced by the University of Washington High Throughput Sequencing Genomics Center , on an Illumina Next-Seq . Approximately 750k paired-end Illumina Next-Seq reads per sample were mapped against the genome of B . subtilis strain JH642 ( GenBank: CP007800 . 1 ) using Bowtie 2 with the—no-mixed option . This option prevents unpaired alignments , such that only reads that aligned uniquely at both ends were mapped [49] . As discordant mapping was minimal and did not significantly alter the resulting profile , discordant mapping was active . The resulting . sam file was processed by SAMtools , view , sort , and mpileup functions [50] , to produce wiggle plots . We tested the effect of removing PCR-based and optical duplicates using Picard v1 . 3 and found that the same gene regions were identified in subsequent analyses , but at a slightly lower signal:noise ratio . Therefore , in the presented data sets , duplicates were not removed . Myc-PcrA ChIP-Seq data and antibody control data were first normalized to input samples ( signal in the ChIP sample minus the signal in the corresponding input sample ) . Normalized antibody control IP ( Mock IP ) data representing non-specific signal enrichment was subtracted from the normalized ChIP signal . For DnaC ChIP-Seq , ChIP samples were first normalized to inputs ( ChIP minus input ) . The normalized + PcrA DnaC ChIP sample signal was then subtracted from the normalized—PcrA DnaC ChIP signal at each nucleotide position , establishing the differential signal . ChIP-Seq data were quantified as follows: for each gene and intergenic region , the maximal signal , average signal , area under the curve , and area under the curve divided by total gene length ( normalized area under the curve ) were calculated . Local background was independently determined for regions proximal to oriC ( the 0–300k nucleotide region , where background signal is slightly higher due to higher chromosomal copy number ) , or distal to oriC , by calculating the average maximal signal in ~100 kb regions that were devoid of peaks . Genes containing a maximum signal of more than 5-fold above background were called as peak-containing regions .
|
In bacteria the concurrence of DNA replication and transcription leads to potentially deleterious encounters between the two machineries . These encounters can destabilize the genome and lead to mutations . Both eukaryotic and prokaryotic cells possess conflict resolution factors that reduce the detrimental effects of these collisions . In this study we show that without the essential Bacillus subtilis accessory DNA helicase , PcrA , the replication machinery slows down at certain regions of the chromosome in a transcription-dependent manner . PcrA is essential to life but incomplete depletion of PcrA only partially inhibits cell survival . We find that , under these conditions , partial survival defects are significantly exacerbated in the presence of a single severe conflict . In summary our work identifies a high degree of conservation for accessory helicase function in conflict resolution , directly establishes PcrA’s role in co-directional conflict resolution , and maps the natural chromosomal regions where such activities are routinely needed . Because both conflicts and accessory helicases are found in all domains of life , the results of this work are broadly relevant .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
The B. subtilis Accessory Helicase PcrA Facilitates DNA Replication through Transcription Units
|
Aedes albopictus is a highly invasive disease vector with an expanding worldwide distribution . Genetic assays using low to medium resolution markers have found little evidence of spatial genetic structure even at broad geographic scales , suggesting frequent passive movement along human transportation networks . Here we analysed genetic structure of Aedes albopictus collected from 12 sample sites in Guangzhou , China , using thousands of genome-wide single nucleotide polymorphisms ( SNPs ) . We found evidence for passive gene flow , with distance from shipping terminals being the strongest predictor of genetic distance among mosquitoes . As further evidence of passive dispersal , we found multiple pairs of full-siblings distributed between two sample sites 3 . 7 km apart . After accounting for geographical variability , we also found evidence for isolation by distance , previously undetectable in Ae . albopictus . These findings demonstrate how large SNP datasets and spatially-explicit hypothesis testing can be used to decipher processes at finer geographic scales than formerly possible . Our approach can be used to help predict new invasion pathways of Ae . albopictus and to refine strategies for vector control that involve the transformation or suppression of mosquito populations .
The Asian Tiger Mosquito , Aedes albopictus , is one of the world’s most dangerous invasive species ( Global Invasive Species Database , http://www . issg . org/database/ ) . It has been implicated in recent outbreaks of chikungunya [1] and dengue [2] in the tropics and even temperate regions [3 , 4] , and its hyperaggressive diurnal bloodfeeding reduces the quality of human environments [5 , 6] . Minimising the negative impact of Ae . albopictus with pesticides has proven increasingly difficult due to insecticide resistance , particularly among populations within the native distribution of this species [7–10] , which extends from India to South East Asia to Japan [11] . Modification of mosquito populations using the endosymbiotic bacterium Wolbachia has been proposed as a viable alternative to pesticide-based population suppression [12] . This study considers an urban population of Ae . albopictus in Guangzhou , China , that has undergone releases of Wolbachia-infected males to suppress populations at two locations since March 2015 . Guangzhou region is within the native distribution of Ae . albopictus , and experiences occasional dengue outbreaks despite local absence of the primary vector of dengue , Ae . aegypti [10] . Aedes albopictus females in Guangzhou are normally superinfected with two Wolbachia strains , wAlbA and wAlbB [13] , and Wolbachia-based population suppression has been conducted by releasing males that carry an additional strain , wPip , that has been transferred from Culex pipiens ( Diptera: Culicidae ) [14] . This population suppression effort has been successful , with population reductions of more than 90% maintained over two years ( Xi pers . comm . ) . However , whether Wolbachia releases are able to suppress Ae . albopictus populations sustainably over a large geographic region depends on the frequency at which suppressed sites are reinvaded by mosquitoes dispersing into the sites from outside . This reinvasion determines the frequency and intensity at which releases need to be repeated to maintain population suppression . To estimate the susceptibility of a region to reinvasion an understanding of basic dispersal parameters in Ae . albopictus is needed . Dispersal in Ae . albopictus is generally thought to represent a combination of short-range active flight of adults and long-range passive movement of immature stages along human transportation networks [11 , 15–18] . The success of Ae . albopictus as an invasive species likely reflects its ability to disperse along human transportation networks such as shipping routes [11] and roads [15] . This has produced little genetic differentiation at broad scales throughout its range [19–23] , with most genetic variation observed within sampling sites [16 , 24] . For an invasive species experiencing founder effects this pattern of genetic structuring may be unsurprising , however , Ae . albopictus populations also show low differentiation within their native range [17 , 18 , 25 , 26] . Some studies in recently invaded regions have reported mild genetic structure [24 , 27] , though the spatial dependence of this has not been tested . Although negligible genetic differentiation has been observed between Ae . albopictus samples within countries [18 , 19] and cities [28] , structure at these scales is regularly observed in Ae . aegypti , the primary dengue vector [29–33] . This seems incongruous given that the estimates of the flight range potential in Ae . albopictus are comparable to those in Ae . aegypti ( 100m–500m: [11 , 34–38] . At finer scales we expect active dispersal to have a stronger influence on genetic structure and contribute to patterns of spatial genetic structure such as isolation by distance ( IBD [39] ) . However , at broader scales that exceed species’ active dispersal capacity , a relationship between spatial distance and genetic distance such as IBD weakens and genetic structure is driven more by patterns of passive dispersal such as via trade or traffic routes . Because strong genetic structure among Ae . albopictus populations has yet to be observed at any scale , passive dispersal has been proposed as the most important determinant of gene flow in this mosquito [40] . The lack of observed genetic differentiation at finer spatial scales may be an artefact of using genetic markers that lack the power to detect subtle genetic patterns [16] . Analyses of Ae . aegypti at genome-wide single nucleotide polymorphisms ( SNPs ) discovered through double-digest restriction-site associated DNA sequencing ( ddRADseq [41] ) have provided a wealth of new ecological information at fine scales [31–33 , 42 , 43] , including direct evidence of passive dispersal and spatial genetic structure at distances < 4km [33] . Until now , genetic patterns in Ae . albopictus have been studied using allozymes , mtDNA , and microsatellites ( [18–23 , 40]; see review in Goubert et al . [16] ) , that have a much lower resolution than genome-wide SNPs in deciphering genetic patterns in Ae . aegypti [33 , 43] . In this study , we used ddRAD sequencing to genotype Ae . albopictus from Guangzhou , China at genome-wide SNPs to test explicit hypotheses about the processes shaping the genetic structure within the city . Specifically , we determined the relative effects on genetic structure of IBD , passive dispersal along human transport networks , and barriers to dispersal . We sampled Ae . albopictus in sites closely situated among networks of human transportation to test hypotheses of the contribution of active and passive dispersal on genetic structure . We also investigated whether human transportation networks such as highways or rivers could act as barriers to dispersal . Highways have been observed to restrict dispersal in Ae . aegypti [33 , 44 , 45] and , given that Ae . albopictus tends to prefer sylvan , rural and suburban habitats over urban ones [11 , 46] , large urban highways such as those in Guangzhou may pose particularly effective barriers to mosquito dispersal . There is considerable evidence for passive dispersal along human transport networks and no evidence of IBD within countries [15 , 16 , 28 , 47 , 48] , and our study is the first to employ high-density genomic markers to investigate gene flow within a city . This paper describes how with these markers we were able to find evidence for both IBD and passive dispersal along human transportation networks , but no evidence for dispersal barriers .
Aedes albopictus were collected from 12 sites on public land in Guangzhou , China , between September 23rd and October 22nd , 2015 . The sample sites were distributed across an area of 380 km2 through nine administrative districts , and were separated by distances ranging from 2 . 73 km to 49 . 34 km ( Fig 1 ) . All subsequent references to sites follow the numerical designations in Fig 1 . At each site , natural containers were searched for larvae and pupae that were collected and then combined into a single sample to be reared in the laboratory until eclosion . Samples from Sites 1–8 were raised in tap water containing shrimp powder , while samples from Sites 9–12 were raised in tap water containing shrimp powder , bovine liver powder and nutritional yeast . Following eclosion , adults were fed on 10% sugar solution and allowed to mate , but not bloodfeed . After 4–6 days , 20 females from each sample were killed by freezing and stored in ethanol until DNA extraction . Genomic DNA was extracted using Roche DNA Isolation Kit for Cells and Tissues ( Roche , Pleasanton , CA , USA ) , with an additional step of RNAse treatment . From the 12 sample sites , we selected 152 individuals for ddRAD sequencing . Analyses of population structure are generally biased when closely-related individuals such as full-siblings are included in the analyses [59] . To identify full-sibling relationships among our samples , we calculated Loiselle’s k [60] using the program SPAGeDi [61] . First and second-degree kin relations can be ascertained with confidence using large SNP datasets [33 , 41 , 62–64] , wherein individuals with pairwise k > 0 . 1875 are full-siblings , and those with 0 . 1875 > k > 0 . 09375 are half-siblings , if each pair is assigned the most likely kinship category [65] . We sampled one individual from each putative full-sibling group with the smallest percentage of missing data , leaving 116 individuals for analysis of genetic structure . We were also interested in observing related pairs at different sampling sites . Larval Ae . aegypti full-siblings have been caught in traps up to 1 . 3 km apart [33] , well above the active dispersal range of Ae . albopictus [11 , 34–37] . As the minimum distance between sites in this study was more than double this distance , movement at these scales would most likely also be human-mediated . To avoid incorrectly identifying the relationship of closely-related pairs we used the program ML-Relate [66] to perform specific hypothesis tests of relationship . For each pair we ran one test that estimated the relationship assuming that the kinship category assigned using k was more likely than the next most likely kinship category , followed by tests that assumed that the kinship category assigned using k was less likely to be correct . Thus , for pairs with k > 0 . 1875 , tests would determine whether the pair were full-siblings or half-siblings , while for pairs with 0 . 1875 > k > 0 . 09375 tests would help determine whether the pair were full-siblings , half-siblings or unrelated . Tests were run using 10 , 000 , 000 simulations of random genotype pairs for each .
Comparisons across sample sites revealed four probable full-sibling kinship groups ( 0 . 203 ≤ k ≤ 0 . 474 , 0 . 10% of pairs ) and a single probable half-sibling group ( k = 0 . 156; 0 . 02% of pairs ) with members spread between two sample sites 3 . 657 km apart ( Fig 2 ) , Jiuwei Village and Zhucun Village ( Sites 8 and 9 ) . This distance is much greater than flight range estimates of Ae . albopictus [34–37] , which suggests that this dispersal likely reflects passive transportation by humans . Specific tests of relationship using maximum-likelihood estimation found that , of the nine putative full-sibling pairs found across sampling sites , six ( k > 0 . 258 ) were definitely full-siblings ( P < 0 . 001 ) and not half-siblings ( all P > 0 . 052 ) , while the remaining three ( 0 . 224 < k < 0 . 248 ) could be of either relationship category . Sites 8 and 9 were connected by highways which may facilitate passive dispersal between them . Out of 854 comparisons between individuals sampled within the same site , 25 pairs ( 2 . 93% ) were full-siblings ( 0 . 200 ≤ k ≤ 0 . 517 ) and 7 pairs ( 0 . 82% ) were half-siblings ( 0 . 105 ≤ k ≤ 0 . 183 ) . A histogram of k scores showed fewer scores near k = 0 . 1875 , the full-sibling/half-sibling category boundary , with the number of full-sibling pairs increasing sharply with k ( Fig 2 , inset ) . We did not observe the same pattern at the half-sibling/unrelated category boundary , but the most distantly related half-sibling pair had a k score 31 . 2% larger than the most genetically similar pair of unrelated individuals . The AMOVA showed significant genetic structuring among sampling sites ( P < 0 . 001 ) , accounting for 8 . 1% of the observed variation ( η2 = 0 . 081 ) . A simple Mantel test on matrices of individual genetic distance and log-transformed geographic distance showed a small but significant correlation ( Table 1 , r = 0 . 062 , P < 0 . 001 ) . Mantel tests for other landscape variables showed that “Port Isolation” , “River Isolation” and “Highway Barriers” were all better predictors of genetic distance than geographical distance . ( Table 1 ) . Neither “Port Isolation” nor “River Isolation” showed any correlation with distance , while the other variables all did . Mitochondrial DNA haplotype variants were broadly distributed throughout the study area ( Fig 4 ) . A single common haplotype was found at all 12 sites and two others were found at seven and three sites respectively . No other haplotypes were found at more than one location . Haplotype variants were separated from the common haplotype by 4 polymorphic sites at most , representing a divergence of 0 . 024% . Analyses demonstrating an overall lack of spatial clustering among mitochondrial haplotypes are described in S2 Table and S1 Fig . Most individuals recorded the same wAlbB haplotype , however , three groups of individuals each had wAlbB haplotypes differing from this haplotype by a single SNP . As Wolbachia is maternally inherited [87] , individuals within each of these groups likely have a shared maternal ancestry . One of these SNPs was observed in three individuals from different sites ( Sites 5 , 9 and 11; see Figs 1 & 4 ) , while the other 2 SNPs were restricted to single sites . Comparing the ratios of wPip to wAlbB alignments among the Shazi Island ( Site 11 ) individuals with all individuals , there was no statistically higher rate of alignment to wPip in Shazi Island ( x¯ = 0 . 489 ± 0 . 037 ) than the average ( x¯ = 0 . 479 ± 0 . 083 ) , so we considered the sample not to be biased by possible accidental releases of females with the wPip infection . The Breteau Index was strongly and positively correlated with “Highway Isolation” ( R2 = 0 . 641 , P < 0 . 002 ) . “River Isolation” ( R2 = 0 . 291 , P < 0 . 071 ) had a negative but nonsignificant correlation with the Breteau Index and “Port Isolation” ( R2 = 0 . 002 , P < 0 . 900 ) showed no correlation . The same general pattern was observed between the NDVI and “Highway Isolation” ( R2 = 0 . 300 , P < 0 . 066 ) , “River Isolation” ( R2 = 0 . 072 , P < 0 . 400 ) and “Port Isolation” ( R2 = 0 . 019 , P < 0 . 666 ) , though none of the relationships were significant at the 95% confidence level . These results suggest that as connectivity to the highway network increases there is a corresponding decrease in Ae . albopictus habitat quality .
Our findings point to the complex processes of gene flow among Ae . albopictus in Guangzhou , facilitated by human transport networks . Similarly-sized cities could show similar genetic patterns , though this will likely depend on whether the local highway network occupies areas of suitable Ae . albopictus habitat . Genetic differentiation may be weaker in recently invaded cities due to founder effects , though this would depend on the frequency of repeated introductions [95] and their populations of origin . Our molecular and analytical approach could be employed to detect inland dispersal between cities , and quantify the relative passive effects of transport ( civilian and freight ) between bus and railway stations , airports , and truck depots . At global scales , they can be used to assign individuals to regional groups with high confidence [43] , allowing for precise modelling of global migration routes throughout the Ae . albopictus range .
|
Aedes albopictus , the Asian Tiger Mosquito , is a highly invasive disease vector with a growing global distribution . Designing strategies to prevent invasion and to control Ae . albopictus populations in invaded regions requires knowledge of how Ae . albopictus disperses . Studies comparing Ae . albopictus populations have found little evidence of genetic structure even between distant populations , suggesting that dispersal along human transportation networks is common . However , a more specific understanding of dispersal processes has been unavailable due to an absence of studies using high-resolution genetic markers . Here we present a study using high-resolution markers , which investigates genetic structure among 152 Ae . albopictus from Guangzhou , China . We found that human transportation networks , particularly shipping terminals , had an influence on genetic structure . We also found genetic distance was correlated with geographical distance , the first such observation in this species . This study demonstrates how high-resolution markers can be used to investigate ecological processes that may otherwise escape detection . We conclude that strategies for controlling Ae . albopictus will have to consider both passive reinvasion along human transportation networks and active reinvasion from neighbouring regions .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"genetic",
"networks",
"ecology",
"and",
"environmental",
"sciences",
"rivers",
"engineering",
"and",
"technology",
"transportation",
"population",
"genetics",
"animals",
"wolbachia",
"genetic",
"mapping",
"transportation",
"infrastructure",
"roads",
"aquatic",
"environments",
"network",
"analysis",
"bodies",
"of",
"water",
"highways",
"population",
"biology",
"insect",
"vectors",
"bacteria",
"civil",
"engineering",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"research",
"errors",
"marine",
"and",
"aquatic",
"sciences",
"disease",
"vectors",
"insects",
"research",
"assessment",
"arthropoda",
"mosquitoes",
"haplotypes",
"eukaryota",
"freshwater",
"environments",
"heredity",
"gene",
"identification",
"and",
"analysis",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"gene",
"flow",
"evolutionary",
"biology",
"organisms"
] |
2017
|
Genome-wide SNPs reveal the drivers of gene flow in an urban population of the Asian Tiger Mosquito, Aedes albopictus
|
Neuronal synapses transmit electrochemical signals between cells through the coordinated action of presynaptic vesicles , ion channels , scaffolding and adapter proteins , and membrane receptors . In situ structural characterization of numerous synaptic proteins simultaneously through multiplexed imaging facilitates a bottom-up approach to synapse classification and phenotypic description . Objective automation of efficient and reliable synapse detection within these datasets is essential for the high-throughput investigation of synaptic features . Convolutional neural networks can solve this generalized problem of synapse detection , however , these architectures require large numbers of training examples to optimize their thousands of parameters . We propose DoGNet , a neural network architecture that closes the gap between classical computer vision blob detectors , such as Difference of Gaussians ( DoG ) filters , and modern convolutional networks . DoGNet is optimized to analyze highly multiplexed microscopy data . Its small number of training parameters allows DoGNet to be trained with few examples , which facilitates its application to new datasets without overfitting . We evaluate the method on multiplexed fluorescence imaging data from both primary mouse neuronal cultures and mouse cortex tissue slices . We show that DoGNet outperforms convolutional networks with a low-to-moderate number of training examples , and DoGNet is efficiently transferred between datasets collected from separate research groups . DoGNet synapse localizations can then be used to guide the segmentation of individual synaptic protein locations and spatial extents , revealing their spatial organization and relative abundances within individual synapses . The source code is publicly available: https://github . com/kulikovv/dognet .
Automation of synapse detection and large-scale investigation of neuronal organization has seen considerable progress in recent years . Most work has been dedicated to the segmentation of electron microscopy datasets , with modern high-throughput pipelines for automated segmentation and morphological reconstruction of synapses [8–10 , 22 , 23] . Much of this progress may be credited to deep convolutional networks . Segmentation accuracy of these approaches can be increased by making deeper networks [24] , adding dilated/ a-trous convolution [25] or using hourglass architectures [8 , 26] that include downscaling/upscaling parts with so-called skip connections . ConvNets typically outperform random forest and other classical machine learning approaches that are dependent on hand-crafted features such as those proposed in [27 , 28] . At the same time , while it is possible to reduce the number of training examples needed by splitting the segmentation pipeline into several smaller pipelines [10] , the challenge of reducnig the number of training parameters without sacrificing segmentation accuracy remains . Within the context of neuronal immunofluorescence images , synapses are typically defined by the colocalization of pre- and postsynaptic proteins within puncta that have sizes on the order of the diffraction limit of 250 nm . One fully automated method using priors , which quantifies synaptic elements and complete synapses based on pre- and postsynaptic labeling plus a dendritic or cell surface marker , was previously proposed and applied successfully [29] . Alternatively , a machine learning approach to synapse detection was proposed in [30 , 31] , where a support vector machine ( SVM ) was used to estimate the confidence of a pixel being a synapse , depending on a small number of neighboring pixels . Synapse positions were then computed from these confidence values by evaluating local confidence profiles and comparing them with a minimum confidence value . Finally , in [32] , a probabilistic approach to synapse detection on AT volumes was proposed . The principal idea of this approach was to estimate the probability of a pixel being a punctum within each tissue slice , and then calculating the joint distribution of presynapic and postsynapic proteins between neighbouring slices . Our work was mainly inspired by works [32] and [11] , that produced the state-of-the-art results in synapse detection on fluorescence images . More conventional machine vision techniques have also been applied for synapse detection [6 , 11 , 12] . These methods aim at detecting regions that differ in brightness compared with neighboring regions . The most common approach for this task is convolution with a Laplacian filter [12] . The Laplacian filter can be computed as the limiting case of the difference between two Gaussian smoothed images . Since convolution with a Gaussian kernel is a linear operation , convolution with the difference of two Gaussian kernels can be used instead of seeking the difference between smooth images . The usage of Difference of Gaussians for synapse detection was proposed in [11] with manually defined filter parameters . Here , we introduce a new DoGNet architecture that integrates the use of simple DoG filters for blob detection with machine , deep learning , thereby combining the strengths of the preceding published approaches [8 , 11 , 32] . Our approach offers the ability to capture complex dependencies between synaptic signals in distinct imaging planes , acting as a trainable frequency filter .
In classical computer vision , the DoG filter is perhaps the most popular operation for blob detection . As follows from its name , DoG filtering corresponds to applying two Gaussian filters to the same real-valued image and then subtracting the results . As the difference between two different low-pass filtered images , the DoG is actually a band-pass filter , which removes high frequency components representing noise as well as some low frequency components representing the background variation of the image . The frequency components in the preserved band are assumed to be associated with the edges and blobs that are of interest . DoG filters are often regarded as approximations to Laplacian-of-Gaussian filters that require more operations to compute . Depending on the parameterization of the underlying Gaussian filters , DoG filters may vary in their complexity . For example , in the most common case , one considers the difference of two isotropic Gaussian probability distribution functions as the filter kernel: DoG Isotropic [ w 1 , w 2 , σ 1 , σ 2 ] ( x , y ) = w 1 exp ( - x 2 + y 2 2 σ 1 2 ) - w 2 exp ( - x 2 + y 2 2 σ 2 2 ) ( 1 ) This version of the DoG filter depends on four parameters , namely the amplitude coefficients w1 and w2 , as well as the bandwidth parameters σ1 and σ2 . The shape of the resulting function is depicted in Fig 2 ( b ) . The amplitudes w1 and w2 can be replaced by normalizing coefficients 1/2πσ1 and 1/2πσ2 respectively , reducing the number of trainable parameters to just two . The four- and the two-parameter DoG filters described above are suitable for detecting isotropic blobs . For anisotropic blob detection , pairs of anisotropic Gaussians with zero means and shared orientations may be more suitable . In this case , we parameterize an anisotropic zero-mean Gaussian as: G w , σ x , σ y , α ( x , y ) = wexp ( - a x 2 - 2 b x y - c y 2 ) ( 2 ) where for an orientation angle α ∈ [0; π ) the coefficients a , b , c are defined as: a = cos 2 α 2 σ x 2 + sin 2 α 2 σ y 2 ( 3 ) b = - sin 2 α 4 σ x 2 + sin 2 α 4 σ y 2 ( 4 ) c = sin 2 α 2 σ x 2 + cos 2 α 2 σ y 2 ( 5 ) The anisotropic DoG filter is then defined as: DoG Ansotropic [ w 1 , w 2 , σ 1 , x , σ 1 , y , σ 2 , x , σ 2 , y , α ] ( x , y ) = G w 1 , σ 1 , x , σ 1 , y , α - G w 2 , σ 2 , x , σ 2 , y , α ( 6 ) We refer to the DoG filter ( 6 ) as the Anisotropic or seven-parameter DoG filter based on the number of associated parameters . The five-parameter DoG filter can be obtained by fixing the constants w1 and w2 to be normalizing , i . e . w i = 1 / 2 π σ i , x σ i , y . An example of anisotropic Difference of Gaussians is depicted in Fig 2 ( c ) . The usage of anisotropic difference of Gaussians allows detecting different kinds of elongated blobs with only three additional trainable parameters per filter ( compared to the two- or four-parameter versions ) . Overall , DoG filters provide a simple way to parameterize blob-detecting linear filters using a small number of parameters . They can also be extended to three-dimensional blob detection in a straightforward manner . Since in three dimensions generic linear filters come with an even larger number of parameters , the use of DoG parameterization is even better justified . Here , one natural choice would be to use differences of Gaussian filters that are isotropic within axial slices and use a different variance ( bandwidth ) along the axial dimensions: G w , σ , σ z ( x , y , z ) = wexp ( - x 2 + y 2 2 σ 2 - z 2 2 σ z 2 ) ( 7 ) DoG 3D [ w 1 , w 2 , σ 1 , σ 2 , σ 1 , z , σ 2 , z ] = G w 1 , σ 1 , σ 1 , z - G w 2 , σ 2 , σ 2 , z ( 8 ) Generally , as axial resolution in 3D fluorescence microscopy is typically lower , σi , z is also taken to be larger than σi . The filter ( 8 ) provides a six-parameter parameterization of a family of 3D blob detection filters ( one of which is visualized in Fig 2 ( d ) ) , whereas a generic 3D filter takes O ( d3 ) parameters , where d is the spatial window size . The shallow ( single layer ) Difference of Gaussians network ( DoGNet ) is a neural network built around DoG filters Fig 2 ( a ) . It takes as an input a multiplexed fluorescence image , applies multiple DoG filters ( 1 ) , ( 6 ) or ( 8 ) to each of the input channels . Subsequently , DoGNet combines the obtained maps linearly ( which in deep learning terminology corresponds to applying 1 × 1 convolution ) . The latter step obtains a single map of the same spatial resolution as the input image . Finally , a sigmoid non-linearity is applied to convert the applied maps into probability maps . More formally , we define a single-layer DoGNet as Ψ ( X ; θ = { γ , β , ζ } ) = S ( ( X ⊛ D o G β ) ⊛ γ + ζ ) , ( 9 ) where X denotes the input multiplexed image , ⊛ is the 2D convolution operation , and the vector β denotes the parameters of all DoG filters . Assuming that the input contains N channels , and each channel is filtered with M DoG filters , the application of all DoG results in M × N maps . Those maps are then combined into K maps using a pixel-wise linear operation ( which can be treated as a convolution with 1 × 1 filters ) . The tensor corresponding to such linear combination and containing K × M × N values is denoted γ . To each of the obtained K maps , the bias value ζk is added , and finally all obtained values are passed through the element-wise sigmoid non-linearity S ( x ) = 1/ ( 1 + exp ( −x ) ) . Overall , θ in ( 9 ) denotes all learnable parameters of the DoGNet . In the case of the single-layer DoGNet , the output has a single map ( i . e . K = 1 ) . Except for the last sigmoid operation , the single-layer DoGNet contains only linear operations and can be regarded as a special parameterization of the linear filtering operator that maps the input M maps to several output maps , usually two maps . The deep DoGNet architecture is obtained simply by stacking multiple DoGNet layers ( 9 ) : Φ ( X ; θ = { θ 1 … θ T } ) = Ψ ( Ψ ( … Ψ ( X , θ 1 ) … ; θ T - 1 ) ; θ T ) , ( 10 ) where T is the number of stacked single layers DoGNets , and θt denotes the learnable parameters of the t-th layer . The final number of maps KT is once again set to one , so that the whole network outputs a single probability map . However , the numbers of layers Kt that are output by the intermediate DoGNet layers would typically be greater than one . In our experiments the number of sequential layers T was set to three . Inspired by an idea from [32] , instead of producing a single probability map , our network delivers two independent maps and using the element-wise product of those maps we get the final map . We have implemented this approach as a separate layer and that does not require any trainable parameters . In the case of synapses , this step allows reducing the effect of displacement between pre- and postsynaptic punctae by learning probability maps independently for pre and postsynaptic signals . Given several probability maps ( for pre- and postsynaptic punctae ) the element-wise products will act as a logical operator “AND , ” highlighting the intersection between those maps , where the synaptic cleft is located . In our research we use element-wise multiplication not only for DoGNets but for baselines as well , they all benefit from this layers . We have found that appropriate parameter initialization is key to obtaining reproducible results with our approach . Popular neural networks have a redundant number of parameters and are initialized by sampling their values from a Gaussian distribution . This initialization is not suitable for DoGNets because of the relatively small number of parameters . Instead , we use a strategy from object detection frameworks [34] . This approach consists of initialization with a range of reasonable states ( priors ) . An optimization procedure selects the best priors and tunes their parameters . In DoGNet we use Laplacian of Gaussians with different sizes that are sampled from a regular grid as priors . Specifically , we obtain the Gaussian variance ( sigma ) by splitting the line segment [0 . 5 , 2] into equal parts . The number of parts depends on the number of DoGs reserved for each image plane ( in our experiments that number was set to five ) . We set the difference-variance in the Laplacian of Gaussians to 0 . 01 . For example , if we set the number of DoGs for a channel to 3 , the sigmas will be 0 . 5 , 1 . 25 , and 2 , respectively . We train the described architecture by minimizing the softdice loss ( 11 ) proposed in [35] between the predicted probability map Ψ ( X; θ ) and a ground truth mask Yg: L θ ( X , Y g ) = 1 - 2 ∑ Y g Ψ ( X ; θ ) ∑ Ψ ( X ; θ ) 2 + ∑ Y g 2 ( 11 ) Here , sums are taken over individual pixels , and in the ground-truth map Yg all pixels belonging to synapses are marked with ones , while the background pixels are marked with zeros . In the experiments we found that on the imbalanced data typical for synapse detection problems , this loss performs better than standard binary cross entropy . In order to optimize this loss function , partial derivatives with respect to DoGNet parameters dL/dθ must be obtained , which may be accomplished via backpropagation [33] . The backpropagation process computes the partial derivatives with respect to the filter parameters at each of the spatial positions within the spatial support of the filter ( which we limit to 15 pixels ) . The partial derivatives with respect to the DoGNet parameters are then obtained by differentiating formulas ( 1 ) , ( 6 ) or ( 8 ) at each spatial location and multiplying by the respective derivatives . The ground truth mask Yg as well as the input images X for the training process are obtained using a combination of manual annotation and artificial augmentation . The synapse detection in FM images is a challenging and arguably ambiguous task even for human experts . Furthermore , even a small , 100 × 100 pixel region of an image might contain more than 80 synapses . In practice it is impossible to annotate the borders of each synapse accurately , therefore the experts were asked to mark the centroid of synapses only , corresponding to the synaptic cleft , after which all pixels within a radius of 0 . 8μm were assigned to the corresponding synapse . We trained DoGNets for 5000 epochs . Each epoch is a set of ten randomly cropped subsamples 64 × 64 from the annotated training dataset . Because DoGNets have few parameters , we found that the training processes converged rapidly typically requiring only several minutes on an NVidia Titan-X GPU for the datasets described below . Once trained , inference can be performed on a CPU as well as on a GPU using the implementations of Gaussian filtering that may be optimized for a particular computing architecture . Our implementation uses the PyTorch deep learning framework [36] , which allows for concise code and benefits from automatic differentiation routines . Because both shallow and deep versions of DoGNet produce probability maps rather than lists of synapse locations and parameters , these probability maps need to be postprocessed in order to identify synapse locations and properties . Toward this end , first , we reject points with low confidence by truncating the probability maps using a threshold of τ of 0 . 5 . In order to extract synapse locations from the probability map produced by the DoGNet , we need to find local maxima . In standard fashion , we greedily pick local maxima in the probability map , traversing them in the order of decreasing probability values while suppressing all maxima within a cut-off radius R = 1 . 6μm from previously identified maxima ( so called non-maxima suppression ) [37] . The output of this procedure is the x and y locations of synaptic puncta . The next step is to describe each detected punctum with a vector containing the information about the detected synapse . To obtain a descriptor for a synapse , we select a small window of the same radius R = 1 . 6μm around its location , fit Gaussian distributions to each of the input channels , and for each protein marker we store the average intensity , the displacement of the Gaussian mean with respect to the window center , the Gaussian orientation , and its asymmetry . Evaluating the quality of such a descriptor is left for future work .
The proposed method and a set of baselines were evaluated on four independent datasets for which synapses were annotated manually: [Collman15] dataset of conjugate array tomography ( cAT ) images [16] , [Weiler14] dataset of array tomography ( AT ) images [17] , [PRISM] dataset of multiplexed confocal microscopy images [6] , and a synthetic dataset that we generate here . Each published experimental dataset was obtained using fluorescence imaging based on commercially available antibodies , with synapsin , vGlut , and PSD-95 markers common to the datasets . At the end of section , we additionally perform comparisons using synthetic dataset with excitatory and inhibitory synapse sub-types . In each of our trials we compared several DoGNet configurations with several baseline methods including reduced version of the fully convolutional network ( FCN ) [15] , and an encoder-decoder network with skip connections ( U-net ) [8] . An exhaustive comparison between different deep architectures is a nearly impossible task , mostly because of an infinite number of possible configurations . Nevertheless , we have done our best to tune the parameters of the baseline methods . The best-performing variants of the baseline architectures ( FCN , Unet ) were used in the experiments and are described in detail in the supplementary material . To make our evaluation more direct , we have designed the competitive networks to have the same receptive field ( FOV ) ( arbitrarily chosen to 15 pixels ) . We have also evaluated two manually-tuned methods , namely the probabilistic synapse detection method [32] and the image processing pipeline proposed in [38] . Detailed technical background on these architectures are described in supplementary materials . The DoGNet architecture has two major options: Shallow and Deep , with the Shallow option corresponding to a single layer and the Deep option corresponding to number of sequential layers . The second word in our notation Isotropic or Anisotropic indicates the number of degrees of freedom in the DoG parameterization , e . g . Isotropic denotes four-degree DoG ( 1 ) . The number of DoG filters for each channel was arbitrary set to five . We also evaluated a simple ablation denoted as Direct that takes the Shallow Isotropic DoGNet architecture and replaces DoG-parameterized filters with 15 × 15 unconstrained filters ( thus using Direct parameterization ) ( see Supplementary Information ) . The quality of synapse detection was estimated using the standard metrics: precision , recall , and F1-score , with the output of each method consisting of the set of points denoting synapse coordinates . True positives were estimated as the number of paired points between annotation and detection provided the distance between them was less than half of the mean synapse radius ( ρ = 0 . 6μm ) . To avoid multiple detections of synapses ( false positives ) , we require that each detected point can be matched at most once . Detections and annotations without pairs were considered to be false positives and false negatives , respectively . The precision measure was then computed as the ratio of true positives to all positives , and the recall measure as the ratio of true positives to all synapses contained in the annotation . The F1-score combines the precision and recall in one criterion by taking the double product of recall and precision divided over their sum . For evaluation purposes , we also added the AUC criterion corresponding to the area under the ROC curve obtained by varying the confidence threshold τ . This criterion is stable to the threshold choice and depends on the quality of the probability map produced by a method . For different thresholds , we estimated the conjunctions between probability map and ground truth binary segmentation pixel-wise . For quantitative comparison , we have also used the absolute difference in counting ( |DiC| ) . This metric merely computes the difference between the number of synapses detected using a method and the ground truth . This measure does not answer the question of how well a synapse was localized but still gives additional insight into quantitative results . Since the training procedure is a probabilistic process depending on initialization and data sampling , we estimate each value as the mean of five independent runs . To verify our method on PRISM data [6] , we performed manual dense annotation of several image regions of a dataset of FM images obtained using this technique . The manual annotation was performed by two experts using synapsin , vGlut , Bassoon and PSD-95 channels . Each expert annotated three regions . The total set was made of six regions and split into training , validation ( 392 synaptic locations ) and testing subsets ( 173 synaptic locations ) . Each subset consisted of two regions annotated by different experts , with test regions overlapped in order to estimate inter-expert agreement . For synapse annotation , we developed a graphical user interface . This software allows selecting image channels and regions . As we solve the task of semantic segmentation during the training , we need a densely annotated image region . We mark each synapse with a point approximately at the synaptic cleft . Evaluation against baselines is presented in Table 1 . Due to circular puncta shape and the relatively small displacement of markers , the optimal method was Shallow Anisotropic with only 107 trainable parameters . This configuration also performed considerably better than the Direct Ablation approach , highlighting the advantage of using DoG parameterization in place of direct parameterization of the filters . We performed several analyses in order to evaluate agreement between three independent human experts as well as between the experts and our method ( Table 2 ) . Importantly , the proposed network agreed with the Experts similarly to the agreement between the Experts themselves . In this dataset , the alignment of electron microscopy ( EM ) and array tomography ( AT ) images provides the ground truth for synapse detection using fluorescence markers . Using high resolution EM data synaptic clefts and pre- versus post- synaptic sites can be identified unambiguously , which was used as validation for the synapse detections from fluorescence data ( Fig 3 ( a ) ) . The dataset contains 27 slices of 6310 × 4520 pixels each , with a resolution of 2 . 23 × 2 . 23 × 70 nm , and contains annotation with pixel-level segmentation of synaptic clefts . In order to fit our training procedure , we have used only synaptic cleft centroid coordinates . The EM resolution is much greater , so AT data were interpolated to be aligned with EM data . Provided we utilize solely AT data , its original resolution of 0 . 1μm per pixel can be recovered without losing any information . The first five slices were used as the train dataset , whereas the remainder ( slices 6-27 ) served as the test dataset . Results of our evaluation ( Table 3 ) show that shallow DoGNets exhibit highest performance in terms of the F1-measure . The receptive field 15 × 15 pixels followed by inter-channel element-wise multiplication allow capturing highly displaced markers puncta combinations . Displacements in marker punctae occur because synapses are 3D objects with random orientations . Therefore , the presynaptic and postsynaptic signals in the image plane produce displaced peaks up to a half of a micron . The closest-performing ConvNet architecture was U-net with 622 trainable parameters; increasing the number of its parameters led to overfitting and therefore lower performance on the test dataset examined here . The AT stains include markers specific for excitatory ( vGlut , PSD95 ) and inhibitory ( GABAergic , gephyrin ) synapses . In our experiments , the use of inhibitory markers did not improve the detection scores . Moreover , the precision of all trainable methods was considerably lower using only inhibitory markers ( synapsin , GABA , gephyrin ) . The Weiler dataset [17] consists of 12 different neural tissue samples . Each sample was stained with a number of distinct antibodies including synapsin vGlut , and PSD-95 . For each stain , 70 aligned slices were acquired using array tomography ( AT ) . Each slice was a 3164 × 1971 pixel image with spatial resolution of 0 . 2μm per-pixel . This dataset does not have any published annotation . We investigated the ability of DoGNets to generalize across distinct datasets by applying networks trained on the well-annotated [Collman15] dataset , which was annotated using serial electron microscopy data , to the previously unlabeled AT dataset [Weiler14] [17] . Generally , the staining of [17] is similar to the Collman15 dataset [16] . Thus , we first performed a coarse alignment by resizing [Collman15] images and applying linear transforms to the intensities of each channel so that the magnification factors , means , and standard deviations of the intensity distributions were matched . The architectures trained on [Collman15] were then evaluated on [Weiler14] . Qualitative examples of this cross-dataset transfer are shown in Fig 3 . For quantitative validation we generated manual annotations of two randomly selected regions of the [Weiler14] dataset using the same software that we have used for [PRISM] annotation . We observed that the levels of agreement between the results of the DoGNet Shallow Anisotropic trained on [Collman15] dataset and each of the experts were similar to the level of inter-expert agreement ( in terms of the F1 score ) . The results of this cross-dataset validation are shown in Table 4 . Importantly , while the performance of compared methods , did not diminish dramatically . In fact , the DoGNets actually improved in their performance , which we attribute to the fact that in the Weiler dataset all expert annotations were based on FM images , rendering the analysis more straightforward in comparison with the [Collman15] synapses that are visible in EM data but not in the FM data that were not included . In order to further evaluate our approach rigorously in a fully controlled setting , we also applied it to a synthetic dataset . The goal of the evaluation of DoGNet using synthetic data was to estimate the quality of synapse detection compared with baseline procedures for distinct levels of signal-to-noise ratio; including the presence of spurious synapses; and for different presynaptic-to-postsynaptic markers displacements on image planes to emulate the 3D structure of synapse . Further , this systematic evaluation using synthetic data addresses questions regarding meta-parameter choice , methodological limitations , and the justification of neural network usage for synapse detection tasks . Because the number of training samples was unlimited , deep networks with a large number of parameters were unlikely to overfit the data . Our dataset models three entities: true synapses , spurious synapses that emulates false bindings , and random noise . We emulated true synapses and spurious synapses using Gaussian probability density functions placed in different image planes with additive white noise , where each image plane refers to a specific protein marker such as synapsin , vGlut , PSD-95 , vGat or gephyrin . To assess the generalized performance of different architectures , in our synthetic experiments we simulated both excitatory and inhibitory synapses . Spurious synapses are made to emulate false bindings in combination with random noise in order to act as a distraction for the classifier to evaluate its robustness . An actual synapse has intensity peaks at least in one presynaptic and in one postsynaptic image plane , while spurious synapses have peaks only in presynaptic or postsynaptic channels , but never in both . An example of a true excitatory synapse might be a signal that has a punctum in synapsin , vGlut and PSD-95 markers separated by a distance less than a half of a micron . An inhibitory synapse would have punctae in synapsin , vGat and gephyrin . The displacement in markers punctae , caused by the 3D structure of synapses , makes the process of differentiation between actual and spurious synapses considerably more challenging , thereby rendering the simulation more realistic . The intensity of the synaptic signal were emulated using Log-Normal distribution with zero mean and σ = 0 . 1 . Modeling synapses using isotropic Gaussians in our synthetic dataset enables the initial evaluation of purely isotropic DoGNets . First the sensitivity of the approach to signal-to-noise ratio was evaluated ( Fig 4 ) . Results indicate that small convolutional neural networks are sensitive to initialization and may become trapped in local minima , whereas DoGNet performance was more robust , although DoGNets initialized randomly rather than using our initialization scheme also suffered from local minima . Importantly , deeper architectures were capable of handling larger displacements between punctae ( Fig 5 ) . This result is anticipated because multi-layer architectures have larger receptive fields and capture more non-linearities , allowing the capture of more complex relations in the data . For example , in the presence of substantial displacements , at least one additional convolution layer followed by an element-wise multiplication was needed to perform a logical AND operation between pre and post synaptic channels after blob detection [32] . We also present a study of training with limited examples . We have evaluated trainable methods ( Direct , FCN , U-Net , Shallow Isotropic , Deep Isotropic ) on fixed size crop without any augmentation in search of minimal size of image region when each method starts work suitable the signal-to-noise ration was sent to approx 4 . 5 and the maximal displacement to two pixels . We present the results of this study in ( Fig 6 ) . We show that Shallow and Deep DoGNets are able to learn a simple signal like a multiplexed blob form only few samples .
We present DoGNet—a new architecture for blob detection . While DoGNets are applied here to synapse detection in multiplexed fluorescence and electron microscopy datasets , they are more broadly applicable to other blob detection tasks in biomedical image analysis . Due to their low number of parameters , DoGNets can be trained in a matter of minutes , and are suitable for non-GPU architectures because the application of a pretrained DoGNet amounts to a sequence of Gaussian filtering and elementwise operations . In our experiments , DoGNets were able to robustly detect millions of synapses within several minutes in a fully automated manner , with accuracy comparable to human annotations . This computational efficiency and robustness may prove essential for the application of multiplexed imaging to high-throughput experimentation including genetic and drug screens of neuronal and other cellular systems .
|
Multiplexed fluorescence imaging of synaptic proteins facilitates high throughput investigations in neuroscience and drug discovery . Currently , there are several approaches to synapse detection using computational image processing . Unsupervised techniques rely on the a priori knowledge of synapse properties , such as size , intensity , and co-localization of synapse markers in each channel . For each experimental replicate , these parameters are typically tuned manually in order to obtain appropriate results . In contrast , supervised methods like modern convolutional networks require massive amounts of manually labeled data , and are sensitive to signal/noise ratios . As an alternative , here we propose DoGNet , a neural architecture that closes the gap between classical computer vision blob detectors , such as Difference of Gaussians ( DoG ) filters , and modern convolutional networks . This approach leverages the strengths of each approach , including automatic tuning of detection parameters , prior knowledge of the synaptic signal shape , and requiring only several training examples . Overall , DoGNet is a new tool for blob detection from multiplexed fluorescence images consisting of several up to dozens of fluorescence channels that requires minimal supervision due to its few input parameters . It offers the ability to capture complex dependencies between synaptic signals in distinct imaging planes , acting as a trainable frequency filter .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"fluorescence",
"imaging",
"medicine",
"and",
"health",
"sciences",
"neural",
"networks",
"engineering",
"and",
"technology",
"nervous",
"system",
"condensed",
"matter",
"physics",
"light",
"microscopy",
"electrophysiology",
"anisotropy",
"neuroscience",
"microscopy",
"materials",
"science",
"optical",
"equipment",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"imaging",
"techniques",
"fluorescence",
"microscopy",
"prisms",
"physics",
"anatomy",
"synapses",
"equipment",
"physiology",
"electron",
"microscopy",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"material",
"properties",
"neurophysiology"
] |
2019
|
DoGNet: A deep architecture for synapse detection in multiplexed fluorescence images
|
One of the main outcomes of quantitative genetics approaches to natural variation is to reveal the genetic architecture underlying the phenotypic space . Complex genetic architectures are described as including numerous loci ( or alleles ) with small-effect and/or low-frequency in the populations , interactions with the genetic background , environment or age . Linkage or association mapping strategies will be more or less sensitive to this complexity , so that we still have an unclear picture of its extent . By combining high-throughput phenotyping under two environmental conditions with classical QTL mapping approaches in multiple Arabidopsis thaliana segregating populations as well as advanced near isogenic lines construction and survey , we have attempted to improve our understanding of quantitative phenotypic variation . Integrative traits such as those related to vegetative growth used in this work ( highlighting either cumulative growth , growth rate or morphology ) all showed complex and dynamic genetic architecture with respect to the segregating population and condition . The more resolutive our mapping approach , the more complexity we uncover , with several instances of QTLs visible in near isogenic lines but not detected with the initial QTL mapping , indicating that our phenotyping accuracy was less limiting than the mapping resolution with respect to the underlying genetic architecture . In an ultimate approach to resolve this complexity , we intensified our phenotyping effort to target specifically a 3Mb-region known to segregate for a major quantitative trait gene , using a series of selected lines recombined every 100kb . We discovered that at least 3 other independent QTLs had remained hidden in this region , some with trait- or condition-specific effects , or opposite allelic effects . If we were to extrapolate the figures obtained on this specific region in this particular cross to the genome- and species-scale , we would predict hundreds of causative loci of detectable phenotypic effect controlling these growth-related phenotypes .
Fine-tuning plant growth throughout development and in response to environmental limitations is a decisive process to optimize fitness and population survival in the wild . As a sessile organism , plants have to cope with environmental fluctuations and evolved a wide range of responses . This is well illustrated by their great phenotypic plasticity and their ability to colonize very diverse habitats , through intraspecific genetic diversity as revealed in most pathways [1] . Aerial and below-ground growth represent a balance between resource investment in the structures and resource acquisition ( respectively photosynthesis and water / nutrient uptake ) . Thus , growth is a highly complex trait controlled by many genes with constitutive or more specific roles depending on developmental stage , tissue , timing , environment [2–7] . In this context , plant growth can be considered as a model complex trait to increase our knowledge in the genetics of evolution , as well as to improve plant performance . Forward mutant analysis plays a central role in plant biology to blindly identify gene functions associated with a phenotype [8] , but sometimes remains limiting to reveal genes with modest phenotypic effect , or when addressing genes from redundant families . With regard to growth and stress tolerance , these limitations are likely to be relevant given the multigenic nature of growth phenotypes , the low mean effect at each locus and/or epistatic interaction they involve [9 , 10] . Thus , the use of naturally-occurring variation through quantitative genetic approaches designed to map quantitative trait loci ( QTLs ) is interesting notably to complement the search for alleles selected during evolution which may not be brought out with classical loss-of-function approaches . Linkage mapping and genome-wide association lead to the identification of large amount of alleles involved in intraspecific phenotype variation from different plant species [1 , 11] . With the drop of sequencing and genotyping costs , phenotyping clearly is the limiting factor for quantitative genetics approaches [12] . However , the complexity of the genetic architecture of a given trait , which depends on the contribution and the number of loci controlling a trait and their interactions with the genetic background and the environment , has direct consequences on how much phenotyping remains limiting . Highly heritable traits with a limited number of contributing loci ( in a given segregating material , or at the species scale ) are more likely to be well understood than more complex traits . For instance , a large part of the phenotypic variation for flowering time in Arabidopsis thaliana maps to a limited number of loci [13–16] , including FRIGIDA and FLC genes [17–19] and thus has a relatively simple genetic architecture , although many more loci make smaller contributions -at least in some environments- and allelic heterogeneity also interferes [15 , 16 , 20 , 21] . By contrast , traits like fitness or growth can be expected to have a more complex basis as they integrate many upstream traits , and consequently many genes , each prone to residual variation and heterogeneity . Smaller contributions from individual loci means that , although one can still estimate total heritabilities , the accuracy and throughput of phenotyping will be limiting to confirm individual QTLs' contributions . Heritabilities for flowering time-related traits will often be above 80% , while biomass accumulation or fitness' heritabilities are essentially found in the range 20–60% [22–27] . Another factor that will influence the genetic complexity of a trait is its response to the environment through phenotypic plasticity [28] . Part of the environmental fluctuations may be controlled in an experimental design , while another part may contribute to the residuals . Whether the sensitivity of a pathway or trait to the environment depends on the number and architecture of the contributing loci remains an open question , however the relationships between higher plasticity and lower heritability are described [26] . Water availability is an environmental factor that varies through space and time and shows great heterogeneity which certainly constrains plant growth and shapes plant distribution in nature and in agricultural systems . Prevalence of drought episode is expected to increase with global climate change making the understanding of plant response to drought one of the major challenge of the next decades [29 , 30]; this includes deciphering the genetic basis for variation in mechanisms such as drought escape , avoidance and tolerance [31] . Hence , this environmental parameter is definitely a good candidate to understand the genetic architecture of GxE . However , drought is both difficult to control and hard to predict , because of interactions with almost all other factors in the environment ( temperature , air flow , light ) and interplay with other constraints ( especially nutrient-related or osmotic ) . The development of robotic phenotyping tools throughout the community makes it now feasible to acquire traits on hundreds or thousands of plants in precisely controlled and reproducible conditions [32–35] , pushing a bit further one of the main limitation for a better decomposition of the genetic architecture of these complex traits . Still , regarding plant growth variation in nature , mainly genomic regions with relatively large effect were identified in Arabidopsis and were often related to development , immunity or major hormones ( for instance [36–43] ) . A limited number of non-theoretical studies seem to confirm that many genes with smaller effect–potentially involving epistasis and linked loci–would be responsible for part of the phenotypic variation of such complex traits ( for instance [44–46] ) . Here , we undertook a precise analysis of plant growth genetic architecture under both optimal watering condition and mild drought stress ( as revealed under artificial conditions of reduced and constant soil water content ) , using a classical linkage mapping approach on 4 biparental segregating populations [47] . As a dynamic trait , we chose to follow growth during the vegetative phase using a high-throughput phenotyping robot ( the Phenoscope [34]; https://phenoscope . versailles . inra . fr/ ) to map major- to small- effect QTLs as well as their interaction with drought stress . Zooming in on the loci , we use near-isogenic lines to validate these QTLs and reveal in more detail the genetics behind a single QTL peak . We then focus more precisely on a region where a major Quantitative Trait Gene ( QTG ) is segregating ( = CRY2 , a known polymorphic actor with major pleiotropic phenotypic consequences ) , and show that other loci with additive or opposite effect are also present in its vicinity , illustrating the complexity of growth genetic architecture .
The four RIL sets used in this study ( BurxCol , CvixCol , BlaxCol , YoxCol ) were conducted under well-watered ( WW ) and moderate water deficit ( WD ) conditions on our high-throughput phenotyping platform . Ensuring that growth occurs in a highly controlled and homogeneous environment , the Phenoscope records a number of image-based quantitative traits describing shoot development ( Fig 1 ) . Taking daily pictures gave access to cumulative ( Projected Rosette Area , PRA ) and dynamics ( Relative Expansion Rate , RER ) growth parameters for individual plants ( Fig 1C & 1D ) as well as other descriptive or derived traits ( rosette morphology and RGB colour components ) [48] . A principal component analysis ( PCA ) was performed using all picture-based phenotypes at the final day of the experiment , 29 Days After Sowing ( DAS; hence 'PRA29' etc ) and relative expansion rate calculated between 16 and 29 DAS ( RER16-29; Fig 2 ) . The first axis explained a major part of the total variance , essentially through final rosette size ( PRA29 ) and expansion rate ( RER16-29 ) . However , PRA29 and RER16-29 variables were not perfectly correlated , with genotypes exhibiting moderate PRA29 despite high RER16-29 . The red ( Red29 ) and green ( Green29 ) components colour phenotypes mainly contributed to the second axis and were positively correlated , and both were negatively correlated with rosette compactness at 29DAS ( Compactness29 ) which was the main trait contributing to the third axis . Individual projection showed that the first axis strongly structured the individuals according to the watering treatment ( WW versus WD ) while the axes 2 and 3 represented cross ( RIL set ) effects , differentiating CvixCol and BurxCol ( axis 2 ) and BlaxCol ( axis 3 ) from the other RIL sets . PRA29 , RER16-29 and Compactness29 are complementary growth phenotypes that were investigated further in this study to quantify different aspects of shoot development variation: final projected rosette area is a cumulative proxy for biomass and photosynthetically-active surface , rosette compactness is an informative parameter describing the rosette morphology , and relative expansion rate highlights the dynamics of growth . Phenotypic distribution among the RILs compared to their parents ( S1 Fig ) revealed extensive transgressive segregation for most of the traits and the crosses studied . As expected , mild drought stress ( WD ) condition impacted the distribution of the RILs for PRA29 and RER16-29 with generally reduced values for both traits . Interestingly , compactness distribution was much more robust to stress , which indicated that overall the morphology of the rosette is less affected by mild drought . In order to estimate the part of phenotypic variation that is explained by genetic factors for each traits , heritabilities were calculated ( S1 Table ) and were essentially below 0 . 2 for RER ( except in BurxCol where they were higher ) , essentially around 0 . 5 for cumulative PRA and generally above 0 . 5 for Compactness ( this trait is certainly less sensitive to shifts in developmental stage that could be induced , for instance , by small changes in germination time ) . Overall , heritabilities were also lower in stress conditions than in control , as if the stress was inducing noisier phenotypic variations . According to ANOVA analyses ( S2 Table ) , all traits and RIL sets showed significant variation according to both the watering condition ( WD versus WW ) and the genotype ( within each RIL set ) , except for Compactness29 response to stress in BlaxCol . There were also weaker ( compared to the genotype and watering condition effects ) but still significant biological replicates’ effects ( i . e . independent Phenoscope experiments ) , but less genotype x experiment interactions ( with the exception of PRA and RER traits in BlaxCol for instance ) . PRA is more prone to genotype x experiment interactions than other traits , especially in CvixCol and YoxCol . Genotype x condition interactions are often milder than genotype or condition effects , and overall compactness–or YoxCol–show much less genotype x condition interactions than other traits/sets . The phenotypic values of each RIL were then corrected for inter-experiment differences ( indicated by the significant biological replicates’ effect ) . Our experimental design allowed the identification of many QTLs for all combinations of traits , conditions and RIL sets , and also for the GxE interaction term using genotype x condition effects from the ANOVA model for each trait ( Fig 3 and S3 Table ) . Globally 112 QTLs were identified all along the genome when conditions are studied independently ( 62 under WW + 50 under WD ) likely corresponding to at least 18 independent loci ( genetically distant-enough from each other to be most likely considered as independent ) , yielding a median of 4 QTLs per modality of cross x trait x condition ( ranging from 1 to 8 QTLs ) . QTL hotspots across RIL sets and traits were identified for instance at the beginning of chromosome 1 , at the bottom of chromosome 2 and 5 . These hotspots include very highly significant QTLs with LOD scores above 10 , and up to 32 . Chromosome 3 appeared to show less significant QTLs in all crosses , especially for PRA29 and RER16-29 . Individual QTL contributions to phenotypic variance ( R2 ) ranged from 1 up to 30% , and showed a L-shaped distribution of effect ( S2 Fig ) . Using empirical significance boundaries according to the observed distribution of QTLs effects , ~10% of the QTLs could be considered as showing major effects and significance ( R2>10% and/or LOD>15 ) ; ~25% of the QTLs could be considered as showing intermediate effects and significance ( 5%<R2<10% and/or 7<LOD<15 ) ; the remaining 2/3rd of the QTLs could be considered as showing minor effects and significance R2<5% and/or LOD<7 ) . Many more potential QTLs not listed here were only suggestive with LOD score peaks just below our permutation-based threshold ( <2 . 4 LOD ) . Most of the QTL profiles are stable across the 2 watering conditions , especially for the major-effect loci . However , QTLs specific for one condition were detected , e . g for RER16-29 under WD in YoxCol on chromosome 4 , and at the top of chromosome 1 under WW . We also mapped QTLs for the interaction term with the drought treatment , yielding 19 QTLs ( Fig 3 and S3 Table ) . These essentially emphasize large effect QTLs showing a modulation of their effect in response to stress ( especially for PRA , see chromosome 1 and 5 ) , with no new loci revealed . For RER16-29 in YoxCol , this would confirm the interaction of the above-mentioned locus on chromosome 4 ( although its exact position is questionable ) , but not for the one at the top of chromosome 1 . There may be some power issues when comparing across conditions due to lower heritabilities of traits under WD . Although derived from PRA29 , Compactness29 showed an independent genetic basis , as exemplified by the major peaks on the bottom of chromosome 4 in BurxCol or 5 in BlaxCol . Although contributing to RER16-29 , PRA29 does not always share the same contributing loci , for instance on the top of chromosome 1 in CvixCol and BurxCol . Other more complex cross x trait patterns are apparent , like at the bottom of chromosome 2 where a major QTL for Compactness29 in three of the four RIL sets seems to colocalize with a significant PRA29 and RER16-29 QTL ( in the same direction ) , but only in one cross . It may be that these Compactness29 QTLs are actually independent in each RIL set . A two-dimensions search for epistatic interactions was performed across all traits , conditions and RIL sets ( S3 Fig ) . Overall , the BurxCol and CvixCol RIL sets showed more significant epistasis compared to the two other sets . Interestingly , pairwise interactions controlling growth phenotypes are overall quite different depending on RIL set and growth phenotypes . Shared epistasis effects are potentially detected in both watering conditions: they appear as symmetrical across the diagonal on S3 Fig . One of the most significant epistatic interaction was observed between 2 loci on the top and bottom of chromosome 4 in the BurxCol cross ( S3 Fig; interaction component highly significant for RER16-29 in WW: 5 . 1 LOD ) . Positions and directions of effect match perfectly with the previously published SG3 x SG3i interaction known to segregate in this cross [49] . Based on its significance , another relevant interaction ( 3 . 34 LOD in WW and 3 . 12 LOD in WD ) was observed for PRA29 in the BlaxCol cross between the bottom of chromosomes 4 and 5 ( Fig 4 ) . The effect of the bottom of chromosome 5 QTL on PRA29 is observed only when RILs carry Bla alleles at the bottom of chromosome 4 . As a consequence of this epistasis , these QTLs appear barely significant in single QTL scans ( Fig 3 ) . We also performed dynamic QTL detection on daily-recorded traits ( PRA and Compactness ) to reveal the evolution of QTL effect throughout the experiment on the Phenoscope: these interactive QTL profiles can be accessed at http://www . inra . fr/vast/PhenoDynamic . htm Most of the PRA QTLs observed after 29 days of growth correspond to locus that become gradually significant across the experiment and are essentially not time-specific . These are most likely contrasted allelic effect on growth that cumulate their effect over time . There are only a few exceptions for PRA , like the bottom of chromosome 5 locus segregating in BlaxCol which remains significant only until 13DAS and thereafter is canceled out . For Compactness , the picture is rather different with numerous examples of QTLs that are essentially significant around specific time-points , even sometimes in successive waves of significance ( BlaxCol , bottom of chromosome 5: QTL peaking at Days 11 , 17 and 27 –providing that this is a unique locus ) . In order to take advantage of the different crosses to Col-0 in our experimental setup , QTL mapping was performed on the whole dataset using MCQTL ( Multi-Cross QTL ) tool to compare allelic effects in a multicross design and potentially reveal shared QTLs . The combined QTL maps obtained ( Fig 5 for PRA29 , S4 Fig for RER16-29 , S5 Fig for Compactness29 ) highlight 10 independent loci , including at least 4 regions with contrasting allelic effect on PRA29: for instance , the middle of chromosome 1 region shows contrasted phenotypic consequences in different crosses , particularly when comparing Bur and Yo alleles ( with respect to Col ) . At this scale , it can be difficult to distinguish between different alleles at the same QTL and different QTLs . Conversely , combining the information of multiple crosses sometimes allows to predict narrower QTL intervals than with the initial QTL mapping , enabling the detection of distinct linked loci , for instance for PRA on the bottom half of chromosome 2; a location where the dynamic QTL analysis for PRA in CvixCol was already showing signs of 2 different segregating loci with slightly distinct dynamics over time . Another striking example is for Compactness on chromosome 5 , with neighbouring QTLs predicted to show opposite allelic effects ( e . g . BlaxCol ) . To confirm and investigate further the complexity of the genetic architecture of these traits , we used near-isogenic lines to mendelize QTLs and assess in more details the role of smaller chromosomal regions . Using 81 independent Heterogeneous Inbred Families ( HIFs ) scattered across the genome or chosen to decompose candidate regions in specific crosses [50] , we tested a total of 79 QTL effects from the 24 modalities of RIL set x trait x condition ( Fig 6 for PRA29; S6 Fig for RER16-29 and Compactness29 ) . Globally , 60% of the HIFs with a segregating region covering a candidate region previously identified showed significant effect with consistent direction , thus validating the QTL . Larger effect QTL were more often validated in HIF , with ~75% of the tested major or intermediate effect QTLs confirmed , compared to ~50% of the minor effect QTL . Specifically , for PRA29 trait in the two conditions studied ( Fig 6 ) , 28 QTLs were assessed with at least one HIF , among which 17 ( 60% ) were confirmed . Some ( minor-effect ) QTL x condition interactions detected in the RIL set were also significantly confirmed using HIF , such as the top of chromosome 5 locus in BurxCol which has significance ( PRA29 ) only under WW . We are also interested in positive HIF results that do not match with the results of the initial QTL mapping . This was particularly possible in regions with high HIF coverage and shows that several LOD score peaks were explained by more than one underlying QTL , as also suggested by the MCQTL analysis . A good example lies in the CvixCol cross , where the bottom half of chromosome 4 seemed to control PRA29 as a single locus in the initial QTL analysis ( Fig 3 ) and is now subdivided in at least 2 independent loci with opposite allelic effects after the HIF analysis ( Fig 6 ) . The 2 adjacent HIFs do not only show opposite allelic effect , but also different allelic effect amplitude on PRA29 ( -1 . 4cm2 versus +0 . 9cm2 ) , potentially explaining why only a Col-negative allelic effect was detected in the initial QTL mapping in this region , since the effect in this direction is stronger . Another example segregates in YoxCol at the top of chromosome 3 where 2 QTLs with opposite-effect on Compactness29 under WW conditions seem to localize closely according to the HIFs ( S6 Fig ) but wasn't detected at all in the QTL mapping at 29DAS ( Fig 3 ) and only remained significant at intermediate stages around 16DAS according to the dynamic analysis . Here , the 2 adjacent HIFs show very similar allelic effect amplitude on Compactness29 ( +0 . 0301 versus -0 . 0295 ) , likely explaining why no QTL was detected in this region for this trait . These could be examples of a lack of power of the RIL design to detect complex patterns due to the confusing effect of linked loci; alternatively , the confusing effect of epistasis could also interfere when comparing QTL mapping results from RILs and HIFs , because of the specific genetic background of each HIF . Finally , we can occasionally exploit the localization of the segregating region ( s ) of the tested HIFs to narrow down the candidate QTL region , like for a PRA29 QTL in CvixCol at the bottom of chromosome 2 which is narrowed down to the extremity of the chromosome and shown to be distinct ( = confirmed in non-overlapping HIFs ) from another nearby QTL segregating in the same cross ( with allelic effect in the same direction ) , as predicted by the multicross analysis above . Here again the dynamic analysis also helped to distinguish these loci based on their effect through time . Still , the 'precision' remains approximately at the Mb level at this stage . To tackle further the question of the complexity of the genetic architecture at a higher resolution than with simple HIFs , we decided to systematically dissect a region of 3Mb at the very beginning of chromosome 1 in CvixCol: in this region all previous approaches have predicted a single QTL with intermediate to major effect significance on PRA29 ( Fig 3 ) , which was confirmed in an HIF ( Fig 6 ) . This HIF was used to zoom in on the region across 30 bins ( stairs ) of ~100kb , defined by successive recombination breakpoints ( S4 Table ) and phenotypically evaluated individually in the same conditions as above ( an approach coined 'microStairs' ) . We selected this interval to test if a region harbouring a large-effect growth QTL may typically also include independent loci , maybe of smaller effect or hidden by the main QTL . A pairwise comparison of their growth phenotypes allows to test the impact of Cvi versus Col alleles over relatively short physical intervals of expected average size 100kb , with a maximum ~200kb , depending on recombination breakpoints location and marker interval . We either compared only the 2 successive recombinants ( 'stair by stair' ) , or we took advantage of the support from all pairwise comparisons ( 'staircase' ) to increase our power to detect QTL ( Fig 7 for PRA29 and S7 Fig for RER16-29 and Compactness29 ) . The strongest phenotypic effect for PRA29 in WW and WD ( but also–to a lesser extent–for RER16-29 in WD ) was fine-mapped down to the stair between recombinants 111 and 112 , covering potentially a physical interval ( bin ) extending to the maximum between 1 . 070 and 1 . 291 Mb ( an interval including 54 genes ) . In the middle of this region lies the obvious candidate gene CRYPTOCHROME 2 ( AT1G04400; CRY2 , a blue-light photoreceptor; S5 Table ) that is known to harbour a functional variant in Cvi ( a single amino-acid change ) and impact plant photomorphogenic development especially in short days [51 , 52] . This is a good test case of our approach , as it is very likely that CRY2 is primarily responsible for the growth difference overall observed in this HIF . Still , on either side of this locus , a few other QTLs also affect growth , underlying a much more complex genetic architecture than expected after the initial HIF results: in both watering conditions , a milder PRA29 QTL , with opposite allelic effect than CRY2 , was detected in the stair defined by recombinants 115 and 116 ( 1 . 495 to 1 . 703 Mb positions ) . Another PRA29 QTL was predicted in the bin 125/126 ( 2 . 509 to 2 . 710 Mb ) . For Compactness29 specifically under WD treatment , CRY2 did not seem to be causal and the causative locus for the observed HIF phenotype would most likely be in the bin 118/119 ( 1 . 801 to 2 . 027 Mb ) ; this QTL was not clear from the initial QTL analysis for this trait on Day 29 ( S6 Fig ) , but appeared significant earlier in the experiment ( cf dynamic analyses for CvixCol around 14DAS and later ) . There are probably even more loci , especially at the very beginning of the region for PRA29 ( stairs between recombinant 101 and 104 at least ) , but we reach the limits of our experimental design and phenotyping precision to be able to conclude accurately , with either too complex genetic architecture in this region or not enough recombinant lines to robustly support each intervals' effect . We then looked for high impact polymorphisms ( premature stop codon , frameshift or non-synonymous mutations ) likely affecting gene function between Col-0 and Cvi-0 within the most promising bins to identify candidate genes . Because of the numerous non-synonymous changes between these accessions ( S5 Table ) , we decided to arbitrarily filter the genes with the criteria of at least 3 non-synonymous mutations to increase our chance to detect genes with altered function or degenerated sequences after loss-of-function mutations . Here , we discuss some of those polymorphic genes where previous publications have indicated a putative function or effect that could relate to our phenotypes . Within the bin 115/116 lie at least 2 interesting candidates: URIDINE DIPHOSPHATE GLYCOSYLTRANSFERASE 74E2 ( AT1G05680; UGT74E2 ) is an auxin glycosyltransferase whose overexpression was shown to modify plant morphology and the size of the petioles , to delay flowering , and to increase drought tolerance [53] . The gene is also known for ample natural variation in expression , including potentially cis-acting variants [54]: http://www . bioinformatics . nl/AraQTL/multiplot/ ? query=AT1G05680 . GLUTAMATE RECEPTOR 3 . 4 ( AT1G05200; GLR3 . 4 ) is a calcium-dependent abiotic stimuli-responsive gene [55] expressed throughout the plant and impacting at least lateral root initiation [56] . It harbours several non-synonymous variants in Cvi-0 ( compared to Col-0 ) , but also higher expression in Col-0 as shown by several local-eQTLs in different crosses ( http://www . bioinformatics . nl/AraQTL/multiplot/ ? query=AT1G05200 ) confirmed to be cis-regulated through Allele-Specific Expression assay [57] ( reported in S6 Table ) . Within the bin 125/126 lie at least 2 interesting candidates: AT1G08130 encodes DNA LIGASE 1 ( LIG1 ) , a ligase involved in DNA repair , which mutation causes severe growth defects [58] . Its expression is also known to be controlled by a local-eQTL ( most likely cis-acting ) in LerxCvi: http://www . bioinformatics . nl/AraQTL/multiplot/ ? query=AT1G08130 AT1G08410 is DROUGHT INHIBITED GROWTH OF LATERAL ROOTS 6 ( DIG6 ) , encoding a large 60S subunit nuclear export GTPase 1 that impacts several developmental processes regulated by auxin , including growth [59] . Further work is required to prove any link between the observed phenotypic variation and these candidate genes .
Owing to its fine regulation throughout development and in interaction with the environment , plant growth represents a highly complex trait potentially controlled by numerous factors and interactions . Little is known on the actual genetic architecture of plant growth natural variation , with essentially a few genes of major effect being identified until now and only a few exceptions of more complex genetics revealed [44] . Association genetics hold great promise to dissect the underlying molecular bases of complex traits [12] , however one can wonder if the genetic architecture of highly integrative traits like growth or fitness is amenable to genome-wide association studies ( GWAS ) at the species scale: GWAS especially lacks power to decompose traits controlled by many loci of small effects when the underlying alleles have low frequency in the mapping population . For instance when exploiting worldwide collections of accessions , very little ( if any ) significant associations for growth parameters were found [60] , even when using very anti-conservative thresholds [2 , 61] or using morphological parameters with higher heritabilities [62] . Even with more targeted growth traits like root cell length , Meijon et al . did not detect any significant signal above the threshold , although the first peak just below the threshold identified a causal gene [63] . Overall , it is argued that linkage and GWA studies are complementary in the loci that they are able to reveal , depending on the genetic architecture of the trait in the population considered ( e . g . [21 , 64] ) . Here , by studying four different crosses to the reference Col-0 , we find many cross-specific loci , especially of mild effects , several of which might correspond to low frequency-alleles that would not likely be pictured in GWAS ( not enough power due to low frequency x effect size ) . Whether linkage or association mapping , these approaches are both similarly phenotyping-intensive and prone to interaction with uncontrolled environmental parameters ( increasingly so with the scale of the experiments ) . Using our high-throughput phenotyping robots to grow individual plants under tightly controlled conditions , we intended to dissect the genetic basis of plant growth under optimal and limiting watering conditions , to a level of accuracy rarely reached so far . We focused on vegetative growth from days 8 to 29 ( after sowing ) and selected three non-fully-correlated variables allowing to characterize plant growth dynamically: rosette area ( PRA ) , relative growth rate ( RER ) and compactness . PRA is a typical cumulative trait: what is observed on day n is not independent of what has occurred from day 1 to n-1 . Compactness , which basically represents a measure of PRA normalized by rosette width , is rather independent of cumulative phenomenons and hence shows much more age-specific QTLs; this is particularly striking when comparing the dynamic analyses for these traits . Finally , RER integrates growth rates over a specific period of time , and is independent of plant size ( = relative ) . Because estimated during the exponential phase of growth , RER is very much similar to what is obtained by fitting exponential models to the PRA data and exploiting model parameters [2] . Hence , growth-related phenotypes , depending on how they are exploited , will present different genetic architectures throughout age , with individual loci making different contributions to cumulative or age-specific traits . It has previously been shown that heritabilities for growth-related traits change over time [6 , 65]; one possible explanation is that QTLs are more or less likely to act at specific time points . For instance , studying growth dynamics in maize [66 , 67] and root tip growth in Arabidopsis [68] allowed to identify marker-trait associations that would not be detectable by considering the cumulative trait only at a single ( final ) time point . At the other end of time-resolution for growth , going into much more details of the dynamics ( several images per day ) may result in noisy raw data requiring further treatment before exploiting , for example due to projected growth estimates interfering with circadian leaf movements [65] . Our work has been performed under two environmental conditions , a control condition and one that moderately limits growth due to water ( but not nutrient ) availability [34] , i . e . a mild drought treatment . The QTL profiles obtained at the genomic scale are very robust to mild drought with most of the large effect QTL showing no clear signs of interaction with water availability in our conditions . Some of them still change their level of significance with conditions , but it is difficult to know if this is a real interaction with drought , a change in trait variance under stress , or a change in the rest of the genetic architecture of the trait ( which will impact the significance of individual QTL ) . Condition-specific QTLs detected here always are of small effect , which also raises the question of the power of these comparisons across different conditions/experiments due to false negative in mapping QTL . Still , this result ( a relatively smaller part of the phenotypic variation is plastic rather than constitutive ) is similar to what was found previously for instance in linkage [69] and association mapping [61] , showing an overlap in the network of genes that regulate plant size under control and mild drought conditions [60] . Drought stress might have pleiotropic effects on different tightly interrelated phenotypic traits and impose strong constraints on them , reducing trait variability [31] . Of course , this may highly depend on the type of stress ( intensity , stage of application , stability … ) that is applied . Here , we chose a mild stress intensity , to remain physiologically relevant and avoid the squeezing of trait variation concomitant to strong stress levels . Also , our robot is compensating ( twice a day ) the individual plant size effect on transpiration that may otherwise artificially increase stress intensity according to intrinsic plant size difference . One advantage of using a star-like cross design , increasingly used in nested association mapping ( NAM ) populations , is to combine the power of individual crosses and take advantage of the comparison of multiple alleles , with respect to the reference allele in order to identify allelic variants more efficiently [70–73] . These could correspond to variants originating from Col-0 or to shared allelic variants among the other parents ( same direction of the allelic effect in all crosses ) , or to allelic series ( other parents could have divergent direction/intensity of effect with respect to the reference parent ) . This multi-population study has confirmed the effect of several loci across traits and environments , with particular power for compactness , and in some instances already allows to predict that independent loci actually underlie major peaks . However , this approach is still limited by the mapping resolution which makes it difficult to distinguish shared variants from linked loci . Considering the precision of our phenotyping ( which has an impact on the part of phenotypic variation that is amenable to genetic dissection ) , the need for higher-resolution approaches to better describe the trait's genetic architecture is obvious here . If the architecture of variation in our crosses is more complex than just a limited number of loci independently segregating for intermediate or major effects , then the density of recombination observed in a simple RIL set will not allow to decipher the full architecture [74 , 75] . We first went deeper in resolution by phenotyping numerous pairs of near-isogenic lines ( actually , HIFs ) that each interrogate a specific portion of a chromosome ( 2-3Mb on average ) in an otherwise fixed genetic background . This nicely confirms a majority of loci but also shows some effects that were unexpected after the initial QTL mapping results , already indicating more complex genetic architecture than anticipated; this includes single peaks splitting up in independent loci or complex patterns of linkage versus pleiotropy ( when comparing different traits ) and linkage versus GxE ( when comparing traits in different conditions ) . It seems that QTL colocalization among crosses ( linkage versus shared variants ) is also often questioned , although this requires to be able to compare multiple HIFs showing positive and/or negative results , which can be difficult for several reasons . Indeed , many factors can explain that a QTL is not validated in an HIF: the QTL could be under epistatic interaction with another locus ( thus , a specific HIF may not represent the adequate genetic background ) , the QTL could be mislocalised by QTL mapping ( thus , out of the HIF segregating region ) or the HIF harbours a more complex genotype than expected at the segregating region , such as a double recombined region ( thus , the HIF doesn't actually allow to test the whole region ) . This makes the comparison between HIFs difficult , even in the same cross , and negative HIF results should particularly be interpreted carefully . In this context , the rate of QTL validation obtained here is rather satisfying . The difficulties to identify genes responsible for complex phenotypes also depends on their involvement in epistatic interactions . HIFs are particularly sensitive to epistasis ( compared to traditional NILs ) as they each harbour a different genetic background , which means that they allow epistasis to be interrogated providing that one can test enough independent HIFs , otherwise they have to be compared with care . Epistatic interactions are detected in all crosses / traits ( although not always with very high significance ) even when the trait heritability or variation is not so high in a cross , illustrating another factor of the complex genetic architecture . Some interactions seem to be condition-specific , but power issues are likely to be limiting in understanding these patterns of GxGxE . Furthermore , HIF have an expected candidate segregating region of several Mb usually , so our observations are likely just a glimpse at the real complexity of growth as it is known that linkage and epistasis is also active at a very local scale [44] . Still , the sensitivity of our approach here is validated by the detection of QTL colocalizing with several already-known QTG expected to segregate in our crosses , like CRY2 as discussed above [52 , 76] , MPK12 which would explain nicely the bottom of chromosome 2 QTL in CvixCol [77 , 78] , or SG3 detected here through its epistasis with SG3i in BurxCol [49 , 79] . To avoid genome-wide epistasis and better reveal local-scale architecture , we have investigated in further details a single HIF background for a specific 3Mb region containing a known QTG of large and pleiotropic effect ( CRY2 ) in an original approach . Our analysis reveals that there are at least 3 other QTGs in this interval controlling one of the traits in at least one condition . For PRA29 , the picture seems to be even more complex with traces of at least one more locus; here , it seems that phenotyping accuracy becomes again limiting after all . Obviously these loci with opposite allelic effects , different patterns of pleiotropy and interaction with the environment , and just a few hundreds of kb from each other , remain cryptic in simple QTL mapping . This major result of local-scale independent complex genetic architecture for different traits and conditions should lead us to a lot of caution when interpreting colocalizing QTLs from different traits / conditions / age , as these may very well be independent loci rather than a single pleiotropic locus , as shown here for PRA and Compactness . If we were to extrapolate the figures obtained on this specific region in this particular cross to the genome- and species-scale , we would expect hundreds of causative loci of detectable phenotypic effect controlling these growth-related phenotypes . One way to approach these individual loci would be to decompose their independent signature based on different dynamics or underlying traits ( transcriptomics , metabolomics … ) in a 'systems genetics' strategy [54 , 60 , 80] . Complex genetic architecture as revealed in this study has consequences on quantitative genetics experimental design and interpretation , arguing in favor of linkage mapping or GWAS depending on the balance between genetic complexity due to linked loci ( where association is expected to behave better than linkage mapping ) and genetic complexity due to small effect/rare alleles ( where association will behave poorly ) . Other intermediate experimental designs like multiparental populations or nested-association mapping should bring more power [12 , 81] . Resolution is improved by pushing recombination densities to its limits and it was shown to help resolve more complex genetic architecture in yeast [82] . In plants , using 'hyper-recombinant' mutations to generate new segregating populations could also be a strategy in the future [83] .
The 4 RIL sets used for this work were generated at the Versailles Arabidopsis Stock Center , France ( http://publiclines . versailles . inra . fr/ ) and were either described previously [47] or on the Publiclines website where all relevant information ( description and associated data ) is gathered . Versailles stock center ID are indicated as 'xxxAV' and 'xxRV' for Accessions and RILs respectively ) . They are derived from crosses between the following pairs of accessions , chosen to maximize genetic and phenotypic diversity [84]: RIL set 'BlaxCol' ( ID = 2RV ) : Bla-1 ( 76AV ) x Col-0 ( 186AV ) / 259 RILs RIL set 'CvixCol' ( ID = 8RV ) : Cvi-0 ( 166AV ) x Col-0 / 358 RILs RIL set 'BurxCol' ( ID = 20RV ) : Bur-0 ( 172AV ) x Col-0 / 283 RILs RIL set 'YoxCol' ( ID = 23RV ) : Yo-0 ( 250AV ) x Col-0 / 358 RILs HIFs have been selected in each RIL set to cover several regions of the genome , some of which are expected to segregate for QTLs while others were chosen at random locations . As described previously [85] they are derived from the progeny of one RIL which is heterozygous only at the locus of interest . Hence , one HIF family is composed of 2 or 3 lines fixed for one parental allele at the segregating region and 2 or 3 lines fixed for the alternate parental allele at the segregating region , in an otherwise identical genetic background . Each HIF ( ID = 'xxHVyyy' ) is named after the RIL ID ( 'xxRVyyy' ) from which it has been generated ( 'xx' is the ID of the RIL set , 'yyy' is the ID of the RIL ) : for example , the family 2HV142 is fixed from the RIL 2RV142 . The tentative QTL validation is based on the phenotypic comparison of these fixed lines within the HIF family . We have used 81 HIF families . The complete dataset gathering genotypic information on the RILs used to generate HIFs has been submitted to INRA institutional data repository ( https://data . inra . fr ) [50] , with the genotypic conventions and ID from the Versailles Arabidopsis Stock Center ( http://publiclines . versailles . inra . fr/ ) . The progeny of CvixCol RIL 8RV294 ( also used to generate HIF family 8HV294 ) , segregating for the first 3Mb of chromosome 1 , has been screened to detect recombined individuals . 4000 individuals were genotyped with markers at the edge of the heterozygous region and then 29 evenly distributed recombinants were selected using markers spaced every ~100kb . These recombinants were genotyped and fixed for the remaining segregating region in such a way that they each differ genotypically from the next recombinant by a ~100kb bin on average ( S4 Table ) . Similarly , as for the HIF , the descendance of 3 similarly-fixed lines are saved and phenotyped to account for possible maternal environmental effect . This is similar to the approach taken by Koumproglou et al . [86] , but at a much finer scale , hence the name 'microStairs' . Phenotyping was performed on the Phenoscope robots as previously described [34] ( https://phenoscope . versailles . inra . fr/ ) . Every RIL set and their respective parental accessions have been phenotyped in 2 independent Phenoscope experiments ( = biological replicates ) , except for CvixCol ( 3 biological replicates ) , with 1 individual ( plant ) per RIL per condition . In short , the peatmoss plugs' soil water content ( SWC ) is gradually adjusted for each plant individually as a fraction of the initially-saturated plug weight . We worked at 2 watering conditions: 60% SWC for non-limiting conditions ( called 'WW' for well-watered ) and 30% SWC for mildly growth-limiting watering conditions ( called 'WD' for water deficit ) . The growth room is set at a 8 hours short-days photoperiod ( 230 μmol m-2 sec-1 ) with days at 21°C/65%RH and nights at 17°C/65%RH . A picture of each individual plant is taken every day at the same day-time and a semi-automatic segmentation process ( with some manual corrections when required ) is performed to extract leaf pixels . From this we exploit different traits: Projected Rosette Area ( PRA ) , circle radius , convex hull area , average Red , Green and Blue components ( leaf pixels , RGB colour scale ) , and derived phenotypic traits are calculated , such as the compactness ( ratio PRA / convex hull area ) and the Relative Expansion Rate ( RER ) over specific time windows ( Fig 1 ) , as previously described [34] . The complete raw phenotypic dataset has been submitted to INRA institutional data repository ( https://data . inra . fr ) [48] . Principal component analysis ( PCA ) was performed using the ade4 R package based on phenotypic data from all the RIL sets in WW and WD conditions . Heritabilities were calculated based on the mean squares ( MS ) of the following ANOVA model , used as estimators of genetic and residual variances: Yij∼μ+αi+εij where Yij is the phenotypic value , μ is the mean , αi is the Genotype factor and εij is the Residuals . The genetic variance VarG was estimated by ( MSα - MSε ) /n ( where n is the number of replicates ) . The residual variance VarR was estimated by MSε . h2 = VarG/ ( VarG+VarR ) . Then , for QTL detection and further analyses , the phenotypic values were corrected for experiment effects . Corrected phenotypic values were calculated using the intercept ( μ ) , the condition ( α ) , genotype ( γ ) , and genotype*condition ( σ ) effects of the following linear model: Yijkl∼μ+αi+βj+γk+δij+λjk+σik+εijkl where Yijkl: phenotype; μ: mean; αi: effect of the condition; βj: effect of the experiment; γk: effect of the genotype; δij: effect of the interaction condition*experiment; λjk: effect of the interaction experiment*genotype; σik: effect of the interaction condition*genotype; εijkl: residuals . The interaction term ( independently used for QTL mapping ) refers to the condition*genotype interaction component . Note that one of the replicates of the CvixCol phenotypic data was already analyzed for QTL mapping -with a different statistical model- in Tisné et al . [34] . Similarly , the set of near isogenic lines ( microStairs ) were phenotyped and analyzed from 3 full biological replicates in independent Phenoscope experiments . QTL detections were performed using Multiple QTL Mapping algorithm ( MQM ) implemented in the R/qtl package [87 , 88] using a backward selection of cofactors . At first , genotype missing data were augmented , then one marker every three markers were selected and used as cofactors . Important markers were selected through backward elimination . Finally , a QTL was moved along the genome using these pre-selected markers as cofactors , except for the markers in the 25 . 0 cM window around the region of interest . QTL were identified based on the most informative model through maximum likelihood . According to permutation results ( computed with the mqmpermutation ( ) function in R/qtl ) , a general LOD threshold of 2 . 4 was chosen for all QTL maps to ensure a FDR below 0 . 05 and remain conservative . Interactive QTL maps for time-course series were generated using the R/qtl charts package [89] . All QTL positions were projected on the consensus genetic map of the 4 crosses built with R/qtl ( this is the map shown along the x axis on Figs 3 , 5 and 6 ) . A joint genotype dataset was constructed with ‘A’ alleles coding Col alleles ( the common parent ) , ‘B’ alleles for non-Col alleles , and monomorphic markers in a cross coded as missing . The linkage groups were considered known from the individual maps and the physical position of markers , and a first marker order was calculated using orderMarkers ( ) function . In case of conflicting marker order between individual , physical and consensus maps , the function switch . order ( ) was used to retain the most probable order ( i . e with the lowest number of recombination ) . The consensus genetic map markers' identities and positions are available together with the HIF dataset in Dataverse [50] . Epistatic interactions were identified using the scantwo ( ) function of the R/qtl package . LOD scores were calculated for additive , interaction and full models for all pairwise combination of markers , except for adjacent markers and a general conservative LOD threshold ( 3 LOD ) was determined from permutations . Effect plots for the pairs of markers were drawn using the R/qtl package . QTL mapping in the multi-cross design was performed with the MCQTL package [90] . The model was described as additive ( no dominance effect ) and connected ( Col-0 centered design ) and the following 3 steps process was applied . Step 1: thresholds were calculated by trait on the whole genome using 1000 resampling replicates ( PRA29_WW = 3 . 73; PRA29_WD = 3 . 84; RER16-29_WW = 4 . 22; RER16-29_WD = 3 . 95; Compactness29_WW = 3 . 64 Compactness29_WD = 3 . 30 ) . Step 2: QTL detection was performed using iQTLm method with a general and conservative threshold of 4 LOD . To perform this detection , cofactors were automatically chosen by backward selection with a threshold of 2 . 8 LOD among a skeleton with a minimal inter distance of 10cM . Search for QTL was not allowed within +/-10cM window surrounding the QTL to avoid linked genetic regions . Step 3: model estimations were performed for each trait and condition using the shared QTL positions identified at step 2 . The phenotypes of the recombined HIFs lines were modeled using the following linear equation: Yij∼μ+αi+βj+εij where Yij is the value of the phenotype; μ is the mean of the phenotype; αi is the effect of the stair ( bin ) i; βj is the effect of the line j ( maternal replicate within each stair ) and εij is the residuals . An anova was performed with this model and the p-value of the stair effects were adjusted by a Benjamini-Hochberg correction . Polymorphic candidate genes ( Cvi versus Col ) were listed for each PRA29 'microStairs' significant interval according to variants listed on the 1001Genomes website ( http://1001genomes . org/ ) , through the Polymorph1001 tool . Differentially cis-regulated variants were extracted from Cvi/Col Allele-Specific Expression ( ASE ) data [57] and from CvixCol local-eQTLs data [91] across the whole region . Two datasets have been submitted to INRA Dataverse repository ( https://data . inra . fr ) . This dataset gathers the main raw phenotypic data obtained and exploited in Marchadier , Hanemian , Tisné et al . ( 2018 ) . It contains data from 4 RIL sets across 9 Phenoscope experiments . For each Phenoscope experiment , Recombinant Inbred Line ( RIL ) and Condition ( 'WW' = Well Watered / 'WD' = Water Deficit ) , the data set indicates the phenotypic value for 6 traits at 21 successive time points . 'Trait . XX' = Trait at XX days after sowing , with 'XX' = 09 to 29 and 'Trait' = PRA ( Projected Rosette Area; in cm2 ) , GreenMean / RedMean / BlueMean ( rosette pixels' colour components; arbitrary unit ) , ConvexHullArea ( area of the convex hull encompassing the rosette; in cm2 ) and CircleRadius ( radius of the smallest circle encompassing the rosette; in cm ) . RIL set IDs and RIL IDs are according to Publiclines http://publiclines . versailles . inra . fr/rils/index Each row represents a single HIF and the genotype of the F7 RIL it originates from is indicated along the chromosomes with RIL ID , markers and genotypic conventions from Publiclines http://publiclines . versailles . inra . fr/rils/index ( i . e . 'A' = Col allele; 'B' = alternate parental allele; 'C' = heterozygous ) . The region highlighted in yellow is the segregating region that is tested in the HIF family through several fixed lines for each parental allele . For each of the 3 growth traits ( Compactness29 = rosette compactness 29 days after sowing; PRA29 = Projected Rosette Area 29 days after sowing; RER16-29 = Relative rosette Expansion Rate between days 16 and 29 after sowing ) in 2 conditions ( 'WW' = Well Watered / 'WD' = Water Deficit ) , whenever significant , the p-value of the comparison between allelic lines ( 'Pval' ) and the direction of the allelic effect ( 'sign' calculated as [Col-Xxx] where Xxx is the alternate parental allele ) are indicated in the last columns of the table . A specific webpage is associated with this work to display interactive graphes for dynamic QTL analyses at http://www . inra . fr/vast/PhenoDynamic . htm
|
The question of the complexity of the genetic variants underlying diversity in plant size and shape is central in evolutionary biology to better understand the impacts of selection and adaptation . In this work , we have combined the high resolution of a robotized platform designed to grow Arabidopsis plants under strictly-controlled conditions and the power of quantitative genetics approaches to map the individual genetic components ( the 'QTLs' ) controlling diverse phenotypes , and hence reveal the so-called 'genetic architecture' of these traits . We show that the more we increase our resolution to map QTLs , the more complex of a genetic architecture we reveal . For instance , by focusing all of our mapping power on a small region representing 2 . 5% of the genome in an unprecedented phenotyping effort , we reveal that several independent QTLs had remained hidden in this region beyond a major-effect QTL that is always clearly visible . If this region is representative of the genome , this means that our current understanding misses potentially hundreds of variants finely controlling traits of evolutionary or agronomical interest .
|
[
"Abstract",
"Introduction",
"Results",
"Materials",
"and",
"methods"
] |
[] |
2019
|
The complex genetic architecture of shoot growth natural variation in Arabidopsis thaliana
|
Viral replication efficiency is in large part governed by the ability of viruses to counteract pro-apoptotic signals induced by infection of host cells . For HHV-8 , viral interferon regulatory factor-1 ( vIRF-1 ) contributes to this process in part via inhibitory interactions with BH3-only protein ( BOP ) Bim , recently identified as an interaction partner of vIRF-1 . Here we recognize that the Bim-binding domain ( BBD ) of vIRF-1 resembles a region ( BH3-B ) of Bid , another BOP , which interacts intramolecularly with the functional BH3 domain of Bid to inhibit it pro-apoptotic activity . Indeed , vIRF-1 was found to target Bid in addition to Bim and to interact , via its BBD region , with the BH3 domain of each . In functional assays , BBD could substitute for BH3-B in the context of Bid , to suppress Bid-induced apoptosis in a BH3-binding-dependent manner , and vIRF-1 was able to protect transfected cells from apoptosis induced by Bid . While vIRF-1 can mediate nuclear sequestration of Bim , this was not the case for Bid , and inhibition of Bid and Bim by vIRF-1 could occur independently of nuclear localization of the viral protein . Consistent with this finding , direct BBD-dependent inactivation by vIRF-1 of Bid-induced mitochondrial permeabilization was demonstrable in vitro and isolated BBD sequences were also active in this assay . In addition to Bim and Bid BH3 domains , BH3s of BOPs Bik , Bmf , Hrk , and Noxa also were found to bind BBD , while those of both pro- and anti-apoptotic multi-BH domain Bcl-2 proteins were not . Finally , the significance of Bid to virus replication was demonstrated via Bid-depletion in HHV-8 infected cells , which enhanced virus production . Together , our data demonstrate and characterize BH3 targeting and associated inhibition of BOP pro-apoptotic activity by vIRF-1 via Bid BH3-B mimicry , identifying a novel mechanism of viral evasion from host cell defenses .
Human herpesvirus 8 ( HHV-8 ) specifies a number of proteins expressed during the lytic cycle that have demonstrated or potential abilities to promote virus productive replication via inhibition of apoptotic pathways induced by infection- or replication-induced stress . These proteins include membrane signaling receptors K1 and K15 [1]–[3] , Bcl-2 and survivin homologues encoded by open reading frames 16 and K7 [4]–[7] , viral chemokines vCCL-1 and vCCL-2 [8] , and viral G protein-coupled receptor ( vGPCR ) [9] , [10] . The viral interferon regulatory factor homologues , vIRFs 1–4 , also are believed to play important roles in blocking interferon and other stress responses to virus infection and replication . Their functions include inhibitory interactions with cellular IRFs , IRF-activating pathways , and/or IRF-recruited p300/CBP transcriptional co-activators to IRF-stimulated promoters [11]–[15] . Additionally , the vIRFs inhibit apoptosis via targeting of other cellular proteins; these include p53 ( vIRFs 1 and 3 ) [16]–[18] , p53-activating ATM kinase ( vIRF-1 ) [19] , p53-destabilizing MDM2 ( vIRF-4 ) [20] , retinoic acid/interferon-inducible protein GRIM19 ( vIRF-1 ) [21] , and TGFβ receptor-activated transcription factors Smad3 and Smad 4 ( vIRF-1 ) [22] . To date , the v-chemokines , vGPCR and vIRF-1 are the only HHV-8 proteins that have been demonstrated both to inhibit apoptosis in lytically infected cells and to promote HHV-8 productive replication , in the context of lytic reactivation in endothelial cells in the case of the vCCLs and vIRF-1 and additionally in primary effusion lymphoma ( PEL ) cells for vGPCR [23] , [8] , [10] . In addition to its other cellular binding partners , vIRF-1 also interacts with the pro-apoptotic BH3-only protein ( BOP ) Bim [23] , a protein also targeted for suppression by v-chemokine signaling and demonstrated to be both induced during lytic replication and a very powerful negative regulator of viral replication efficiency [8] . Bim , like other BOPs , functions by virtue of its BH3 domain to target anti-apoptotic members of the Bcl-2 family and to disrupt their interactions with apoptotic executioner proteins Bax and Bak , liberating them for oligomerization and mitochondrial permeabilization [24] , [25] . However , Bim can also interact with and activate Bax and Bak directly , via induced conformational changes [26]–[28] . This property of direct activation of Bax and/or Bak is shared by BOPs Bid and Puma , although other BOPs appear to act indirectly via BH3-mediated interactions with Bcl-2-family proteins [26] , [27] , [29] , [30] . Activities of several BOPs , such as Bim , Bmf and Bad , are regulated via phosphorylation , to effect activation , inactivation , or alteration of protein stability [31]–[33] . For example , Bim is activated by JNK-mediated phosphorylation of residue T56 , causing release of Bim from microtubules , inactivated by Akt phosphorylation of S87 , which allows 14-3-3 association and cytosolic sequestration , and ERK phosphorylation of S69 to effect proteasomal degradation of Bim [34] , [32] , [35] . Bid is unique among BOPs in its activation via protease cleavage , typically by death receptor-activated caspases but also by other proteases , such as granzyme B [36]–[38] . Cleavage removes N-terminal sequences ( p7 ) containing a motif , termed BH3-B , which interacts intramolecularly with the BH3 domain to inhibit Bid pro-apoptotic activity [39] , [40] . Mitochondrial membrane targeting of cleaved , truncated Bid ( tBid ) is promoted via surface exposure of hydrophobic residues and N-terminal glycine myristoylation [41] , [42] . While the nature of BH3 interactions with Bcl-2 family proteins , involving BH3 α-helix association with Bcl-2 BH1–3-comprised hydrophobic grooves , is well characterized [43] , [44] , the basis of Bid BH3∶BH3-B interaction is at present poorly understood [40] . Our previous studies of vIRF-1 interaction with Bim identified a unique mechanism of Bim regulation , via nuclear sequestration of the BOP away from mitochondria , and documented the first example of interaction between a Bcl-2 family member and an IRF homologue [23] . While the Bim-binding domain ( BBD ) of vIRF-1 was mapped , to residues 170–187 comprising a putative amphipathic α-helix , the region of Bim interacting with vIRF-1 was not determined . Subsequent comparisons of the BBD primary and predicted secondary structures with those of Bid BH3-B revealed similarities , in terms of conserved residues , α-helical structure and amphipathicity , indicating , by analogy , that BBD may target for binding and direct functional inhibition the BH3 domain of Bim ( in addition to enabling vIRF-1-mediated inactivation of Bim via nuclear sequestration ) , and indeed may be able to bind Bid BH3 also . The data presented here demonstrate that this is indeed the case , that vIRF-1 can inhibit Bid as well as Bim pro-apoptotic activity , and that BBD can also recognize the BH3 domains of certain other BOPs , dependent on residues that are common and particular to these domains . Thus , vIRF-1 BBD mediates BH3-B mimicry , to our knowledge the first example of viral usurpation of this mode of inhibition of BOP function and pro-apoptotic signaling .
We noted previously the amphipathic α-helical structure of the Bim-binding domain ( BBD , residues 170–187 ) of vIRF-1 [23] . This type of structure also is apparent in the so-called BH3-B domain of Bid , which interacts intramolecularly with the BH3 domain to effect inhibition of Bid activity [40] . Indeed , the sequences of BBD and BH3-B are similar to each other and to BH3 domains of other proteins ( Fig . 1 ) . Particularly noteworthy are the BH3-conserved hydrophobic and BH3-B-conserved basic residues of the BBD core sequence ( Fig . 1 ) . The structural similarities of BBD and BH3-B suggested the possibility that BBD might interact with Bim via its BH3 domain and that vIRF-1 might target Bid ( and possibly other BOPs ) in addition to Bim . To test whether the BBD of vIRF-1 interacted with the BH3 domain of Bim , recombinant fusion proteins were made for co-precipitation binding assays . The proteins comprised T7/DsRed-fused wild-type and Bim-binding-refractory GK179AA versions of BBD ( vIRF-1 residues 170–187 ) , and also Bid BH3-B ( BidL residues 34–51 ) , and chitin-binding domain ( CBD ) -tagged GFP-Bim BH3 ( BimEL residues 148–161 ) and GFP ( control ) . Paired T7/DsRed and GFP/CBD fusion proteins were mixed , CBD-tagged proteins precipitated with chitin beads , and precipitated material analyzed by SDS-PAGE and immunoblotting . This experiment revealed binding of BBD , but not BBD ( GK179AA ) or BH3-B , to Bim BH3 , with no detectable background binding to negative control GFP-CBD ( Fig . 2A ) . An analogous experiment using GFP/CBD-fused Bid BH3 ( residues 86–99 ) as the “bait” identified interaction with Bid BH3-B ( positive control ) and also with vIRF-1 BBD ( Fig . 2B ) , thereby identifying Bid BH3 , as well as Bim BH3 , as a target of vIRF-1 BBD interaction . To confirm interaction of full-length vIRF-1 and Bid proteins , as we had done previously for vIRF-1 and Bim [23] , appropriate expression plasmids were used to transfect HEK293T cells , and cell lysates were used for co-precipitation assays . Immunoprecipitation of Flag epitope-tagged Bid , as well as Bim ( positive control ) , enabled co-precipitation of vIRF-1 , demonstrating interaction between vIRF-1 and Bid ( Fig . 2C ) . The relationship between vIRF-1 BBD and Bid BH3-B was tested by substitution of the latter with the former in the context of full-length , uncleaved Bid ( BidL ) and testing the constructions ( Fig . 3A ) for pro-apoptotic activities in appropriately transfected cells . Apoptotic activity of Bid was measured by a GFP-based assay , in which loss of GFP fluorescence in GFP vector-cotransfected cells correlates with loss of cell viability and corresponds to rates of apoptosis , e . g . as measured by TUNEL assay [23] . Transfection of BidL and GFP expression vectors into HEK293T cells led to substantially reduced GFP fluorescence ( ∼39% ) relative to empty vector plus GFP control , set at 100% ( Fig . 3B ) , consistent with previously reported apoptotic activity of uncleaved Bid [45] , [40] . However , this activity was increased substantially by introduction of Bid-BH3 binding-abrogating mutations ( GHE41VLA [40] ) into BH3-B , suppressing GFP fluorescence to ∼20% [Fig . 3B , Bid ( mBH3-B ) ] . Importantly , BBD could substitute fully for BH3-B in this assay , inhibiting BH3-mediated Bid apoptotic activity more effectively than native BH3-B ( 75% GFP fluorescence relative to 39% ) . Increased inhibition by BBD indicates that it may bind with higher affinity than BH3-B to Bid BH3 . Although it is more likely that BBD and BH3-B mediated inhibition of BidL activity occur via intramolecular interactions with Bid BH3 , it is also possible that trans-inhibition may occur . Introduction of mutation GK179AA ( previously shown to abrogate Bim interaction [23] ) into BBD ( mBBD ) abolished its inhibition of BidL activity , leading to GFP fluorescence ( cell viability ) levels similar to those obtained upon mutation of BH3-B . Similar transfections with these constructions were undertaken to confirm apoptotic inhibitory activity of BBD in the context of BidL . Annexin V-Cy3 staining ( Fig . 3C ) and cytochrome c release assays ( Fig . 3D ) were employed to quantify apoptosis and to monitor induction of the apoptotic pathway , respectively . The results derived from annexin V-Cy3 staining mirrored those obtained from the GFP-based assay ( Fig . 3B ) , demonstrating full functional substitution of Bid BH3-B by vIRF-1 BBD . Apoptosis induced by tBid , included in this experiment , was notably higher than that of BidL , BidL-mBH3-B and BidL-mBBD , as expected because of the complete absence of the inhibitory N-terminal region of Bid and efficient mitochondrial targeting of the truncated form . Congruent results were obtained from the cytochrome c release assays , confirming the ability of BBD to substitute functionally for BH3-B in the context of BidL ( Fig . 3D ) . Combined , the data presented in Figure 3 demonstrate functional equivalence of Bid BH3-B and vIRF-1 BBD and provide further evidence , in a biologically relevant context , of BBD interaction with Bid BH3 . In HHV-8 lytically reactivated endothelial cells , Bim is found predominantly in the nucleus , and nuclear location of Bim can be induced by vIRF-1 in transfected HEK293T cells [23] . As nuclear-localized Bim is inactive in respect of apoptotic induction , its nuclear sequestration represents a mechanism of BOP inactivation . To determine the nuclear-cytoplasmic distribution of Bid during HHV-8 lytic reactivation , dual-label immunofluorescence assays ( IFA ) were undertaken to identify Bid induction and distribution in reactivated cells expressing lytic antigen ( vIRF-1 ) . Like Bim , Bid was induced during lytic reactivation in telomerase-immortalized endothelial ( TIME ) cells [46] , here engineered to express HHV-8 immediate-early protein RTA in response to doxycycline ( see Materials and Methods and Fig . S1 ) ( Fig . 4A ) . However , little or no nuclear localization of Bid was apparent , in sharp contrast to the predominant nuclear localization of Bim in lytically reactivated cultures [23] ( Fig . 4A ) . In cells transfected with BidL or tBid expression vectors together with an empty or vIRF-1 expression plasmid , the nuclear-cytoplasmic distribution of each Bid protein was refractory to vIRF-1 influence ( Fig . 4B ) . It is notable that some nuclear localization of BidL was apparent , consistent with previous reports of nuclear localization and associated activities of Bid [47]–[49] , but no nuclear staining was evident for tBid . In contrast to Bid , and consistent with previous findings [23] , Bim distribution was altered in the presence of vIRF-1 , with strong nuclear staining apparent exclusively with vIRF-1 co-expression ( Fig . 4B ) . That vIRF-1 indeed did not influence nuclear-cytoplasmic distribution of Bid was verified by using immunoblotting of cytoplasmic and nuclear fractions of transfected cells . Again , while nuclear localization of a proportion of BidL was detected , this was not detectably influenced by vIRF-1 co-expression , and tBid localization was restricted to the cytoplasm in the absence and presence of vIRF-1 ( Fig . 4C ) . Furthermore , a nuclear localization-defective vIRF-1 variant ( see below ) also did not influence the nuclear-cytoplasmic distribution of BidL , although induction of Bim nuclear localization was abolished ( Fig . 4C ) . Targeting of BH3 domains of Bim and Bid by vIRF-1 , coupled with the inability of vIRF-1 to induce significant Bid nuclear localization , suggested the possibility of direct inactivation of BOP apoptotic activity by vIRF-1 binding . To address this issue , we generated a nuclear localization-defective version of vIRF-1 for use in functional assays . Each of four potential nuclear localization signals ( NLS ) was mutated ( Fig . 5A ) , and the respective vIRF-1 proteins were tested for nuclear localization in expression vector-transfected cells . vIRF-1 ( RGRRR163AGAAA ) was found to be defective for nuclear localization , as determined by IFA ( Fig . 5B ) . This was verified using a functional assay based on p53-inhibitory activity of vIRF-1; while wild-type vIRF-1 was able to suppress reporter-detected transactivation by nuclear p53 , NLS-mutated vIRF-1 was completely inactive ( Fig . 5C ) . As mentioned above , the vIRF-1 variant was also confirmed to be unable to induce nuclear localization of Bim ( Fig . 4C ) . The wild-type and NLS-mutated versions of vIRF-1 were used in GFP-based viability assays to compare their abilities to inhibit Bim and Bid activities . Both vIRF-1 proteins were able to protect cells from Bim- and Bid-induced apoptosis , with very similar dose-activity profiles ( Fig . 5D ) . Therefore , vIRF-1 inhibition of both Bim and Bid can be mediated independently of nuclear-localization and nuclear-localized functions of vIRF-1 . To address the hypothesis that vIRF-1 may act directly at the mitochondrion to suppress BOP-induced apoptosis , we isolated mitochondria from vIRF-1 vector-transfected HEK293T cells and undertook SDS-PAGE and Western analysis for detection of vIRF-1 in the mitochondrial fraction . Both wild-type and Bim-refractory ( GK179AA ) vIRF-1 proteins were present in mitochondrial fractions , representing approximately 2% of total vIRF-1 present in the transfected cell lysates ( Fig . 6A ) . To assess whether this vIRF-1 was likely to be membrane inserted/associated or mitochondrial outer membrane ( OM ) protein associated , in vitro mitochondrial binding assays were used . These assays employed purified recombinant vIRF-1 ( T7 epitope-tagged ) added to isolated mitochondria prior to and after proteinase K treatment . Binding of vIRF-1 to mitochondria , apparent absent treatment , was completely abrogated by proteinase K pre-treatment ( Fig . 6B ) , indicating mitochondrial association of vIRF-1 via interaction with a cytoplasmic-exposed mitochondrial protein . Furthermore , endogenously-expressed vIRF-1 was present in mitochondrial fractions prepared from HHV-8+ primary effusion lymphoma ( PEL ) cells , BCBL1-TRE/RTA [50] , with or without lytic induction ( +Dox ) , and this vIRF-1 was susceptible to proteinase K digestion , demonstrating peripheral association of vIRF-1 with mitochondria in HHV-8 infected cells ( Fig . 6C ) . vIRF-1 has been reported previously to be expressed during latency in PEL cells but to be induced during lytic reactivation [51] , consistent with our detected patterns of vIRF-1 expression in BCBL1-TRE/RTA cells . Mitochondrial localization of vIRF-1 was verified by IFA in TIME-TRE/vIRF-1 cells [23] in which vIRF-1 expression is inducible upon addition of doxycycline to culture medium . These results demonstrated localization of detectable levels of vIRF-1 to a subset of loci staining positively for mitochondrial marker TOM20 ( Fig . 6D ) . In both Dox-inducible BCBL1-TRE/RTA PEL cells and HHV-8-infected TIME-TRE/RTA endothelial cells ( see Materials and Methods and Fig . S1 ) , vIRF-1 was found by immunoblotting of mitochondrial fractions to localize in part to mitochondria during productive replication ( Fig . 6E ) . The proportion of vIRF-1 localizing to mitochondria in the BCBL-1 cells was comparable with that measured in transfected cells ( Fig . 6A ) ; the level in the TIME cells ( +Dox ) was >4 times higher at 9% . For the PEL cells , which express vIRF-1 also during latency but at reduced levels , the proportion of mitochondrial-localized vIRF-1 was increased from 0 . 9% to 2% in Dox-treated , lytically-induced cultures , possibly reflecting biological significance of vIRF-1 activity at this site during productive replication . It should be noted that as only a subset of these cells support lytic reactivation , the proportion of vIRF-1 localized to mitochondria in lytically infected cells is likely to be substantially greater than the 2% level observed for the culture as a whole . Similar fractionation experiments in HHV-8+ TIME-TRE/RTA cells treated with Dox verified Bid , as well as vIRF-1 , localization to mitochondria during lytic reactivation ( Fig . 6F ) . Confocal immunofluorescence microscopy detected at least some colocalization of vIRF-1 and Bid in these cells ( Fig . S2 ) . Combined , our data provide evidence of mitochondrial association of vIRF-1 , in both transduced and infected cells , and indicate that vIRF-1 may target BOPs for inhibition at this site . Bid and Bim BH3 domain-targeting by vIRF-1 , nuclear localization-independent inhibition of BOP pro-apoptotic activity , and partial mitochondrial localization of vIRF-1 suggested the likelihood of direct BOP inactivation via BBD∶BH3 association . This was tested by using an in vitro mitochondrial permeabilization assay to assess the abilities of wild-type and a ΔBBD ( BOP-refractory ) variant of vIRF-1 to inhibit tBid-induced cytochrome c release . The vIRF-1 proteins and tBid were expressed as T7/intein/CBD and thioredoxin/His6/S-tag fusion proteins in bacteria and were subsequently purified and cleaved to release the respective T7- and S-tagged proteins ( see Materials and Methods ) . SDS-PAGE and Coomassie staining ( Fig . 7A ) verified their purity prior to use . Addition of recombinant tBid ( 1 . 5 µg/ml , 100 nM ) to mitochondrial preparations induced the release of cytochrome c into the soluble fraction of the mitochondrial suspension and led to a corresponding decrease in the level of cytochrome c in the mitochondrial pellet , as determined by Western analysis ( Fig . 7B ) . Inclusion of vIRF-1 ( 8 µg/ml , 100 nM ) blocked all detectable cytochrome c release , but vIRF-1ΔBBD was inactive in respect of tBid inhibition . These data demonstrate that BBD∶BH3 interaction alone is sufficient to inhibit tBid-induced apoptosis . Further experiments were undertaken to determine if BH3-B could substitute functionally for BBD in the context of vIRF-1 to block tBid-induced mitochondrial permeabilization and also to confirm direct inhibitory activities of the BBD and BH3-B domains , expressed as fusions with GST . The recombinant proteins were isolated and purified from bacterial extracts and their purities and concentrations checked prior to experimental use ( Fig . 7C ) . T7-fused wild-type vIRF-1 and its BBD-substituted counterpart each were able to inhibit tBid-induced cytochrome c release from mitochondrial preparations ( Fig . 7D , top ) . Similarly , both GST-BBD and GST-BH3-B were able to inhibit tBid-induced cytochrome c release in this assay ( Fig . 7D , bottom ) . These data confirm the functional equivalence of BBD and BH3-B , in the context of vIRF-1 , and the direct role of these domains and their interaction with Bid BH3 in the inhibition of Bid pro-apoptotic activity . We next investigated whether BBD could target additional BH3 domains , in particular those of other BOPs . The respective BH3 coding sequences were cloned in-frame with the GFP open reading frame in plasmid vector pTYB4 and expressed in and purified from bacteria; BBD was expressed and isolated similarly , as a GST fusion protein ( Materials and Methods ) . BH3 domains tested comprised those of BOPs Bad , Bmf , Bnip3L , Hrk and Noxa , along with Bim , Bid and “BH3-only” beclin [52] , and BH3s from multi-BH domain proteins Bcl-2 and Mcl-1 ( anti-apoptotic ) and Bak , Bax and Bok ( pro-apoptotic ) . Results from these co-precipitation assays identified Bik , Bmf , Hrk and Noxa BH3 domains as additional targets of BBD interaction ( Fig . 8A ) . Interactions between the corresponding full-length proteins and vIRF-1 were tested by co-immunoprecipitation of vIRF-1 with Flag-tagged BOPs from transfected cell lysates; all but Bik were able to co-precipitate vIRF-1 in this experiment ( Fig . 8B ) . Bcl-2 was essentially negative , with barely detectable levels of vIRF-1 in the co-precipitate , as was BOP Puma [not included in the BBD∶BH3 experiment ( Fig . 8A ) ] . Therefore , of the Bcl-2 protein family members tested , BOPs Bim , Bid , Bmf , Hrk and Noxa are demonstrably targeted by vIRF-1 , via BBD∶BH3 association , and Bik BH3 can also bind BBD . Comparisons of the BH3 domains of vIRF-1/BBD-interacting BOPs identified a single unique and conserved residue among the BBD-binding BH3 sequences , namely an alanine at position φ1+1 ( Fig . 9A , left ) . Mutagenesis of this position within the context of Bim BH3 was undertaken to determine its significance with respect to BBD binding; it was changed to each of the collinear residues of the non-binding BH3 domains . Additionally , residues φ1+1 to φ1+3 ( SEC ) of BBD-refractory Bax BH3 were changed to the equivalents ( AQE ) in closely related Bim BH3 and the reciprocal changes were made in Bim BH3 to determine if these “diverged” residues in combination could , respectively , confer and abrogate BBD binding . The various changes made are shown in Fig . 9A ( right ) . As before , these sequences were expressed as GFP fusions for use in GST-BBD-based coprecipitation assays . Other than wild-type Bim BH3 , glutathione bead-precipitated GST-BBD was able to efficiently co-precipitate only cysteine- and serine-substituted alanine φ1+1 , with weak binding apparent for the leucine-substituted BH3 ( Fig . 9B ) . The AQE substitution of Bax SEC residues was able to confer at least some BBD-binding capacity to Bax , demonstrating the contribution of these residues , either directly or via structural influence , to binding; the converse substitution in Bim abrogated binding . Interestingly , a sequence isolated from a phage-display dodecamer-peptide library using GST-BBD as bait had some resemblance to BH3 sequences in respect of conserved basic residues and , importantly , possessed an alanine residue at the equivalent of position φ1+1 . This sequence also showed some binding in the in vitro coprecipitation assay . Taken together , these data indicate the likely central importance of alanine at position φ1+1 for BBD interaction , although the residue's contribution is likely indirect and evidently context dependent , as particular collinear small side-chain residues ( serine and cysteine ) from non-binding BH3 domains can substitute for alanine in Bim BH3 and the AQE motif from Bim can confer only weak binding to the closely related BH3 domain of Bax . Previous studies from this laboratory identified Bim as a potent inhibitor of virus productive replication and the importance of vIRF-1 BBD-mediated interactions in countering lytic cycle-induced apoptosis and promoting HHV-8 production in TIME cells [23] , [8] . In view of the present findings that vIRF-1 inhibits Bid activity in a BBD-dependent fashion ( Fig . 7 ) and that Bid is induced in lytically reactivated TIME cells ( Fig . 4A ) , we wanted to test the significance of Bid in HHV-8 replication . To do this , we utilized lentiviral vector-delivered shRNAs directed to Bid mRNA sequences to deplete Bid in HHV-8+ ( latently infected ) TIME cells and compared levels of cell-released encapsidated viral genomes produced following TPA induction to those obtained from non-silencing ( NS ) shRNA-transduced control cultures . Bim shRNA-transduced TIME cells were also included to provide a positive control for the experiment . The data from this experiment ( Fig . 10A ) revealed that each of the two Bid shRNAs led to small but significant increases in virus production from TPA-treated TIME cells; as before , Bim depletion led to substantial amplification of virus titers . Similar experiments were undertaken in TIME-TRE/RTA cells to confirm the effect on replication of Bid depletion; here , lytic replication was induced by addition of doxycycline to the cultures . Again , Bid depletion led to increased virus production ( Fig . 10B ) , measured in this experiment by titration of released infectious virus in culture media via inoculation of naïve TIME cells and detection of virus infection by immunofluorescence assay for latency-associated nuclear antigen ( LANA ) . These data demonstrate that Bid , in addition to Bim , is a contributor to negative regulation of HHV-8 infection and suggest that its control by vIRF-1 is likely to be important for optimal virus productive replication . It is important to note , however , that the positive effects on virus replication of Bid and Bim depletion by shRNA transduction demonstrate also that vIRF-1 is not completely effective at suppressing the activities of these BOPs . This situation is not unexpected and is probably universal amongst such viral regulators of cellular activities .
Viruses have numerous and diverse mechanisms of apoptotic inhibition , necessitated by the pro-apoptotic signals induced upon de novo infection of a cell and by the processes associated with virus replication ( reviewed in [53]–[55] ) . These mechanisms include inhibition by various means of interferon induction and signaling and inactivation and suppression of p53 , pro-apoptotic proteins such as BOPs , and caspase mediators of apoptotic signaling . Examples are p53-inhibitory activities of simian virus 40 T-antigen and adenovirus E1B-55K [56] , [57] , caspase inhibition by HHV-8 specified K7/vIAP and gammaherpesvirus viral FLICE inhibitory ( vFLIP ) proteins [6] , [7] , [58] , [59] , and the Bcl-2 homologues specified by herpesviruses ( gammaherpesvirus vBcl-2s and human cytomegalovirus UL37x1/vMIA ) , poxviruses ( e . g . , fowlpox virus 039 ) , African swine fever virus ( A179L ) , adenovirus ( E1B-19K ) , and others [60] , [61] . These viral Bcl-2 homologues , while not necessarily readily identifiable at the amino acid sequence level , have been demonstrated or are believed to preserve the essential BH-like helical domains and overall three-dimensional hydrophobic groove structure of cellular Bcl-2 proteins to allow their interactions with and inhibition of pro-apoptotic BH3-only proteins and/or apoptotic executioners Bax and Bak . Thus , Bcl-2 “mimicry” is a commonly used mechanism of viral evasion from innate host cell defenses via apoptotic induction . However , alternative mechanisms of direct BOP inhibition by viral proteins have not previously been reported , to our knowledge . HHV-8 specifies a number of proteins that have predicted abilities to inhibit apoptosis induced by virus de novo infection and lytic replication and therefore have the potential to promote virus infection and establishment of latency and/or productive replication [62] . However , demonstration of such activities in the context of virus infection is largely lacking , although vIRF-1 , in part via BBD-dependent interactions , has been found to promote cell survival under lytic-induced stress and to enhance HHV-8 productive replication in culture [23] . Our previous studies identified Bim as a potent negative regulator of HHV-8 productive replication , mapped the binding domain ( BBD , residues 170–187 ) involved in its inhibition , and noted induced nuclear localization of Bim as a means of its inactivation by vIRF-1 to promote HHV-8 replication and cell survival [23] , [8] . However , the region of Bim interacting with vIRF-1 was not mapped and the possibility of direct inhibition of Bim activity , in addition to inhibition via vIRF-1-induced nuclear sequestration , was not considered . The present identification of BH3 as the target of BBD binding to Bim , as well as Bid and other BOPs , and finding of direct inactivation by vIRF-1 of Bid-induced mitochondrial permeabilization in vitro , Bim and Bid inhibition by nuclear localization-defective vIRF-1 , and inability of vIRF-1 to induce significant Bid ( BidL , tBid ) nuclear localization , demonstrate that vIRF-1 can inhibit BOPs independently of nuclear translocation . Indeed , in contrast to Bim , [23] , nuclear localization of Bid was not apparent in lytically infected cells ( Fig . 4 ) . The mechanism of induced nuclear translocation of Bim could comprise cytoplasmic-to-nuclear chaperoning by vIRF-1 and/or nuclear capture of Bim translocating independently or by other means . The latter would be analogous to HHV-8 latency-associated nuclear antigen ( LANA ) -induced nuclear localization of predominantly cytoplasmic GSK3β [63] , for example . The fact that Bid has been reported to localize in part to the nucleus , and indeed to function here as a component of the DNA repair machinery and as an apoptotic mediator in response to DNA damage [47]–[49] , indicates that simple “nuclear capture” by vIRF-1 is unlikely to be a mechanism of vIRF-1-induced nuclear sequestration of Bim , as this would be expected to operate for Bid also . Specificity of Bim nuclear chaperoning by vIRF-1 is , similarly , difficult to explain , as Bim and Bid interactions with vIRF-1 occur by the same means ( BH3∶BBD binding ) and these BOPs , able to move between cellular compartments , appear to be equally susceptible to translocation . It is possible that while both BOPs can enter the nucleus , independently or promoted by vIRF-1 , only Bim can , via its association with vIRF-1 , form stable interactions with other nuclear proteins to effect its sequestration in this compartment . In trying to resolve this issue , it would be informative to determine whether vIRF-1 can induce nuclear translocation of any of the other BBD-targeted BOPs or if this activity is restricted to Bim . Regardless of mechanism , however , it is apparent that vIRF-1-induced nuclear sequestration of Bim represents a mechanism of inactivation of its pro-apoptotic activity , in addition to its direct inactivation via BH3 binding . It is possible that this is necessary for biologically sufficient inhibition of this powerful negative regulator of HHV-8 productive replication . It is also conceivable that there are as-yet unrecognized nuclear functions of Bim that contribute to HHV-8 lytic replication . An important finding of the present study is that vIRF-1 can interact , via its Bid BH3-B-like BBD region , with the BH3 domains of BOPs Bid , Bik , Bmf , Hrk and Noxa in addition to Bim BH3 , while refractory to interaction with other tested Bcl-2 family members ( other BOPs and multi-BH-domain proteins ) . Thus , the “Bim-binding domain” of vIRF-1 should more appropriately be referred to as the “BOP-binding domain” . These additional interactions of BBD not only identify multiple new BOP interaction partners of vIRF-1 that could be targeted for inhibition , of likely relevance in the context of HHV-8 biology , but also provide the tools to better understand the molecular basis and specificity of BBD and BH3-B interactions with BH3 domains . Previous studies of the latter identified , via mutagenesis of murine Bid BH3-B sequences and analysis of BH3-B∶BH3 interaction and Bid activity , residues L35 ( first hydrophobic , BH3/BBD-conserved ) and G39 , R40 and/or E41 ( GHE in human Bid ) as important for BH3-B∶BH3 binding in vitro and for both physical and functional interactions , respectively [40] . Our own in silico , mutagenesis , and binding studies indicate that for BBD∶BH3 interactions , an alanine residue corresponding to BH3 position “φ1+1” ( Fig . 9A ) is both conserved and specific to all BBD-interacting BOP BH3 domains and important for BBD∶BH3 interaction . An alanine at this position was also identified in BH3-like sequences [RVADSLATLMMN ( Fig . 9B ) , SIANTIASVQFM , and IFAALDYNLGRH; φ1+1 underlined , collinear hydrophobic residues italicized ) isolated from a phage-display screen using BBD as bait . However , because of its simple methyl side chain , it is likely that this alanine residue permits adoption of the required local structure and “space” for binding via other BBD/BH3-residue interactions rather than being involved directly in binding . Thus , while binding was abrogated by mutation of this alanine to most other collinear residues of non-binding BH3s , mutation to cysteine or serine ( present in BH3s of Bok and Bax , respectively ) was compatible with binding to BBD . Also , introduction of this alanine along with adjacent glutamine and glutamate residues ( as present in Bim ) into BBD-refractory Bax BH3 was sufficient to confer BBD interaction , although the binding was weaker than that of Bid BH3 . Taken together , our data indicate that the φ1+1 alanine appears to be preferred and important for BH3 interaction , but that it is unlikely to contribute directly to binding and that other residues , in appropriate context , are directly involved in BH3 association and specificity . In regard to specificity of BH3 binding , we have established definitively that while BBD can recognize Bim and Bid BH3 domains , the former is not targeted by Bid BH3-B . Alignments of BBD and BH3-B , although revealing significant similarities , with three identical and three highly-related residues within the BH3-binding core regions , also show considerable divergence within and outside these central sequences ( Fig . 1 ) . It is noteworthy that previous di-alanine scanning mutagenesis of the 174–183 region of BBD identified residues within 174–181 as essential for interaction with Bim-BH3 [23] . Interestingly , however , combined mutation of the first two conserved hydrophobic residues , L174 ( qe ) I177 , did not abrogate binding ( unpublished data ) , although identical or related hydrophobic residues are conserved in BH3 domains and are key contributors to the amphipathicity of the predicted α-helix . Single-residue mutagenesis across BBD , to alanine or to collinear BH3-B residues would be warranted to further delineate those amino acids and associated properties contributing to interaction with BH3 domains and to Bim- and Bid-BH3 selectivity . In summary , data presented here identify similarity between vIRF-1 BBD and Bid BH3-B , corresponding inhibitory interactions of vIRF-1 BBD with the BH3-domains of Bim and Bid ( both induced during and inhibitory to HHV-8 productive replication ) , and additional BBD binding of the BH3 domains of BOPs Bik , Bmf , Hrk and Noxa . To our knowledge , this is the first example of BOP targeting and inhibition via Bid BH3-B domain mimicry and thereby our data reveal a novel mechanism of viral evasion from host cell , apoptosis-mediated defense against viral infection and replication .
Telomerase-immortalized endothelial ( TIME ) cells [46] and genetically engineered derivatives were cultured in EGM-2 MV medium ( Lonza; Walkersville , MD ) containing 5% fetal bovine serum ( FBS ) and cytokine supplements . TIME cell lines expressing vIRF-1 or RTA in doxycycline ( Dox ) -inducible fashion were generated using the Retro-X Tet-On Advanced system ( Clontech Laboratories; Mountainview , CA ) . Briefly , the pRetroX-Tet-On Advanced plasmid was transfected into Phoenix cells and the supernatant was used to transduce TIME cells which were selected in G418 ( 400 µg/ml ) to obtain TIME/Tet-On cells . The RTA coding region was derived from an existing eukaryotic expression vector as an EcoR1 restriction fragment and ligated into the EcoR1 site in the pRetroX-Tight-Pur plasmid . This was transfected into Phoenix cells , virus-containing supernatant used to infect TIME/Tet-On cultures , and transduced cells ( TIME-TRE/RTA ) selected in puromycin ( 1 µg/ml ) . Cloning discs were used to isolate individual colonies , and derived cells were screened by immunofluorescence assay for RTA expression following Dox induction . HeLa , HEK293 , and HEK293T cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% FBS and gentamicin . BCBL-1/TRE-RTA [50] cells were maintained in RPMI 1640 medium containing 15% FBS and gentamicin . HHV-8 virus stocks were derived from doxycycline induced BCBL-1/TRE-RTA cultures and used to infect TIME cells as described previously [23] . For cell viability or immunofluorescence assays , cells were transfected using Fugene 6 ( Roche Applied Science; Indianapolis , IN ) . For immunoprecipitation , cells were transfected using standard calcium-phosphate method or Lipofectamine 2000 ( Invitrogen; Carlesbad , CA ) . For reporter assays , HEK293 cells were transiently transfected with plasmids expressing vIRF-1 and p53 along with the PG13-luc reporter vector ( Addgene; Cambridge , MA ) for 24 h and then lysed with passive lysis buffer ( Promega; Madison , WI ) . Luciferase activity was measured by standard methods using D-luciferin and luminometry . For lentivirus production , HEK293T cells were transfected with virus vector and gag/pol-encoding plasmids using standard calcium-phosphate precipitation and virus was harvested after 48 h by centrifugation at 49 , 000×g . Cells were transduced with lentivirus in the presence of 5 µg/ml polybrene for 12 h and then cultured in complete media . For bacterial expression of T7-tagged proteins , coding sequences of T7 were first cloned between the NcoI and SalI sites of pTYB4 ( New England Biolabs; Ipswich , MA ) ; coding sequences of vIRF-1 or enhanced green fluorescent protein ( EGFP ) were then inserted between the SalI and SmaI sites of pTYB4-T7 . The EGFP cDNA was amplified from vector pEGFP N1 ( Clontech Laboratories ) . For bacterial expression of EGFP-fused BH3 peptides , EGFP coding sequences were inserted between the SalI and EcoRI sites of pTYB4 and BH3 sequences ( ds-oligonucleotides ) were then inserted between the NheI and SalI sites of pTYB4-EGFP . For the bacterial expression of Discosoma red fluorescent protein ( DsRed ) -fused peptides , coding sequences for DsRed were amplified from vector pDsRed2 ( Clontech Laboratories ) and inserted between the SalI and EcoRI sites of pTYB4-T7 . Coding sequences of vIRF-1 Bim-binding domain ( BBD ) or Bid BH3-B domain were inserted between the EcoRI and SmaI sites of pTYB4-T7-DsRed . The BBD sequence was also inserted between the BamHI and EcoRI sites of pGEX4T-1 ( GE Healthcare Life Sciences; Piscataway , NJ ) for expression of GST-BBD . For generation of recombinant S-tagged tBid ( see below ) , the coding sequence of tBid ( comprising codons 61–195 of BidL [64] ) was inserted between the BamHI and EcoRI sites of pET-32a ( + ) ( Novagen; Madison , WI ) . BOP and Bcl-2 cDNA sequences linked to Flag were cloned between the BamHI and EcoRI site of pcDNA3 . 1 ( Invitrogen ) for expression in transfected cells . Plasmids expressing vIRF-1 and Bim were described previously [23] . Mutagenesis was performed by a PCR-mediated method using Pfx DNA polymerase ( Invitrogen ) with oligonucleotide primers containing deletion or substitution mutations . Two short hairpin RNAs ( shRNA ) for Bid were cloned into pYNC352/puro ( a derivative of pTYB6 [65] , [66] ) using BamHI and MluI enzyme sites . The target sequences of the shRNAs correspond to 5′- GGGATGAGTGCATCACAAACC -3′ ( sh1 ) and 5′-CTTTCACACAACAGTGAATTT-3′ ( sh2 ) . Commercially obtained antibodies used in this study were as follow: T7 , Novagen ( Madison , WI ) , catalog number 65922; GFP , Bcl-2 and cytochrome c , Epitomics ( Burlingame , CA ) , catalog numbers 1533-1 , 1017-1 and 1896-1; Flag M2 and β-actin , Sigma ( St . Louis , MI ) , catalog numbers F3165 and A5441; polyclonal Flag and Bim antibodies , Cell Signaling Technology ( Beverly , MA ) , catalog numbers 2368 and 2819; Bax ( N-20 , catalog number sc-493 ) , Bid ( 5C9 , sc-56025 ) , GST ( B-14 , sc-138 ) , prohibitin ( H-80 , sc-28259 ) , TOM20 ( F-10 , sc-17764 ) , VDAC1 ( 20B12 , sc-58649 ) , lactate dehydrogenase ( H-160 , sc-33781 ) , and histone deacetylase 1 ( H-11 , sc-8410 ) antibodies were purchased from Santa Cruz Biotechnologies ( Santa Cruz , CA ) ; LANA ( LN53 ) , Advanced Biotechnologies Inc . ( Columbia , MD ) , catalog number 13-21-100 . vIRF-1 rabbit antiserum was provided by Dr . Gary Hayward . For immunoblotting , cells were lysed in lysis buffer ( 50 mM Tris-HCl [pH 8 . 0] , 150 mM NaCl , 1 mM EDTA , 1% IGEPAL CA-630 , and 0 . 25% sodium deoxycholate ) freshly supplemented with protease inhibitor cocktail ( Sigma ) for 1 h on ice . After centrifugation at 12 , 000×g for 20 min , the supernatant was used as a whole cell extract . For immunoblotting , proteins were size fractionated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-PAGE ) and transferred to a nitrocellulose or polyvinylidene fluoride membranes . Immunoreactive bands were detected with enhanced chemiluminescence solution ( GE Healthcare Life Sciences ) and visualized on X-ray film or digitally using a chemiluminescence imager . For immunofluorescence assays , cells were grown on a 0 . 1% gelatin-coated coverglass or a chamber slide and were fixed and permeabilized in chilled methanol . Following incubation with Superblock blocking buffer in phosphate-buffered saline ( PBS ) ( Thermo Scientific Inc . ; Rockford , IL ) , coverslips were incubated with primary antibody , washed with PBS , and then incubated with appropriate fluorescent dye-conjugated secondary antibody . The coverslips were mounted in 90% glycerol in PBS containing 10 mg/ml p-phenylenediamine , an antifade reagent . For immunoprecipitation , HEK293T cells transfected with plasmids encoding vIRF-1 or Flag epitope-tagged BH3-only proteins ( BOPs ) were lysed in lysis buffer , and cell extracts were incubated with anti-Flag M2 affinity gel ( Sigma ) for 3 h at 4°C . After washing with lysis buffer , immune-complexes were eluted with 30 µl of 3× Flag peptide ( 150 ng/µl ) , subjected to SDS-PAGE , and analyzed by immunoblotting using vIRF-1 antiserum or polyclonal Flag antibody . For in vitro binding assays , fluorescent protein-fused peptides or T7-tagged proteins expressed in E . coli from vector pTYB4 ( New England Biolabs ) were purified according to the manufacturer's protocol . Bacterially expressed glutathione-S-transferase ( GST ) -fused proteins were purified by standard methods . 0 . 5 µg of EGFP or EGFP-BH3 , fused to intein-chitin binding domain ( CBD ) and immobilized on chitin beads , was mixed with 1 µg of purified T7-DsRed-fused BBD or BH3-B peptides and incubated for 1 h at room temperature . After washing four times with Tris-buffered saline ( TBS ) supplemented with 0 . 1% Tween 20 , bead-associated proteins were size-fractionated by SDS-PAGE and analyzed by immunoblotting using T7- or GFP-specific antibodies . To screen for BBD interaction with a variety of BOP BH3 domains , 1 µg of GST or GST-BBD protein immobilized on glutathione sepharose 4B beads was mixed with 2 µg of EGFP-BH3 fusion proteins . GST-BBD and its associated proteins were eluted with 10 mM reduced glutathione in TBS , size-fractionated by SDS-PAGE , and analyzed by immunoblotting using GFP- or GST-specific antibodies . For nucleo-cytoplasmic fractionation , cells were homogenized in buffer A ( 20 mM Tricine-KOH [pH 7 . 8] , 5 mM MgCl2 , 25 mM KCl , 0 . 25 M sucrose , and protease inhibitor cocktail ) using a Dounce homogenizer . After centrifugation at 1 , 000×g for 10 min , the supernatant was used as the cytoplasmic fraction , and the pellet was subjected to 25–35% iodixanol discontinuous gradient centrifugation . Nuclei were collected from the interface between 30 and 35% iodixanol and resuspended in buffer C ( 20 mM HEPES [pH 8 . 0] , 1 . 5 mM MgCl2 , 420 mM NaCl , 0 . 2 mM EDTA , and protease inhibitor cocktail ) . For in vitro cytochrome c release assays , mitochondria were isolated by sucrose density gradient centrifugation . Briefly , HEK293T cells grown to subconfluence in a 10 cm dish were washed in ice-cold PBS and resuspended in mitochondrial isolation buffer ( MIB: 210 mM mannitol , 70 mM sucrose , 1 mM EDTA , and 10 mM HEPES [pH 7 . 5] ) supplemented with protease inhibitor cocktail . Next , cells were homogenized with 40 strokes of Dounce homogenizer and centrifuged at 2 , 000×g for 10 min . The supernatant was further centrifuged at 13 , 000×g at 4°C for 10 min . After resuspending in MIB , the resulting pellet was layered on top of a discontinuous sucrose gradient consisting of 1 . 2 M sucrose in buffer H ( 10 mM HEPES [pH 7 . 5] , 1 mM EDTA , and 0 . 1% BSA ) on top of 1 . 6 M sucrose in buffer H . Following centrifugation at 131 , 000×g for 2 h at 4°C , mitochondria were recovered at the 1 . 6–1 . 2 M sucrose interface , washed in MIB , centrifuged at 13 , 000×g at 4°C for 10 min , and resuspended in 100 µl of MIB . For other studies , mitochondria were purified using Axis-Shield OptiPrep ( Sigma ) according to the manufacturer's protocol . For proteinase K treatment , mitochondria in MIB ( 0 . 5 mg/ml ) were incubated on ice for 30 min with 20 µg/ml proteinase K; the reaction was stopped by the addition of 2 mM PMSF , and mitochondria were then washed twice in 1 ml of MIB and resuspended in MIB or MRM buffer ( 250 mM sucrose , 10 mM HEPES [pH 7 . 5] , 1 mM ATP , 5 mM sodium succinate , 80 µM ADP , 2 mM K2HPO4 ) . For mitochondrial binding assays , 1 µg of recombinant T7-EGFP or T7-vIRF-1 protein was added to the proteinase K-pretreated or untreated mitochondria in MRM buffer ( 50 µg protein/50 µl ) supplemented with PMSF and 0 . 1 mg/ml BSA . The mixtures were incubated for 1 h at 30°C and then centrifuged at 12 , 000×g for 5 min . The mitochondrial pellets were washed twice in MRM buffer , and the final washed pellets were resuspended in SDS sample buffer . Trichloroacetic acid ( TCA , 10% final concentration ) was added to the supernatants to precipitate proteins prior to SDS-PAGE and immunoblot analysis . For in vivo cytochrome c release assay , transfected HEK293T cells were subjected to Dounce homogenization ( 30 strokes ) in MIB buffer . The homogenate , an aliquot of which was used as a total cell extract , was centrifuged at 1000×g for 10 min at 4°C to remove nuclei and unbroken cells . The resulting supernatant was centrifuged at 100 , 000×g for 1 h at 4°C to yield the final soluble cytosolic fraction ( S100 ) . In vitro mitochondrial permeabilization assays based on cytochrome c release were undertaken essentially as described by Arnoult [67] and outlined below . Thioredoxin/His6/S-tagged tBid protein encoded from pET-32a ( + ) was purified using Ni-NTA His-tag affinity chromatography , and S-tBid ( 100 µg/ml ) was eluted following thrombin ( 0 . 5 unit/ml ) treatment for 2 h at room temperature . Thioredoxin and His sequences were retained on the Ni-NTA resin . S-tBid was purified away from thrombin using protein S agarose , eluted with 3 M MgCl2 , and dialyzed against dilution buffer ( 25 mM HEPES-KOH [pH 7 . 4] , 0 . 1 M KCl ) . Following preincubation of 10 nM or 100 nM of S-tBid with or without 500 nM of GST-fusion peptides or 100 nM of T7-vIRF-1 proteins ( wild-type or mutated ) in dilution buffer supplemented with 1 mg/ml of fatty acid-free bovine serum albumin ( FA-BSA ) , purified mitochondria were added ( 50 µg protein/50 µl of mitochondrial buffer: 125 mM KCl , 0 . 5 mM MgCl2 , 3 mM succinic acid , 3 mM glutamic acid , 10 mM HEPES-KOH [pH 7 . 4] , 1 mg/ml FA-BSA , and protease inhibitor cocktail ) . The reaction mixtures were incubated at 30°C for 30 min and then centrifuged at 12 , 000×g for 5 min at 4°C to pellet the mitochondria . The supernatants were quickly removed , and the pellet was resuspended in 70 µl of mitochondria lysis buffer ( 50 mM Tris-HCl [pH 7 . 4] , 150 mM NaCl , 2 mM EDTA , 2 mM EGTA , 0 . 2% Triton X-100 , and 0 . 3% IGEPAL CA-630 ) . The amounts of cytochrome c in the supernatant and pellet fractions were determined by immunoblotting . After plasmid transfection of 293T cells on a coverslip , the cells were stained with Cy3-conjugated annexin V ( Biovision Inc . ; Mountain View , CA ) in AV binding buffer ( 10 mM HEPES [pH 7 . 4] , 140 mM NaCl , and 2 . 5 mM CaCl2 ) , fixed with 2% formaldehyde in AV binding buffer for 10 min , and mounted in glycerol medium containing DAPI . Annexin V-positive cells , fluorescent under UV , were counted from three randomly selected low-magnification microscopic fields . The Ph . D . -12 library , comprising a complexity in excess of one billion independent clones , was purchased from New England Biolabs ( Beverly , MA ) . Screening was performed by a solution-phase panning method with affinity bead capture as described by the manufacturer's protocol . In brief , a mixture of purified GST-BBD ( 500 nM ) and the library ( 2×1011 pfu ) was incubated for 20 min at room temperature and added to glutathione 4B beads pre-blocked with BSA . After incubating for 15 min and washing the mixture with Tris-buffered saline ( TBS ) containing 0 . 1% Tween 20 , phage were eluted and amplified . Negative selection using purified GST was performed prior to the second round of panning . After the third round of panning , the mixture was washed with TBS containing 0 . 5% Tween 20 , and phage were eluted and plaque-purified prior to DNA preparation and sequencing . For HHV-8 infection , TIME cells were centrifuged at 1 , 000×g for 1 h in the presence of HHV-8 virions and then cultured in fresh complete medium for 7 days to allow establishment of latency in the absence of ongoing lytic replication . After lentiviral transduction of control ( NS ) , Bid , or Bim shRNAs for 2 days into HHV-8+ TIME cells , lytic replication of HHV-8 was induced by treatment with TPA ( 20 ng/ml ) or 1 µg/ml doxycycline ( TIME-TRE/RTA cells ) . For determination of encapsidated HHV-8 genome copy number , viral DNA was isolated using standard phenol extraction and glycogen/ethanol precipitation methods following pre-treatment of virus suspensions with DNaseI for 20 min at 37°C to remove any unencapsidated DNA . For the determination of the viral genome copy number , all qPCRs were performed in a 96-well microplate using an ABI Prism 7500 detection system ( Applied Biosystems; Foster City , CA ) with SYBR green/ROX master mix ( SuperArray Bioscience Corp . ; Frederick , MD ) . For induced TIME-TRE/RTA cells , infectious virus titers were measured by application of induced culture media-derived virus to naïve TIME cells and immunofluorescence staining for HHV-8 latency-associated nuclear antigen ( LANA ) .
|
Viruses possess mechanisms of subverting host cell defenses against infection and virus replication; these mechanisms are essential to the virus life cycle . Here , we identify and characterize a novel mechanism of HHV-8 mediated inhibition of virus-induced programmed cell death ( apoptosis ) . This function is specified by viral interferon regulator factor homologue vIRF-1 , which binds to and directly inhibits pro-death activities of so-called BH3-only proteins ( BOPs ) , induced and activated by stress signals such as those occurring in infected cells . The BH3 domains of BOPs mediate their pro-apoptotic functions , and it is these domains that are targeted by vIRF-1 , via a region resembling a BH3-interacting and -inhibitory domain , termed BH3-B , present in one of the vIRF-1 targeted BOPs , Bid . The targeted BOP BH3 domains share characteristic and conserved features . As shown previously for Bim , depletion of Bid leads to enhanced HHV-8 productive replication , demonstrating that Bid , also , is a biologically significant negative regulator of virus replication and suggesting that its control by vIRF-1 is of functional importance . To our knowledge , this is the first report of viral targeting and inhibition of BOP activity via Bid BH3-B mimicry; our studies therefore expand the known mechanisms of viral evasion from antiviral defenses of the host .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"death",
"cellular",
"stress",
"responses",
"molecular",
"cell",
"biology",
"cell",
"biology",
"virology",
"biology",
"microbiology",
"host-pathogen",
"interaction"
] |
2012
|
Human Herpesvirus 8 Interferon Regulatory Factor-Mediated BH3-Only Protein Inhibition via Bid BH3-B Mimicry
|
Mobile genetic elements either encode their own mobilization machineries or hijack them from other mobile elements . Multiple classes of mobile elements often coexist within genomes and it is unclear whether they have the capacity to functionally interact and even collaborate . We investigate the possibility that molecular machineries of disparate mobile elements may functionally interact , using the example of a retrotransposon , in the form of a mobile group II intron , found on a conjugative plasmid pRS01 in Lactococcus lactis . This intron resides within the pRS01 ltrB gene encoding relaxase , the enzyme required for nicking the transfer origin ( oriT ) for conjugal transmission of the plasmid into a recipient cell . Here , we show that relaxase stimulates both the frequency and diversity of retrotransposition events using a retromobility indicator gene ( RIG ) , and by developing a high-throughput genomic retrotransposition detection system called RIG-Seq . We demonstrate that LtrB relaxase not only nicks ssDNA of its cognate oriT in a sequence- and strand-specific manner , but also possesses weak off-target activity . Together , the data support a model in which the two different mobile elements , one using an RNA-based mechanism , the other using DNA-based transfer , do functionally interact . Intron splicing facilitates relaxase expression required for conjugation , whereas relaxase introduces spurious nicks in recipient DNA that stimulate both the frequency of intron mobility and the density of events . We hypothesize that this functional interaction between the mobile elements would promote horizontal conjugal gene transfer while stimulating intron dissemination in the donor and recipient cells .
Mobile group II introns are remarkable retroelements , based on large catalytic RNAs that encode reverse transcriptase ( RT ) that is required for their movement [1] , [2] . Despite their bacterial origin , they are of interest for their putative ancestral relationship to nuclear spliceosomal introns and their ability to invade DNA and spread by a mechanism similar to non-LTR retrotransposons in metazoans [3]–[6] . Besides being at the nexus of eukaryotic evolution , group II introns often co-exist with other mobile elements in bacteria [7] . The molecular underpinning of one such liason is explored below . Self-splicing group II introns can efficiently insert into intronless alleles by a retrohoming process , or integrate at lower frequencies into ectopic sites by retrotransposition [2] , [8] , [9] . Both retrohoming and retrotransposition occur by reverse splicing into DNA , where the intron RNA is copied into cDNA by the RT activity of the intron-encoded protein ( IEP ) . In the high-efficiency retrohoming reaction , integration is into dsDNA and the primer for cDNA synthesis is provided by a nick introduced by the IEP's endonuclease activity [8] , [10] ( Fig . 1A ) . In low-frequency retrotransposition to ectopic sites , integration is predominantly into ssDNA and primers can be provided by Okazaki fragments at replication forks [11] , [12] ( Fig . 1B ) , leading to broad intron dispersal . Curiously , many group II introns reside on bacterial mobile elements including plasmids . Examples are provided by conjugative plasmid pRS01 in Lactococcus , virulence plasmid pX01 in Bacillus , and Pseudomonas-derived toluene-catabolic self-transmissible plasmid pDK1 [13] , [14] . The pRS01 plasmid , a broad-host-range conjugative element , mediates high-frequency transfer of the genes encoding lactose utilization among lactococci [15] . In bacteria , conjugation is one of the most common ways to increase genome plasticity via horizontal gene transfer . The mobile conjugative plasmids carry , among others , the genes encoding proteins necessary for conjugative DNA transfer , including a relaxase enzyme which is necessary for the initiation of conjugation . The relaxase nicks the plasmid DNA at the origin of transfer ( oriT ) and ssDNA is transferred to the recipient in a site- and strand-specific manner during conjugation [16]–[18] . This reversible reaction creates a covalent linkage between the active-site tyrosyl hydroxyl oxygen and ssDNA 5′-phosphate , yielding a 3′- OH , which can later act as a nucleophile to reverse the transesterification and release the relaxase . Interestingly , the Ll . LtrB group II intron resides within the relaxase ltrB gene of pRS01 plasmid ( Fig . 2A ) , where intron splicing is required for expression of the functional LtrB relaxase protein , which initiates conjugation [13] . Moreover , intron retrotransposition is stimulated in the recipient cell by conjugation [19] , suggesting either that the transfer process or factors encoded by the conjugative plasmid aid intron mobility . Thus , conjugative plasmids not only transfer the intron between bacterial cells and even across genera and species [19] , [20] , but aid dissimination of the intron in the recipient . In the present study we report on the discovery that the conjugative relaxase , LtrB , promotes Ll . LtrB group II intron retrotransposition . To provide a mechanistic basis for enhanced retrotransposition , biochemical experiments were performed with the purified relaxase and the distribution of the retrotransposition events reported by a retromobility indicator gene ( RIG ) was explored . To this end we developed a targeted sequencing approach specifically for retrotransposition insertion sites , called RIG-Seq . Together , the data show the dynamic interaction between two different mobile elements: Intron splicing is essential for active relaxase expression required for conjugation; in turn , relaxase nicks recipient DNA to provide 3′ ends that prime and thereby stimulate retrotransposition to novel sites . Moreover , given that conjugation facilitates intron retrotransposition in both donor and recipient cells , one can infer that this co-habitation of different mobile elements promotes gene transfer and genome plasticity .
To study the effect of the conjugative plasmid pRS01 on mobility of the Ll . LtrB group II intron , we used a kanR RIG cassette where kanamycin resistance is acquired after splicing of a group I intron from an RNA intermediate [11] ( Fig . 2B ) . Whereas retrotransposition frequencies in the control donor strain were ∼1×10−5/recipient ( strain D-C ) , the presence of the conjugative plasmid pRS01 stimulated retrotransposition levels to 4 . 7×10−5 on average ( median 3 . 6×10−5; strain D-pRS01 ) ( Fig . 2C; for strains see Table 1 ) . To test if the relaxase itself was the stimulatory factor , a deletion of the relaxase gene from pRS01 was tested ( strain D-pRS01ΔltrB ) . The retrotransposition frequency dropped to ∼1×10−5 , the same as that of the D-C control . Additionally , with relaxase expression from pLtrB::Ll . LtrB ( strain D-pLtrB ) , retrotransposition increased to 3 . 2×10−5 on average ( median 3 . 7×10−5; Fig . 2C ) . This elevation in retrotransposition frequency is not due to effects of donor plasmid pLNRK-RIG copy number , which is slightly depressed in cells carrying pRS01 and pLtrB ( Figure S2A in Text S1 ) , resulting in an underestimate of stimulatory effects by as much as a factor of 2 . Mutation of the putative catalytic tyrosine ( Y21 in LtrB relaxase ) to alanine ( strain D-pLtrB ( Y21A ) ) resulted in reduction in retrotransposition to control levels ( Fig . 2C; Figure S1 in Text S1 ) . Mutants in non-catalytic active site residues had a lesser effect on retrotransposition ( Figure S1 in Text S1 ) . To further investigate the effect of relaxase on retrotransposition , we used several bisphosphonates , etidronic acid ( ETIDRO ) , imino-bis ( methylphosphonic acid ) ( PCNCP ) , and methylenediphosphonic acid ( PCP ) , which are specific inhibitors of relaxase [21] . Elevated retrotransposition levels in D-pRS01 and D-pLtrB strains were reduced 2- to 3-fold by all three relaxase inhibitors ( Fig . 2D ) . We next asked whether the presence of relaxase affects the specificity and distribution of the intron retrotransposition events . We developed a strategy for targeted high-throughput detection of retrotransposed introns , called RIG-Seq , based on the Ll . LtrB RIG construct ( Fig . 3A ) . The unique splice junction ( SJ ) sequence in the kanR gene , from which the group I intron is removed during retromobility , is present in all retrotransposition events but not in the original intron donor ( Fig . 2B ) . This sequence can be used as the landmark for primer binding to detect exclusively retrotransposition events . Libraries for sequencing were constructed after selection of KanR retrotransposition events for the control strain D-C and the two relaxase-containing strains D-pRS01 and D-pLtrB . The sequences containing the 3′ flanking genomic sequence were analyzed for each library ( Tables S1–S4 in Text S1 , where Table S1 represents the summary of multiple sequence alignments for three strains with sequence listings in Tables S2–S4 in Text S1 ) . Initial mapping of the sequenced fragments showed that many of the intron insertions were in the RIG donor plasmid rather than the chromosome ( Fig . 3B and Figure S2 in Text S1 ) . This result is consistent with our previous report that retrotransposition of Ll . LtrB intron occurs frequently into the donor plasmid , likely because of replication-coupled intron integration into ssDNA and the high number of forks per unit of plasmid DNA [22] . Strikingly , whereas only 5 . 2% of reads were mapped to the chromosome for the control D-C strain , 30 . 4% of the reads were chromosomal in the presence of pRS01 , and 86 . 9% of the reads were chromosomal when the D-pLtrB strain expressed relaxase alone , suggesting the ability of relaxase to facilitate intron retromobility into the chromosome ( Fig . 3B ) . We further established that there are no notable differences in the consensus sequences for the insertion sites among relaxase-positive and -negative libraries , indicating universal underlying mechanisms for retrotransposition in L . lactis under these different circumstances ( Fig . 3C and Table S1 in Text S1 ) . A distinguishing feature between invasion into dsDNA and ssDNA pathways is a T residue at position +5 relative to the insertion point ( Fig . 1A and Fig . 1B ) . T+5 is required for IEP endonuclease-mediated nicking of the bottom strand in the dsDNA pathway . We therefore performed comparative analysis of the flanking sequences upstream and downstream of the intron insertions [23] , [24] . Multiple sequence alignment profiles for each library supported the predominantly ssDNA nature of the insertions because the frequency of the T+5 appeared to be low ( ∼25% ) ( Fig . 1 , Fig . 3C , and Table S1 in Text S1 ) . In contrast , cytosine at position −6 seems to be very important for retrotransposition in L . lactis ( C-6 scored ∼90% across libraries ) irrespective of the presence of relaxase . The chromosomal reads were further mapped to the L . lactis IL1403 reference genome [25] . The total number of the unique chromosomal insertion points ( the DNA loci invaded by the intron ) in the libraries correlated with the presence or absence of relaxase ( Fig . 4A ) . The D-C strain produced only 154 unique intron insertion points , whereas there were 461 and 912 unique insertion sites for D-pRS01 and D-pLtrB , respectively . We also observed striking differences in the integration frequency distribution among libraries . Whereas , the relative retrotransposition frequency plotted throughout the bacterial chromosome for the D-C strain exhibited a few pronounced peaks , the two relaxase-expressing strains , D-pRS01 and D-pLtrB , showed more uniform frequencies among a larger number of reads . The distribution of the strong peacks in the D-C strain likely reflects replication fork dynamics since the majority of these peacks are located either in close proximity to the origin of replication ( oriC ) or chromosomal replication termination region , Ter . Interestingly , the highest frequencies from all the libraries were observed ∼60 Kbp upstream of the putative Ter , which is at position ∼1 , 260 , 000 in L . lactis IL1403 genome [25] ( Fig . 4A ) . It is not clear if there is any advantage for the intron to disrupt the locus upstream to the Ter region at a high frequency and at multiple insertion points in L . lactis or if the silencing of the region improves the fitness of the bacterial host under the conditions of the experiment . Alternatively , clustering at Ter may be related to intron-encoded LtrA protein showing distinctive bipolar localization when expressed in E . coli [26] , [27] . Although relaxase does not affect intron insertion specificity , more potential sites become available for intron retrotransposition when relaxase is present . We hypothesize that relaxase stimulation is related to primer availability for reverse transcription . Off-target , spurious nicking activity of the relaxase would satisfy the primer requirement by providing the 3′-OH for initiating reverse transcription ( Fig . 1C ) . Although during conjugation , relaxases nick plasmid DNAs in a sequence- and strand-specific manner at oriT [16] , there are a few examples of conjugative relaxases with loose specificity for oriT sequences such that these enzymes can process cryptic origins in bacterial chromosomes [28] . Similar spurious nicking at off-target sites may be the basis for enhanced retrotransposition . We tested specificity by the ability of pRS01 relaxase to cleave ssDNA of its cognate oriT site and an unrelated oriT site . The LtrB protein was expressed in E . coli and its activity was analyzed ( Fig . 5; Figures S3–S4 in Text S1 ) . The oriT region of pRS01 was predicted based on sequence similarity with other conjugative plasmids [29] and cloned into pGEM from which ssDNA was prepared . The oriT fragment from heterologous plasmid R388 served as a control ( Fig . 5 ) . The ssDNA was treated with either wild-type relaxase or the Y21A catalytic mutant relaxase and the reaction mixtures were used for primer extension analysis . A single cleavage band was observed only for ssDNA containing oriT from pRS01 treated with relaxase but not with the catalytic Y21A relaxase mutant or when ssDNA carried oriT from a heterologous R388 plasmid ( Fig . 5A and 5B ) . To test if LtrB is able to nick non-cognate chromosomal sites , we selected a 1323-bp genomic fragment containing the glutamine ABC transporter permease gene glnP ( Fig . 6 ) . This fragment was chosen because of the high density of intron integration events in this region in presence of relaxase . If relaxase is providing the 3′-OH for initiating reverse transcription , the sites for off-target cleavage might be detectable in the presence of relaxase in such regions for which many retrotransposition events occur . Six retrotransposition events with high relative frequencies were mapped in glnP and all six were present only in libraries with active LtrB relaxase ( D-pRS01 and D-pLtrB ) . A number of faint primer extension cleavage bands was detected for the glnP gene fragment with wild-type relaxase but not with the Y21A relaxase mutant , demonstrating that active relaxase can cleave off-target sites with low efficiency ( Fig . 6 and Figure S4 in Text S1 ) . Interestingly , the most efficient nick site in glnP has 9 of 15 nucleotides identical to the oriT site ( Fig . 6B ) . Thus , the LtrB relaxase not only nicks its cognate oriT with high efficiency , but it is also able to cleave chromosomal off-target sites with low efficiency in vitro .
In this study , we show that two such cohabiting elements that move by different DNA- or RNA-based pathways interact to promote gene transfer ( Fig . 7 ) . The intron splicing generates LtrB relaxase mRNA and thereby facilitates relaxase expression that initiates conjugation , whereas relaxase promotes retrotransposition , in terms of both the frequency and diversity of events . This stimulation is achieved by the relaxase introducing spurious nicks in recipient DNA , thereby providing 3′-OH ends that prime reverse transcription for retrotransposition . Thus , the promiscuous activity of this nickase is the driver that links conjugal mating and retrotransposition , which is elevated in both donor cells ( this work ) and in the recipient [19] , [20] . In our model for relaxase-mediated stimulation of retrotransposition , we propose that when there is sufficient sequence similarity to the homing site to allow intron RNA and chromosomal DNA interaction upstream of the nick site , the intron will reverse splice into the degenerate homing site , while the 3′-OH of the nicked bottom strand provides a primer for reverse transcription ( Fig . 1C ) . cDNA synthesis and repair will then proceed using standard retromobility reactions ( cf . Figs . 1C with 1A and 1B ) . How the relaxase might be removed from the nick site is not known . Although reversal of the relaxase reaction might be invoked in ligating the nick to complete retrotransposition and remove itself from the DNA , our unpublished experiments suggest that this is not generally the case . Transposition stimulated by DNA damage has been shown for a number of bacterial transposons [30] , [31]; some prophages are also induced following the DNA damage [32] . Nicks introduced by relaxase , if they remain unligated , could be converted into DNA double-strand breaks during a replication fork collapse [33] , [34] . It will therefore be interesting to determine if relaxases stimulate other transposons that utilize various DNA intermediates such as DSBs or 3′ ends to mobilize , e . g . Tn7 transposon in E . coli [30] . Transcription and translation of the ltrB gene is growth phase-dependent [35] suggesting that retrotransposition may be coupled to cellular physiology . Transcription of the relaxase operon ( Fig . 2A ) and the ltrB gene in particular is essential for Ll . LtrB retrotransposition since the intron RNA is the template for the synthesis of the first cDNA strand in chromosomal intron integration ( Fig . 1 ) . At the same time , efficient and precise splicing of Ll . LtrB is crucial for pRS01 conjugative transfer due the requirement for functional relaxase protein for the initiation of conjugation ( Fig . 7 ) . Splicing of the intron depends on the IEP LtrA , which is expressed at low levels under control of an internal promoter [35] , in addition to being autoregulated [36] . This independent regulation of the expression of the IEP LtrA and LtrB relaxase , which control splicing/reverse splicing and conjugation , respectively , provide the opportunity for the intron to act as a sensor to control conjugation , and reciprocally , for the relaxase to promote retrotransposition when conjugation is stimulated and conditions are rife for horizontal gene transfer . This interplay is likely important in stressful environments , when retrotransposition of the group II intron is stimulated , as for example during starvation [37] . Likewise , conjugative elements in different bacterial systems respond to environmental and physiological cues including DNA damage [38] , [39] . Not only does LtrB relaxase provide the first example of a stimulatory factor of RNA-based group II intron retrotransposition in a native host , but also the promiscuous activity of this nickase promotes the link between conjugation and retrotransposition . This scenario is consistent with the unusually high retromobility frequency of the Ll . LtrB intron immediately after conjugation in some transconjugant isolates [19] . Thereby , this link between the mobilization apparatus of a conjugative element and a group II intron provides an example of a stimulatory force for DNA spread among bacterial populations and the infectious transfer of mobile genetic elements which could lead to rapid evolution of bacterial genomes .
All strains and plasmids used in this study are listed in Table 1 . E . coli DH5α , TOP10 and TOP10F′ strains were grown in Luria Broth ( LB ) medium at 37°C with aeration . L . lactis strains were grown in M17 with 0 . 5% glucose ( w/v ) ( GM17 media ) at 30°C without aeration . Where appropriate , the media contained erythromycin ( 150 µg/ml for E . coli and 10 µg/ml for L . lactis ) , spectinomycin ( 50 µg/ml for E . coli and 300 µg/ml for L . lactis ) , or chloramphenicol ( 25 µg/ml for E . coli and 10 µg/ml L . lactis ) . Electroporation of both E . coli and L . lactis was performed with a Gene Pulser apparatus ( BioRad ) . E . coli transformants were recovered in SOC media ( 0 . 5% yeast extract , 2% tryptone , 10 mM NaCI , 2 . 5 mM KCI , 10 mM MgCI2 , 10 mM MgS04 and 20 mM glucose ) for 1 h at 37°C with aeration . L . lactis transformants were recovered in GM17 medium with 0 . 5 M sucrose for 3 h at 30°C without aeration . Plasmid DNA was isolated and purified using a QIAprep Spin Miniprep Kit ( Qiagen ) . The digests and PCR fragments were visualized by electrophoresis in 0 . 7% ( w/v ) agarose gels and stained with ethidium bromide . DNA fragments were purified from agarose gels with QIAquick Gel Extraction Kit ( Qiagen ) . T4 DNA ligase from Promega was used for ligation . The list of oligonucleotides used in this study is in Table 2 . The sequences of all fragments generated by PCR were verified . Construction of the intron donor pLNRK-RIG was accomplished by inserting a fragment of Ll . LtrB:RIG [11] into PstI and SpeI sites of L . lactis/E . coli shuttle vector , pLNRK , a pLE1-based plasmid [40] , containing a nisin promoter and nisR and nisK genes cloned into the ApaLI site . The Ll . LtrB:RIG fragment was generated by PCR with primer pair W1746/W1747 from plasmid pLERIG [11] . The strain D-C ( D = Donor plasmid , Control ) is L . lactis IL1403 transformed with pLNRK-RIG and selected for CamR . To create the D-pRS01 strain , a plasmid pRS01::pTRK28 ( ErmR ) from L . lactis DM2036 [40] was transferred to L . lactis IL1403 by conjugation using plasmid pDL278 ( SpcR ) as a marker for antibiotic resistance in the recipient and selecting for ErmR and SpcR . The pDL278 marker plasmid was cured by growing in the presence of 2 µM ascorbic acid and selecting only for pRS01::pTRK28 ( ErmR ) . The resulting strain ( ErmR , SpcS and CamS ) was transformed with pLNRK-RIG and clones were selected for CamR giving rise to D-pRS01 strain . D-pRS01ΔltrB strain was created in two steps . First , L . lactis IL1403 containing pRS01::pTRK28 was transformed with plasmid pBlueScriptΔltrB::tet ( TetR ) , in which the HindIII fragment of ltrB was replaced by a tet marker , and which does not replicate in L . lactis . Recombinants were selected for ErmR and TetR , and double crossover mutants that resulted in pRS01ΔltrB::tet ( ErmR , TetR ) were screened for the absence of the pBlueScript vector and ltrB gene and the presence of the tetR gene . Recombinants were confirmed by sequencing and the strain carrying pRS01ΔltrB::tet was transformed with pLNRK-RIG followed by selection for CamR . The D-pLtrB strain was made by transformation of the D-C strain with complementation plasmid pLtrB:Ll . LtrB carrying the ltrB relaxase gene under a nisin-inducible promoter . The pLtrB:Ll . LtrB was constructed by amplifying the intron-containing ltrB gene from pRS01::pTRK28 as a template by PCR using primer pair IDT1544/IDT1545 and inserting the resulting fragment into SpeI and SphI sites of the expression vector pCJK21 [41] . The pLtrB:Ll . LtrB plasmid contains an unmarked native copy of the Ll . LtrB intron so as to eliminate the Ll . LtrB group II intron homing site and ensure measurement of retrotransposition and not retrohoming into intron-less ltrB . The relaxase catalytic mutant strains were prepared by transformation of the D-C strain with a plasmid carrying a mutated ltrB relaxase gene . The mutations were introduced using a two-step SOEing ( Synthesis by Overlap Extension ) PCR . First , the fragments of ltrB sequence were amplified , introducing the desired mutations using mismatched primer sequences . Second , the corresponding fragments were mixed and used as the templates to amplify the full-length mutated variants of the ltrB gene . The resulting PCR products were subcloned into TOPO vector using Zero Blunt TOPO PCR Cloning Kit ( Life Technologies ) and clones of interest were detected by colony PCR to confirm the presence of the insert using universal M13 primers . Plasmid DNA was digested with SphI and SpeI for 2 h at 37°C and the insert was ligated into pCJK21 backbone into SphI and SpeI sites . After propagation in E . coli DH5α and confirmation of the insertion with PCR , plasmid DNA was isolated and used for subsequent L . lactis transformations . Retrotransposition assays in L . lactis were performed in specific strains with the intron donor plasmid as previously described [11] . Overnight cultures were grown in GM17 medium supplemented with appropriate antibiotics at 30°C without aeration . Cultures were diluted 1∶100 in fresh GM17 , grown to OD600 of 0 . 2 , and intron expression was induced by addition of nisin at 10 ng/ml . Cultures were grown for an additional 3 h , plated on GM17 medium with and without kanamycin ( 160 µg/ml ) and incubated at 30°C for 2 days . The number of colony-forming units on kanamycin plates was used to estimate retrotransposition frequencies relative to total number of colonies on plates lacking kanamycin . Bacterial colonies growing on the selective plates ( kanamycin 160 µg/ml ) from three replicates of the retrotransposition assay were scooped from the plate surface , washed in phosphate buffered saline ( PBS; 137 mM NaCl , 2 . 7 mM KCl , 10 mM Na2HPO4 , 1 . 8 mM KHPO4 , pH 7 . 4 ) and stored at −70°C . DNA was isolated using DNeasy Blood & Tissue Kit ( Qiagen ) following the protocol designed for purification of total DNA from Gram-positive bacteria . The quality and quantity of the resulting DNA was assessed using electrophoresis in a 0 . 7% ( w/v ) agarose gel stained with ethidium bromide and NanoDrop ( Thermo Scientific ) , respectively . The total DNA ( 2 µg in 100 µl final volume of TE buffer ) was fragmented using a Bioruptor Standard ( Diagenode ) with the following parameters: 30 min of 30 sec on/off cycles at 4°C . The sheared DNA was visualized on a 0 . 7% ( w/v ) agarose gel stained with ethidium bromide and fragments in the range of 200 bp–800 bp were isolated from the gel with a QIAquick Gel Extraction Kit ( Qiagen ) . Preparation of genomic DNA from kanR colonies was followed by ligation of standard adapters , P5A and P7A , containing Illumina flow-cell binding sites . End repair of fragmented DNA was performed using a NEBNext End Repair Module . The NEBNext dA-Tailing Module and NEBNext Quick Ligation Module were utilized for dA-tailing of the end-repaired DNA and adapter ligation of dA-tailed DNA . The Phusion High-Fidelity PCR Master Mix was used for PCR enrichment of adapter-ligated DNA . Purification of the fragments when necessary was performed with QIAquick PCR Purification Kit ( Qiagen ) . Specific primer P-SJ and the universal primer P7 enable amplification of the fragment containing the 5′ end of the kanR gene , the 3′ end of the Ll . LtrB intron , and the flanking genomic sequence of interest . At this step , the amplicons lack the necessary P5 adapter sequence which is reintroduced as ‘chimeric’ primer P5i ( Fig . 3A ) . The region at the 3′ end of P5i is complementary to the short 3′ end fragment of the Ll . LtrB intron , joined by a six-base index sequence and a stretch of six random nucleotides to the 5′ adapter sequence ( Fig . 3A ) . The product of the second PCR has P5A and P7A flow cell binding sites as well as the index sequence , which is specific for each constructed sequence library , allowing multiplex sequencing . Sequence libraries were constructed after selection of KanR retrotransposition events for the different strains , each bar-coded with a unique index sequence ( Fig . 3A and Table 2 ) and sequencing was performed by the Core Facility at SUNY Buffalo ( Buffalo , NY ) . The sequences containing the 3′ flanking genomic sequence , generated on an Illumina HiSeq 2000 , were analyzed , to generate unambiguous profiles for each of the libraries . The adapter and intron sequences were removed from reads using custom BioPhyton script ( scripts available on request ) . Not only were adapter and barcode sequences removed , but also the sequence corresponding to the 3′ end of the intron had to be cut out , leaving only the flanking sequence of interest prior to mapping . The script allows precise detection of the intron fragment within the generated reads followed by the trimming of the intron and the upstream sequence leaving the flanking fragment . The 5 bp at the 3′ end of the reads appeared to be of low quality and were removed as well . The resulting 15–16 bp reads were analyzed . The frequencies and distribution of the reads along the L . lactis IL1403 chromosome could not be compared directly even after digital normalization . This was because Illumina libraries derived from different cultures not only consisted of a different number of chromosomal retrotransposition events ( Fig . 3B ) , but also exhibit striking differences in distribution of these events along the chromosome for different libraries . To overcome this obstacle , we first weeded out all non-chromosomal reads from the original libraries ( including plasmid events and unmapped reads ) . Next , we performed random sampling of the reads for each library , generating new libraries containing only chromosomal reads . BioPython script ( fastq_reservoir_sampling . py ) was used for random sampling of the reads from each library for further comparative analysis . The 1 , 000 randomized subsets of the same size ( 250 , 000 reads ) were automatically generated for each of the libraries . The number of unique insertion points was estimated for each subset and the average numbers are provided in Fig . 4A . The analysis and mapping of the Illumina-generated reads was performed using the Galaxy server ( https://usegalaxy . org/ ) [42] . Bowtie software was used for mapping of the reads under custom settings [43] . The settings were as follows: ‘Maximum number of mismatches permitted in the seed’ was set as ‘0’ ( parameter -n ) , and ‘Whether or not to try as hard as possible to find valid alignments when they exist’ was set to ‘Try hard’ ( parameter -y ) . Additionally , the setting was to report the ‘best’ singleton alignments in terms of stratum ( the number of mismatches ) and in terms of the quality values at the mismatched positions ( parameter –best ) . FASTQ Groomer [44] , SAMtools [45] , deepTools [46] , and BEDTools [47] were also used for analysis among other bioinformatic tools implemented in the Galaxy server ( https://usegalaxy . org/ ) [42] . DNA sequences for 5′ and 3′ flanks of the intron insertion were aligned using ClustalW algorithm incorporated into Unipro UGENE software [48] and compared to the sequence of the Ll . LtrB homing site ( Tables S1–S4 in Text S1 ) . Since we used the whole set of sequences irrespective of the frequencies of the retrotransposition events for each particular insertion point , the scores and nucleotide distribution in each position reflect the whole spectrum of the sequences which might become a potential target for the intron . The alignments were optimized based on the relative frequency of each retrotransposition event to better reflect the overall preferences for the insertion site nucleotide composition in different libraries ( Table S1 in Text S1 ) . Sequence logos were drawn for each of the libraries based on the resulting alignments using WebLogo [49] ( Fig . 3C ) . The cut-off for relative frequency was equal to 5 . 0×10−3 , which means that only retrotransposition events with higher frequencies were used in the present analysis . The nucleotides relative to the insertion point are numbered on the top of each Table ( Table S1 ) : “−” indicates the location upstream relative to the putative insertion site , “+” indicates downstream . The grid profiles and consensus sequences ( Cons . ) were built using Unipro UGENE software [48] and its feature ‘Statistics’ . The E . coli TOP10-pBAD33 host-vector expression system [50] was used to clone and overexpress the ltrB gene in plasmids pLtrB-HIS6 and pLtrB ( Y21A ) -HIS6 . To construct the expression plasmid pLtrB-HIS6 , the full-length intronless ltrB gene was amplified from pCY20 plasmid as template [29] using primer pair IDT3381/IDT3277 . The resulting PCR fragment was digested with SpeI and SphI at 37°C for 2 h . The fragment was ligated into pBAD33 [50] , which had been digested with SphI and SpeI overnight . The expression plasmid pLtrB ( Y21A ) -HIS6 was produced by site-directed mutagenesis from pLtrB-HIS6 with the GeneArt Site-Directed Mutagenesis System ( Invitrogen , Life Technologies ) using the primers IDT3509 and IDT3510 containing a mutation in the corresponding codon to convert Tyr to Ala at position 21 . E . coli TOP10 carrying LtrB-HIS6 or LtrB ( Y21A ) -HIS6 were grown in LB medium supplied with chloramphenicol at 30°C to O . D . ∼0 . 6 . Induction was with 0 . 4% arabinose ( w/v ) for 4 h . Proteins were separated by 10% SDS-PAGE [51] , and visualized by staining with Coomasie brilliant blue . For preparative purposes , cells from 700 ml induced culture were collected by centrifugation and stored at −70°C . Pellets were thawed on ice , suspended in 40 ml of Lysis Buffer ( 50 mM Tris-HCl , pH 8 . 0 , 2 mM EDTA , 10 mM imidazole , 10% [v/v] glycerol ) and incubated on ice for 30 min . Next , cells were lysed by sonication , and debris was removed by centrifugation . The HIS6-tagged proteins were purified using metal chelating chromatography under native conditions at 4°C . Lysis Buffer was used for equilibration of the column . Wash Buffer ( 50 mM NaH2PO4 , 300 mM NaCl , 50 mM imidazole , pH 8 . 0 ) and Elution Buffer ( 50 mM NaH2PO4 , 300 mM NaCl , 200 mM imidazole , pH 8 . 0 ) were used in further purification with the HisTrap HP ( GE Healthcare ) columns containing Ni Sepharose High Performance following the manufacturer instructions . The IDT3584 and IDT3585 oligonucleotides 96-nt in length containing pRS01 oriT in sense and antisense orientation [29] , and the IDT3582 and IDT3583 oligonucleotides 96-nt in length containing R388 oriT in sense and antisense orientation were used for activity assays [52] . Oligonucleotides were labeled with 1–2 U of T4 polynucleotide kinase ( NEB ) and 25 µM γ-[32P]-ATP ( 5×106 cpm/µmol; Amersham ) for 1 h at 37°C followed by purification with illustra MicroSpin G-50 columns ( GE Healthcare ) . The reaction mixture ( 20 µl ) contained relaxase reaction buffer ( 50 mM Tris-HCl pH 7 . 9 , 100 mM NaCl , 10 mM MgCl2 , 1 mM dithiothreitol ) , 100 µg/ml bovine serum albumin ( BSA ) , 20 ng LtrB-HIS6 or LtrB ( Y21A ) -HIS6 protein , and 3 pmol 5′-labeled [32P]-ATP-oligonucleotide primer . After 1 h incubation at 37°C , reaction products were analyzed by denaturing polyacrylamide gel electrophoresis ( PAGE ) ( 8% [w/v] 29∶1 acrylamide/Bis-acrylamide , 7 M urea , 1× Tris/Borate/EDTA buffer ) followed by imaging using phosphor screen in a Typhoon imager ( GE Healthcare ) . The ssDNA was prepared from E . coli TOP10F′ cells transformed with plasmids: pONoriT1 , pONoriT2 , or pONLLglnP-R using M13KO7 helper phage ( NEB ) following the manufacturer protocol . Plasmids pONoriT1 and pONoriT2 were constructed by AT-cloning of the oriT 95-bp fragments from pRS01 [29] and R388 plasmids [52] into pGEM-T vector ( Promega ) in the direct orientation relative to the phage f1 origin ( f1 ori ) carried by pGEM following the manufacturer's recommendation . The following oligonucleotides were used: IDT3584 and IDT3585 which contain oriT from pRS01 [29]; IDT3582 and IDT3583 which contain oriT from R388 [52] . Plasmid pONLLglnP-R was constructed by AT-cloning of the fragment of glnP locus , 1323 bp in length , amplified from L . lactis IL1403 genomic DNA using the following pair of primers: IDT3751 and IDT3752 . Plasmid pONLLglnP-R carried the fragment in direct orientation relative to the phage f1 ori . The ssDNA ( 500 ng ) was incubated for 1 h at 37°C with either LtrB-HIS6 or LtrB ( Y21A ) -HIS6 protein in 50 µl relaxase reaction buffer described above . Treated DNA was used directly in the primer extension reaction . Each extension reaction contained ssDNA ( 50 ng ) , reaction buffer ( 50 mM KCl , 20 mM Tris-HCl pH 8 . 8 , 10 mM MgCl2 , and 100 µg/ml BSA ) , 200 µM dNTP , 3 pmol 5′-labeled [32P]-ATP-oligonucleotide primer , and 1–2 U of Bst DNA polymerase Large Fragment ( NEB ) . The reaction mix was incubated at 65°C for 15 min . The reaction products were analyzed by denaturing PAGE as described above , alongside sequencing ladders ( USB 785001KT ) . The primer IDT3492 is specific for pGEM and was used for all ssDNAs; additionally , glnP-specific primers IDT3752 , and IDT3822 were used .
|
Mobile genetic elements are segments of DNA that capable of “jumping” within a single DNA molecule , between chromosomes or even between cells . They usually encode the enzymes that mediate their own transfer and integration into new DNA locus . The transfer of mobile genetic elements between cells is known as horizontal gene transfer and it is common in Bacteria . Conjugative plasmids are major means to horizontal gene transfer often carrying a variety of putative virulence factors and antibiotics resistance determinants among and within bacterial species . Thus , conjugative plasmids play a crucial role in the plasticity of the genome , allowing bacteria to adjust readily to new environments . Other mobile elements , such as mobile group II introns , were found to be associated with conjugative plasmids . Here , we demonstrated that mobile group II intron and conjugative plasmid interact promoting gene transfer , and potentially providing a mutual benefit to each other .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"lactococcus",
"lactis",
"transfer",
"rna",
"horizontal",
"gene",
"transfer",
"medicine",
"and",
"health",
"sciences",
"genome",
"evolution",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"plasmids",
"gene",
"transfer",
"genetic",
"elements",
"bacterial",
"pathogens",
"genome",
"complexity",
"lactococcus",
"microbial",
"pathogens",
"biochemistry",
"rna",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"mobile",
"genetic",
"elements",
"introns"
] |
2014
|
Interaction between Conjugative and Retrotransposable Elements in Horizontal Gene Transfer
|
The mite Varroa destructor is an obligatory ectoparasite of the honey bee ( Apis mellifera ) and is one of the major threats to apiculture worldwide . We previously reported that honey bees fed on double-stranded RNA ( dsRNA ) with a sequence homologous to that of the Israeli acute paralysis virus are protected from the viral disease . Here we show that dsRNA ingested by bees is transferred to the Varroa mite and from mite on to a parasitized bee . This cross-species , reciprocal exchange of dsRNA between bee and Varroa engendered targeted gene silencing in the latter , and resulted in an over 60% decrease in the mite population . Thus , transfer of gene-silencing-triggering molecules between this invertebrate host and its ectoparasite could lead to a conceptually novel approach to Varroa control .
The European honey bee , Apis mellifera , plays a key role in pollination . One third of the world's food crops as well as many wild plants depend on honey bees for pollination [1] . In recent years , honey bee colonies suffer from severe losses worldwide . Pests and pathogens are mainly involved in the decrease of honey bees vitality and colony losses including: mites , viruses , fungi , bacteria , and other insects [2] . However , the mite Varroa destructor is considered one of the greatest threats to apiculture , not only because of its direct deleterious effect , but also by being a vector of several important bee viruses [3]–[8] . Varroa destructor is an obligatory parasite that feeds on the hemolymph of developing and mature honey bees . Varroa mites invade cells of bee larvae just before they are sealed , feed on the hemolymph of the developing bee and proliferate there . When the adult bee emerges , the attached female mites emerge with it . They may then transfer to another bee or another bee larval cell [3] , [9] . Without proper treatment , honeybee colonies infested with Varroa destructor typically collapse within 2 to 3 years [3] , [10] . Beekeepers use chemicals such as the organophosphate coumaphos , tau-fluvalinate and the formamidine amitraz to control Varroa , but the mites evolve resistance to such chemicals [11]–[14] . Therefore , alternative measures for Varroa control are sought , such as breeding bees for tolerance [3] . Here we report that Varroa gene expression can be modulated by RNA interference ( RNAi ) mediated by the bees , which may lead to a potential new conceptual approach to Varroa control . RNAi is an RNA-mediated sequence specific gene-silencing mechanism [15] . RNAi has been demonstrated to moderate gene expression in a wide variety of organisms including plants , mammals , insects and ticks [16] , [17] . The silencing pathway is initiated by the presence of endogenous or exogenous double-stranded RNAs ( dsRNAs ) that is then cleaved by RNase III-like enzymes resulting in small ( 21–26 bp ) interfering RNAs ( siRNA ) . SiRNAs guide protein complexes to RNAs carrying homologous sequences and target the RNA for degradation , or RNA-directed DNA methylation or chromatin remodeling [16] . Recently , transfer of RNAi from plants to insects and nematodes has been reported [18]–[21] as well as vertical transgenerational transfer of silencing signals [22] . RNAi affecting a germline has also been reported very recently [23] . DsRNA-mediated gene knockdown has also been demonstrated in Varroa by soaking the mite with dsRNA-containing solution [24] . In honey bees , ingestion of dsRNA has been successfully used to investigate gene function as well as for direct application against viral infection and the endoparasite Nosema ceranae [25]–[31] . Our previous indications of the effectiveness of ingested dsRNA-mediated silencing in bees [32] suggested that dsRNA is spread systemically in treated bees . We therefore hypothesized that ingested and systemically spread dsRNA might be horizontally transferred from treated bee to Varroa , in which case the bees could serve as RNAi vectors . Here we demonstrate reciprocal horizontal transfer of dsRNA ingested by honey bees to Varroa mites and then on to Varroa-parasitized bees . A significant phenotype , a measure of Varroa control , was achieved by RNAi that was vectored through bees to silence Varroa-specific genes .
To determine whether dsRNA is transferred from honey bees to Varroa , initial studies were carried out with biologically irrelevant dsRNA marker carrying a segment of the gene for green fluorescent protein ( GFP ) . The use of dsRNA-GFP ( Table S1 ) potentially reduced endogenous sequence “noise” and minimized possible silencing of bee or Varroa gene expression ( the list of dsRNA sequences used throughout this study is shown in Table 1 and the actual sequences are depicted in Table S1 ) . Since Varroa mites suck substantial amounts of hemolymph from both adult and developing bees in sealed brood cells , we tested whether dsRNA is horizontally transferred in both cases . Adult bees were fed a sucrose solution containing dsRNA , resulting in direct transfer of dsRNA from adult bees to phoretic mites feeding on their hemolymph . Adult nurse bees produce jelly that they feed to developing larvae in brood cells , prior to sealing the cells: here , the transfer of dsRNA from a developing bee in a sealed cell to mites feeding on their hemolymph is indirect . To test for direct horizontal transfer , we placed 30 worker bees in plastic containers , and fed them with dsRNA-GFP in a 50% sucrose solution for 8 days . Varroa mites were introduced to the containers on the fifth day of feeding . After 3 days , Varroa that were attached to the bees were removed . Transfer of dsRNA from bees to mites was determined by RT-PCR of templates extracted from the mites and the presence GFP sequence in the mite indicated bee-to-mite transfer ( Figure 1 ) . Mites can absorb dsRNAs and siRNAs by physical contact [24] . To determine any biological function , it was necessary to establish that RNAi triggers were delivered by food ingestion and not by accidental contamination , and that the acquired dsRNA was present in the bee hemolymph ( the food source of Varroa ) . To this end we strapped individual bees to a hollow plastic tube , and fed them directly to their proboscis with dsRNA-GFP . This prevented potential contamination of other bee parts by self or mutual grooming with other bees . On the following day , we released each bee to a separate plastic box , and added two female Varroa mites to each of the above boxes . The Varroa were collected from a remote hive that had not been exposed to dsRNA . On the third day , we collected Varroa attached to the bees , and extracted hemolymph from the bees . DsRNA-GFP was detected in RNA extracted from hemolymph of dsRNA-GFP-treated bees . DsRNA-GFP was also found in Varroa parasitizing the treated bees ( Figure 2 ) , indicating that dsRNA transfer was via ingestion and that treated bees do carry dsRNA in their hemolymph . Pursuant to establishing a hemolymph-mediated RNAi transfer we studied the possibility of indirect horizontal RNAi transfer . We placed ca . 250 worker bees and a laying queen in mini-hives and fed them dsRNA-GFP for 8 days . Varroa mites were introduced into the mini-hives on the fifth day of feeding , and were later collected from sealed larval/pupal cells . The presence of dsRNA-GFP in Varroa mites collected from sealed cells on various days of larval/pupal development indicated indirect transfer of dsRNA from bee larvae ( which were fed by nurse bees that had themselves fed on dsRNA-containing sugar solution ) to Varroa ( Figure 3A ) . Having shown direct and indirect transfer of dsRNA from bees to Varroa , we wanted to test whether a mite that has acquired dsRNA from a bee can then transfer it to another bee that it parasitizes . In the direct transfer experiment , bees were fed dsRNA-GFP in 50% sucrose solution for 8 days , and Varroa mites were added on the fifth day of feeding . To test for bidirectional horizontal transfer , on the eighth day the mites were removed from the dsRNA-carrying bees and introduced into a container with untreated bees for 4 days . DsRNA-GFP was detected in RNA extracts of bees that had not consumed dsRNA , but were parasitized by Varroa mites that previously parasitized dsRNA-carrying bees ( Figure 3B ) . The presence of dsRNA-GFP in the parasitized bees indicated direct reciprocal transfer of dsRNA from bee to Varroa and on to another bee . Once we had established horizontal transfer of physiologically inert dsRNA between bees and Varroa , we explored the possibility that RNAi that originated in the bee might affect gene expression in the mite . When this study was initiated , the Varroa genome had not been elucidated ( recently , partial Varroa genome information has been released [33] ) . Therefore , we designed a number of genes whose silencing was expected to harm the Varroa mite . We chose fundamental housekeeping genes involved in cytoskeleton assembly , energy transfer and transcription . In addition , we chose genes involved in apoptosis inhibition ( assuming that their silencing would enhance apoptosis ) . We aligned several published mite and insect gene sequences and determined conserved regions ( Figure S2 ) . We designed probes to the conserved regions , screened a Varroa cDNA library and isolated the respective Varroa genes . The Varroa dsRNA sequences selected for Varroa gene silencing are presented in Table S1 . To prevent off-target human or bee gene silencing , these sequences did not correspond to any A . mellifera or human genes ( Table S2 ) . In some cases , sequences from the same gene family were selected . The dsRNAs of the selected sequences were prepared as previously described [29] . To determine whether Varroa gene silencing can be mediated via bees that have ingested dsRNA , we placed ca . 250 worker bees and a laying queen in mini-hives . We prepared two mixtures of the Varroa dsRNA: Mixture I contained sequences derived from five Varroa gene sequences ( sequences 1 , 4 , 8 , 12 , and 14 described in Tables 1 and S1 ) and Mixture II contained all 14 Varroa gene sequences ( Tables 1 and S1 ) . Mini-hives fed with Mixture I or II each served as a treatment group . In addition , mini-hives fed with dsRNA-GFP or only sucrose solution served as two control groups . After 1 week of feeding , we introduced Varroa mites every day for a week ( Figure 4 ) . At the end of the 60-day experiment , we sampled Varroa mites from all four treatment groups and determined transcript levels of four selected Varroa genes by real-time RT-PCR ( sequence 4 , homologous to RNA polymerase III; sequence 9 , homologous to vacuolar proton ATPase; and sequence 14 , homologous to apoptosis inhibitor iap1 and 2 ) or semi-quantitative RT-PCR ( sequence 12 , homologous to apoptosis inhibitor FAS ) . The results confirmed that the dsRNA fed to the bees indeed engendered gene silencing in the parasitic Varroa mites , inhibiting expression levels of the tested genes by approximately 35 to 60% ( Figure 5 ) . The transcript levels of the three genes that we analyzed by real-time RT-PCR from Mixture II ( genes 4 , 14 , and 9 ) were significantly lower relative to the untreated and the dsGFP controls ( Figure 5 , A–C , respectively ) . The reduction in transcript levels of genes 4 and 14 did not differ significantly between Mixtures I and II . However , in Mixture I there was a trend for higher transcript levels , especially of the latter gene . Mixture I did not contain sequence 9 , and consistently transcript level of this gene was not affected by Mixture I relative to the untreated and the dsGFP controls . Both mixtures contained the sequence for Varroa gene 12 , and semi-quantitative RT-PCR showed greatly reduced expression of this gene by both mixtures ( Figure 5D ) relative to untreated bees and to bees treated with dsRNA-GFP ( Figure 5E ) , and to expression of actin as a standardizing internal control ( Figure 5F ) . Once we had demonstrated silencing of several Varroa genes , we proceeded to monitor mite survival . First , we tested whether the dsRNA mixtures affect bee survival by counting all mature bees and sealed brood in the mini-hives at the end of the experiment . Bee population size did not differ between control and dsRNA-treated mini-hives ( F3 , 29 = 0 . 62 , P = 0 . 608; Figure 6 ) . The results were similar when brood and adult bees were analyzed separately ( not shown ) . Hence , the dsRNA mixtures were not deleterious to bees , indicating no off-target effect . We proceeded to investigate whether bee-mediated silencing of Varroa genes could reduce the size of Varroa population in infested hives . We determined the number of Varroa individuals per bee by examining the mite population on mature bees and in sealed brood cells at the end of the experiment . Varroa infestation was reduced in mini-hives treated with Varroa dsRNA compared to the controls ( F3 , 29 = 5 . 65 , P = 0 . 0035; Figure 7 ) . The effect was greater with Mixture II , which targeted more genes than Mixture I , reducing Varroa populations by an average 53% compared to the dsRNA-GFP control , and by 61% compared to the untreated control .
Gene silencing following ingestion of dsRNA by honey bees has been previously reported [25]–[32] . Here , we show that the ingested dsRNA can be delivered across species to a bee-hemolymph-dependent parasite . This RNAi transfer may cause silencing of Varroa gene expression and reduce mite populations in hives . Quantitative and semi-quantitative RT-PCR indicated that the dsRNAs affected the targeted Varroa genes . From an RNA biology point of view , the possibility that closely-associated organisms may interact via RNAi pathways should be further explored . The Varroa mite has been demonstrated to be a vector for bee viruses [2] , [6]–[8] . Similarly , we show that the mite can also vector RNAi-triggers , which were acquired from hemolymph of bees that had consumed dsRNA . Additionally , we demonstrated that bees vector biologically active dsRNA to the Varroa mite . The occurrence of such reciprocal interactions raises the hypothesis that bees may be potential vectors for Varroa-affecting viruses . Although the main emphasis of this study is the cross-species transfer and effect of RNAi , it also reflects on a promising concept of Varroa control . With Varroa mites evolving resistance to the chemicals used for their control [11]–[14] , our study provides a novel potential approach for relieving the most serious economic burden on apiculture . Consistent with our findings of the stability of dsRNA in the honey bee colony , it is stable enough to be efficient when administered to the colony in sugar solution [29] , [32] , yet it eventually degrades in an environment favorable to bacterial growth . Varroa mainly devastate colonies when Varroa population grows unchecked . Effective control measures do not necessarily need to completely eradicate the Varroa population [34] , [35] . Further studies would need to monitor the effect of dsRNA treatment on Varroa population dynamics and long-term effect on honey bee colonies under field conditions . Furthermore , the dsRNA formula may be optimized by finding more vital target genes that will lead to a greater reduction of the mite population in hives with less amount of dsRNA fed to infested hive . We prepared two dsRNA formulations: Mixture I targeted 5 Varroa gene sequences and Mixture II targeted 14 Varroa gene sequences . Mixture II tended to reduce Varroa populations more effectively ( Figure 7 ) . This suggests the existence of overlapping metabolic pathways or that some of the gene products are stable and remain active even though their respective gene's expression had been silenced . We selected dsRNA sequences that are not homologous to honey bee ( or human ) sequences . As in other reports [31] , [32] , silencing ( in this case of the Varroa genes ) did not affect the vigor of the bees ( Figure 6 ) . Notably , we did not notice the off-target effects reported by Jarosch and Moritz [36] . Although Varroa infestation was greater in control vs . treatment mini-hives ( Figure 7 ) , this did not affect the strength of the hives at the end of our experiment ( Figure 6 ) . This is not surprising , since Varroa were present in our hives for only about 7 weeks . Varroa damage is cumulative and is minimal in newly infested hives ( hives collapse after 2–3 years [3] , [10] ) .
DsRNA was prepared according to Maori et al . [29] . Sequences were amplified by PCR using specific primers including the 5′ tail of the T7 promoter ( Table S1 ) . PCR products were TA cloned into the plasmid pDRIVE and sequenced . Amplicons were used as template for in-vitro transcription . Total RNA for dsRNA-GFP detection experiments was isolated from a single honey bee or from 10 Varroa mites , using phenol-chloroform extraction ( peqGOLD Trifast , Peqlab ) . Total RNA for Varroa dsRNA experiments or for dsRNA-GFP detection was isolated from 5 Varroa mites or from 50 mites , respectively , with the ZR Tissue & Insect RNA MicroPrep ( Zymo Research ) according to the manufacturer's instructions . Eluted RNA was treated with TURBO DNA-free kit ( Ambion , Austin , TX , USA ) and tested for DNA contamination . Varroa RNA was then co-precipitated with glycogen and 3 M sodium acetate in 70% ethanol and resuspended in 20 µl of RNAse-free water . The amount and quality of the RNA samples were determined by spectrophotometer ( NanoDrop Technologies , Wilmington , DE , USA ) . RNA from hemolymph was extracted using the phenol: chloroform: Isoamyl alcohol method according to published protocol [37] . DsRNA-GFP was detected by RT-PCR using Verso 1-Step RT-PCR ( Thermo Scientific ) with specific GFP primers according to the manufacturer's protocol . Total RNA samples extracted from 10 Varroa or 1 honey bee were used as templates . Northern blot assay for detecting dsRNA-GFP was performed as follows: Samples of RNA ( 500 ng ) were electrophoresed on 1 . 2% agarose gel . The gel was washed with 0 . 25 M HCl , denaturation solution , neutralization solution and 10XSSC before the RNA was transferred to a positively charged nylon membrane ( Roche Diagnostics ) . The membrane was treated according to the manufacturer's protocol , with DIG-labeled probe ( Roche Diagnostics ) of GFP sequence corresponding to the sequence used as template for dsRNA-GFP synthesis . A Varroa cDNA library was prepared using a Smart cDNA construction kit ( Clontech ) according to the manufacturer's instructions . Genes involved in four activity categories were designated . Database-recorded mite and insect genes belonging to those groups were aligned ( Figure S2 ) . Conserved sequences were determined for each group and served as probes for selecting the homologous genes from a Varroa cDNA library . The actual Varroa genes were sequenced . Segments of Varroa genes , 200 to 450 bp in length , which did not correspond in sequence to any bee or human genes ( identity of less than 21 consecutive bases; Table S2 ) , were selected . The selected Varroa sequences and GFP partial sequence are presented in Table S3 . RNA ( 400 ng ) was subjected to reverse transcription with random hexamers using the Verso cDNA synthesis kit ( Thermo Scientific ) . Each sample of the obtained cDNA was diluted 1∶50 before amplification . Real-time quantitative PCR was performed by LightCycler 480 ( Roche ) and was analyzed with the instrument's software . The employed primers and probes are listed in Table S4 . The real-time PCR program was as follows: 95°C for 10 min , followed by 45 cycles of 95°C for 10 s and 60°C for 30 s . At the end , samples were subjected to 40°C for 30 s . 18S rRNA was used as an internal control for the standardization of RNA levels . The semi-quantitative PCR program was as follows: 95°C for 10 min , followed by 40 cycles , each consisting of 95°C for 10 s and 65°C and 55°C for 30 s for the apoptosis inhibitor ( FAS ) and its internal standardization control ( actin ) , respectively , followed by 72°C for 30 s . Every three cycles starting from cycle 31 for FAS and 29 for actin , a tube was taken out , incubated for 5 min at 72°C and stored at −20°C . Samples were analyzed on a 1 . 2% agarose gel . Each qPCR experiment was repeated three times . To test for direct transfer of dsRNA-GFP from adult bee to mite , 1-day-old bees that emerged in an incubator were placed in four plastic containers ( 30 bees per container ) . Two containers were fed with 30 µg dsRNA-GFP in 200 µl of 50% sucrose solution for 8 days , and the other two containers were controls , fed only 50% sucrose solution . Adult female Varroa ( n = 30 ) were introduced into each container on day 5 . After 3 days , Varroa that were attached to bees were removed and collected and their RNA was isolated for dsRNA-GFP analysis . A possible caveat of this experiment was that dsRNA-GFP would be transferred to Varroa through direct contamination with dsRNA-GFP in the containers , or from contact with contaminated bee body parts . We therefore performed an additional experiment to test for dsRNA transfer to Varroa mites via the bee hemolymph . In the morning of the experiment , 150 bees were collected from the area of combs containing open larvae; these tend to be nurse bees . Each bee was strapped into a hollow plastic tube in a manner that ensured their ability to extend their proboscis , and minimized injuries of the bees upon release [38] . The bees were divided into two groups: Individuals in the first group were fed with 2 . 5 µg dsRNA-GFP in a 5 µl 50% sucrose solution . Bees in the second group served as the control group , and were fed with 5 µl 50% sucrose solution only . In order to avoid contamination of dsRNA-GFP on the bee body , the sucrose solution was given directly to the proboscis . In addition , in order to prevent starvation of the bees overnight , both groups ( treated and control ) were fed 5 µl 50% sucrose solution in the evening . The following day , each bee was released gently from the hollow plastic tube , placed in a clean cage and supplied with candy ( 67% sugar powder and 33% honey ) . Two female adult Varroa mites were added to each of the above bees . Varroa were collected from a mite-infested hive that has never been exposed to dsRNA . On the third day , each bee was anaesthetized on ice , and 1–10 µl of hemolymph was collected . The collection of hemolymph was performed by pricking a hole in the inter-segmental membrane between the 2nd and 3rd abdominal segment , and inserting a capillary tube . Prior to hemolymph collection , Varroa mites , which were attached to the bee's body , were collected . All samples ( Varroa mites , hemolymph and bees ) were placed directly on ice , and then stored at −80°C for molecular analysis . To test for bidirectional transfer of dsRNA-GFP from bee to mite and on to another bee , some of the Varroa that had been detached from bees were transferred to containers with newly emerged , untreated bees for 4 days and their RNA was isolated for dsRNA-GFP analysis . Every day , bees in all containers were given an additional 1 ml sucrose solution after finishing their treatment . In addition , bees had free access to a pollen patty consisting of 70% pollen mixed with sugar powder . To test for indirect transfer of dsRNA-GFP from adult bee to bee larva and on to mite feeding on the hemolymph of the developing bee in a sealed cell , a cup of bees ( ca . 250 ) and a laying queen were introduced into each mini-hive ( two replicates in each of two enclosures ) . DsRNA-GFP ( 200 µg per hive ) was provided daily in 5 ml 50% sucrose solution for 8 days . Thirty Varroa mites were introduced to the hives on the fifth day . Adult female Varroa were collected from sealed cells from day 11 till day 30 and their RNA was isolated for dsRNA-GFP analysis . To test the dsRNA stability in the hive , we placed 10 bees in a cage , and exposed them to 10 ml of 50% sucrose solution that contained 200 µg dsRNA-GFP . Hence , dsRNA-GFP final concentration was 20 ng/µl , identical to the dsRNA-GFP concentration that was used in the other experiments . The cage was placed in a vacant second floor of a hive , separated by a screen from the populated bottom floor . Thus the caged bees were exposed to the hive's environment . Samples from the sugar solution were taken on days 1 , 2 , 3 and 6 , and placed on ice . The caged bees died on day 2 , possibly due to a heat wave . Sucrose concentration in the samples was determined with a refractometer , and if needed , equilibration with water was done . Samples were then stored at −80°C until analysis . Samples were analyzed on 1 . 2% agarose gel , loaded with 5 µl from each sample ( 100 ng of dsRNA-GFP at time zero ) . The experiment with Varroa dsRNA was conducted in mini-hives , 12 mini-hives per replicate , and was repeated three times . In each replicate , a cup of bees and a laying queen were placed in each mini-hive . Three mini-hives were randomly assigned to one of four netted enclosures , each representing a different feeding treatment . Bees were fed 5 ml of 50% sucrose solution in troughs placed in each mini-hive . The four treatments were: 1 ) sucrose solution only ( untreated control ) , 2 ) Mixture I ( 200 µg each of five dsRNAs added to the sugar solution ) , 3 ) Mixture II ( 200 µg each of 14 dsRNAs added to the sugar solution ) , and 4 ) dsRNA-GFP ( 200 µg dsRNA ) serving as an inert dsRNA control . Mini-hives that fully consumed the treatment solutions were supplemented with candy ( 67% sugar powder and 33% honey ) . In addition , the bees were routinely fed pollen patties ( 70% pollen and 30% sugar powder ) . Each replicate of the experiment lasted for 60 days ( Figure 4 ) . Bees in each treatment were fed the respective solution daily for the first 10 days and for the last 14 days , and twice a week in the interim . Varroa mites were introduced into each mini-hive from day 7 till day 14 . In the first replicate , 30 mites were introduced into each mini-hive; in the latter two replicates , 100 mites were introduced into each mini-hive . On day 60 , all mature bees were collected , counted and shaken with 70% ethanol overnight in order to collect and count Varroa mites that fell off the bees . All capped brood cells were opened to collect and count Varroa mites . We calculated mites per bee ( mature and developing ) . Varroa mites , adult bees , emerging bees and pupae were stored for molecular analyses . Statistical analyses were conducted with JMP statistical software version 9 ( SAS Institute , Cary , NC , USA ) . Statistical significance was set at P<0 . 05 . To test for significant differences in relative expression , one-way ANOVA was conducted on ddCt values [39] . Treatment was the main factor . To test for differences in Varroa mite population , two-way ANOVA was conducted on numbers of Varroa per bee in a block design with treatment as main effect and experimental replicate as block . To test for differences in total bee population , a similar two-way ANOVA was conducted on the total number of bees ( capped brood and adults ) . Significant differences between treatments were tested by the Tukey-Kramer ( HSD ) test .
|
Acquisition of RNAi components ( dsRNA , siRNA ) by ingestion and their spread within the recipient organism has been previously reported by us and others . Here we extend such observations , demonstrating cross-species horizontal transmission of dsRNA which , upon transmission from one organism to another still retains its biological activity . We show that dsRNA ingested by honey bees is further transmitted to the parasitic mite Varroa destructor that feeds on the honey bee's hemolymph . Reciprocally , dsRNA-carrying Varroa transmits the dsRNA back to bees . Furthermore , we demonstrate that bees ingesting dsRNA of Varroa gene sequences become vectors of dsRNAs , transmitting the signals to the Varroa , thus engendering silencing of mite genes and resulting in a significant phenotype , Varroa mortality . The exchange of active silencing signals between the honey bee and the mite suggests a potential RNA-based interaction between invertebrate hosts and parasites . Furthermore , our results offer a potentially conceptually new control measure for the mite Varroa destructor , which is one of the greatest threats to apiculture .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology",
"veterinary",
"science",
"agriculture"
] |
2012
|
Bidirectional Transfer of RNAi between Honey Bee and Varroa destructor: Varroa Gene Silencing Reduces Varroa Population
|
Lysozymes are nearly omnipresent as the first line of immune defense against microbes in animals . They exert bactericidal action through antimicrobial peptide activity and peptidoglycan hydrolysis . Gram-negative bacteria developed several weapons to battle lysozymes , including inhibitors of c-type lysozymes in the MliC/PliC family and the Neisseria adhesin complex protein ( ACP ) . Until the recent discovery of ACP , no proteinaceous lysozyme inhibitors were reported for the genus Neisseria , including the important human pathogen N . gonorrhoeae . Here , we describe a previously unrecognized gonococcal virulence mechanism involving a protein encoded by the open reading frame ngo1063 that acts to counteract c-type Iysozyme and provides a competitive advantage in the murine model of gonorrhea . We named this protein SliC as a surface-exposed lysozyme inhibitor of c-type lysozyme . SliC displays low overall primary sequence similarity to the MliC/PliC inhibitors , but we demonstrate that it has a parallel inhibitory mechanism . Our studies provide the first evidence that bacterial proteinaceous lysozyme inhibitors protect against host lysozyme during infection based on lack of attenuation of the ΔsliC mutant in lysozyme knock-out mice , and that the conserved residues involved in lysozyme inhibition , S83 and K103 , are functionally indispensable during infection in wild type mice . Recombinant SliC completely abrogated the lytic activity of human and chicken c-type lysozymes , showing a preference towards human lysozyme with an IC50 of 1 . 85 μM and calculated KD value of 9 . 2 ± 1 . 9 μM . In contrast , mutated SliC bearing S83A and K103A substitutions failed to protect fluorescein-labeled cell-wall from lysozyme-mediated hydrolysis . Further , we present data revealing that SliC is a surface-displayed lipoprotein released in membrane vesicles that is expressed throughout all phases of growth , in conditions relevant to different niches of the human host , and during experimental infection of the murine genital tract . SliC is also highly conserved and expressed by diverse gonococcal isolates as well as N . meningitidis , N . lactamica , and N . weaveri . This study is the first to highlight the importance of an anti-lysozyme strategy to escape the innate immune response during N . gonorrhoeae infection .
The first line of host immune defense against bacterial pathogens in plants and in invertebrate and vertebrate animals involves degradation of peptidoglycan through innate immune system components such as lysozymes [1–3] . Peptidoglycan ( murein ) is the major structural component of the bacterial cell envelope that provides resistance against turgor pressure and prevents cell death due to lysis . Peptidoglycan forms a giant three-dimensional network built of linear glycan strands of alternating β- ( 1 , 4 ) linked N-acetylmuramic acid and N-acetylglucosamine sugars that are cross-linked by short peptides [4] . Lysozymes are powerful host weapons exerting bacterial killing by hydrolytic action on the glycosidic bond linking the sugars , which breaks peptidoglycan [3 , 5] . Three main classes of lysozymes have been distinguished within the animal kingdom including c- ( conventional or chicken ) , g- ( goose ) , and i- ( invertebrate ) type lysozyme [2 , 3] . The mammalian lysozyme-like gene family consists of lysozyme c , lactalbumin and calcium-binding lysozyme [5] . Recent studies , however , showed the presence of eight additional diverse types of lysozyme-like genes in the genome of the common ancestor of all extant mammals , and ten lysozyme-like sequences distributed over five chromosomes in humans [6] . Therefore , it is not surprising that bacteria have evolved sophisticated mechanisms to escape killing via murein hydrolysis by chemical modifications of the peptidoglycan backbone and by synthesis of proteinaceous lysozyme inhibitors [7 , 8] . The latter anti-lysozyme strategy targets i- , g- , or c-type lysozymes and encompasses at least five inhibitor families distributed exclusively in Gram-negative bacteria and predominantly in Proteobacteria [8] . Among the inhibitors of c-type lysozyme are Ivy ( Inhibitor of vertebrate lysozyme ) and Ivy-like proteins , the MliC/PliC family ( membrane-bound or periplasmic lysozyme inhibitor of c-type lysozyme , respectively ) , and the Neisseria Adhesin Complex Protein , ACP [8 , 9] . MliC/PliC proteins and ACP have been implicated in host colonization based on their ability to at least partially protect bacteria against challenge with their cognate lysozyme in vitro [8–12] . Surprisingly , to date , only two reports have investigated the role of lysozyme inhibitors in in vivo virulence . These studies demonstrated the importance of MliC from Avian Pathogenic Escherichia coli and Edwardsiella tarda in a subcutaneous chicken infection model and a turbot model , respectively [13 , 14] . N . gonorrhoeae colonizes mucosal surfaces of the reproductive tract , pharynx , rectum and conjunctiva in its sole human host ( 15 ) . During colonization of these different niches , the gonococcus may encounter c-type lysozyme in tears , saliva , urine , serum , vaginal fluid , cervical mucus , and in the lysosomal granules of neutrophils and macrophages ( 3 , 16 , 17 ) . Until the recent discovery of ACP , no proteinaceous lysozyme inhibitors were reported for the genus Neisseria , including the important human pathogen , N . gonorrhoeae [8 , 9] . Here we describe a previously unrecognized N . gonorrhoeae virulence mechanism involving a protein encoded by the open reading frame ( ORF ) ngo1063 in N . gonorrhoeae FA1090 that acts as a weapon to counteract the innate immunity effector c-type lysozyme and significantly impacts N . gonorrhoeae fitness in the female mouse genital tract . We identified NGO1063 in high-throughput proteomic investigations geared towards the discovery of potential gonorrhea vaccine antigens and therapeutic targets in N . gonorrhoeae cell envelopes and naturally released membrane vesicles ( MVs ) [15 , 16] . NGO1063 is annotated in UniProt as a membrane-bound lysozyme inhibitor of the c-type lysozyme family ( Protein Q5F7V4_NEIG1 ) . NGO1063 displays low overall sequence similarity to the MliC/PliC protein family , but we demonstrate that it has a parallel inhibitory mechanism and provide the first evidence that the conserved residues involved in lysozyme inhibition are functionally critical during in vivo infection . Supporting these findings , gonococci lacking NGO1063 show comparable fitness to wild type bacteria during competitive infection in lysozyme-defective mice , but are attenuated in wild-type mice of the same background . We also present data showing that NGO1063 is exposed to the extracellular milieu , expressed during in vivo infection , and conserved in pathogenic and commensal Neisseria , highlighting the importance of this anti-lysozyme strategy during host colonization . We named this protein SliC as a surface-exposed lysozyme inhibitor of c-type lysozyme .
Members of different lysozyme inhibitor families show little similarity at the primary amino acid sequence level [8] . Despite the low sequence identity ( 24–39% ) of the PliC and MliC type inhibitors , these inhibitors display the same specificity and share a specific sequence motif as well as an eight-stranded antiparallel ß-barrel structural topology [8 , 11 , 12] . In silico analysis of the deduced amino acid sequence of our new proteome-derived antigen/drug target [15 , 16] , sliC , revealed a predicted signal peptide with a lipobox motif ( LSLAAC ) containing an invariant cysteine ( C18 ) and a putative MliC domain ( residues 47–118; Figs 1A and 2A ) . Comparison of the primary amino acid sequence of SliC with other characterized MliC proteins from E . coli , Salmonella enterica and Pseudomonas aeruginosa showed only 18 . 8–23 . 3% overall identity ( S1 Table ) . However , the additional conserved regions of the COG3895 domain in the MliC/PliC family , SxSGAxY and YxxxTKG [8] , closely resembled those found in SliC ( Fig 1A ) . Further , the P . aeruginosa MliC residues S89 and K103 , involved in interaction with hen egg white c-type lysozyme ( HEWL ) [11] , were present in SliC ( S83 and K103 ) , which together suggested that ngo1063 encodes a c-type lysozyme inhibitor . To further compare SliC to other lysozyme inhibitors of the MliC/PliC family and Neisseria ACP , a phylogenetic analysis was performed ( Fig 1B ) . Two distinct clusters were revealed in the phylogenetic tree; one cluster contained lysozyme inhibitors found in the genomes of Neisseria species , while the second cluster was comprised of MliC and PliC proteins from more diverse bacterial species . These results indicated that meningococcal and gonococcal SliC and ACP are more closely related to each other than to MliC/PliC lysozyme inhibitors from other bacterial species , despite their low primary sequence identity ( S1 Table ) . To extend these studies , we analyzed SliC sequence diversity among N . gonorrhoeae and other Neisseria species using the translated amino acid sequences from 10 alleles found in N . gonorrhoeae ( Fig 1C ) and the 224 alleles found in the population of over 44 thousand Neisseria isolates present in the PubMLST database ( S1 Fig ) . These phylogenetic analyses showed that all SliC alleles are closely related in pathogenic and commensal Neisseria and allele 29 , of which FA1090 SliC is a member ( Fig 1C , red box ) , represented nearly 98% of the alleles found in N . gonorrhoeae ( 3 , 787 out of 4 , 990 N . gonorrhoeae isolates; 1 , 111 isolates had no value for the sliC locus ) . Analysis of single nucleotide polymorphisms revealed the existence of 9 and 157 polymorphic sites within all alleles found in N . gonorrhoeae ( Fig 1D ) and in all Neisseria ( S2 Fig ) , respectively , further confirming high conservation of SliC . Finally , to assess the conservation at the level of immune-recognition , we probed whole cell lysates from 37 temporally , geographically and genetically diverse N . gonorrhoeae isolates , including common laboratory strains; clinical isolates collected from public health clinics in Baltimore from 1991 to 1994 and Seattle from 2011 to 2013 [16]; and the 2016 WHO reference strains [17] with antiserum against FA1090 SliC ( Fig 1E ) . SliC was detected in all strains tested , and in pathogenic N . meningitidis; commensal N . lactamica; and a human opportunistic pathogen typically associated with canine bite wounds , N . weaveri ( Fig 1F ) . Cumulatively , these findings demonstrated a high conservation of SliC and a similar cellular pool in pathogenic N . gonorrhoeae and N . meningitidis . Lysozyme inhibitors in their natural forms are difficult to obtain with high purity , therefore these proteins have been examined primarily as recombinant proteins [9 , 12 , 14 , 18 , 19] . To initiate characterization of SliC , we engineered recombinant versions of wild type SliC ( rSliC ) and its mutated variant ( rSliC* ) bearing alanine substitutions in residues S83 and K103 . Both recombinant proteins lacked signal peptides and contained a C-terminal Tobacco Etch Virus ( TEV ) protease cleavage site followed by a 6×His-tag ( Fig 2A ) . The two proteins were purified in a soluble form to homogeneity through several chromatography steps and migrated on SDS-PAGE according to the predicted molecular mass of 11 . 65 kDa ( Fig 2B ) . Untagged rSliC was used to obtain polyclonal rabbit antisera . Subsequently , we utilized N . gonorrhoeae FA1090 to create a null mutant in ngo1063 ( ΔsliC ) and complemented strains , ΔsliC/P::sliC and ΔsliC/P::sliC* , in which either wild type sliC or mutated sliC* bearing S83A K103A substitutions were expressed under the control of the native sliC promoter ( Fig 2A ) . The polyclonal rabbit anti-SliC antisera recognized SliC within whole cell lysates obtained from wild type and both complemented strains , but no signal was detected in ΔsliC ( Fig 2C ) . Lipoproteins traverse the cell envelope and may be anchored through their invariant triacylated cysteine residue to the periplasmic side of the inner or outer membrane , or may be translocated across the outer membrane for cell-surface decoration [20–22] . To examine SliC localization , cell sub-fractionation experiments coupled with immunoblotting were performed . Corroborating our proteomic investigations [15 , 16] , these experiments demonstrated the presence of SliC in the cell envelopes and naturally released membrane vesicles . The outer membrane proteins BamA and BamD were also present in this fraction , and as expected , the periplasmic and cytoplasmic markers , SurA and Zwf , respectively , were found primarily in the periplasmic and cytoplasmic sub-proteomes ( Fig 2D ) . Previously applied in silico approaches categorized SliC as an extracellular protein [15]; however , a reliable method to predict whether a given lipoprotein is surface-exposed or faces the periplasmic side of the outer membrane is currently lacking [22] . Therefore , we utilized proteinase K shaving assays and dot blots as two independent experimental methodologies to further examine SliC localization . Exposing intact gonococci to increasing concentrations of proteinase K resulted in detection of lower amounts of SliC , indicating that it is accessible to external protease , similarly to the surface protein marker BamA ( Fig 2E ) . In contrast , BamD , which is anchored to the periplasmic side of the outer membrane [23] , remained unaffected . Likewise , SurA and Zwf resisted proteinase K challenge , verifying the intactness of N . gonorrhoeae cells during the experimental procedures . SliC was also detected on the cell surface of wild type , ΔsliC/P::sliC , and ΔsliC/P::sliC* when analyzed by an immunological dot blot assay , in which suspensions of intact gonococci are spotted on nitrocellulose filters followed by incubation with the antisera ( Fig 2F ) . In these experiments , as expected , BamA was detected on the cell surface in all strains , whereas no signal was observed for BamD , SurA and Zwf unless the cells were lysed . Together , these studies showed that SliC is a surface-displayed lipoprotein released to the extracellular milieu in native membrane vesicles and that alteration of residues predicted to interact with c-type lysozyme does not affect its localization . To examine whether SliC acts as an inhibitor of c-type lysozyme , we employed an assay that takes advantage of a highly sensitive quenched fluorescent lysozyme substrate–fluorescein-labeled cell walls of Micrococcus lysodeikticus . As expected , addition of either HEWL or human lysozyme c ( HL ) to the bacterial suspension triggered hydrolysis of the ß- ( 1 , 4 ) bond between N-acetylmuramic acid and N-acetylglucosamine in peptidoglycan and resulted in robust fluorescence dequenching ( Fig 3A , 3B and 3D ) . These reactions were abrogated upon addition of increasing amounts of rSliC . Comparison of the inhibition of HEWL and HL activity by incubation with molar equivalent concentrations of rSliC demonstrated drastically reduced HL lytic activity in the presence of 1 . 25 μM rSliC , whereas no significant effect was observed for HEWL ( Fig 3C ) . Furthermore , the mutated SliC variant , rSliC* , failed to inhibit HL-driven cell-wall hydrolysis , confirming that S83 and K103 were critical residues for the SliC-HL interaction ( Fig 3D ) . The half-maximal ( 50% ) inhibitory concentration , IC50 , of SliC against the lytic activity of HL towards M . lysodeikticus peptidoglycan was determined to be 1 . 85 μM ( Fig 3E ) . To gain further insights into interactions between SliC and HL , we employed Bio-Layer Interferometry ( BLI ) , a label-free biophysical method that provides kinetic data for protein-protein interactions . It is similar to Surface Plasmon Resonance but is less affected by changes in sample composition [24 , 25] . Biotinylated rSliC was immobilized on streptavidin sensors and incubated with increasing concentrations of HL . BLI experiments were executed using a steady state method and curve fitting of the association and dissociation responses . Curves were fitted to a biphasic binding model , yielding a calculated KD value of 9 . 2 ± 1 . 9 μM ( average ± SEM; Fig 3F ) . This result suggests a moderate binding of SliC to HL in vitro , which is very similar to the ACP-HL interaction with a KD of 11 μM [9] . Cumulatively , our biochemical studies demonstrated that SliC is an efficient inhibitor of c-type lysozyme with functionally pivotal residues S83 and K103 and suggest that SliC is a better inhibitor of HL than HEWL . Complete elimination of SliC or replacing SliC with its mutated variant , SliC* , expressed at the native level ( Fig 2C ) had no effect on the growth rate of ΔsliC or ΔsliC/P::sliC* strains in liquid media ( Fig 4A ) . Similarly , no fitness defects were observed when ΔsliC was exposed to conditions that more closely resemble micro-ecological niches encountered by gonococci in the human host such as iron deprivation , the presence of normal human sera and anaerobiosis ( Fig 4B ) . We next studied whether SliC protects N . gonorrhoeae from the hydrolytic activity of HL in the presence of the outer membrane permeabilizing protein , lactoferrin . Surprisingly , even at HL concentrations higher than are physiologically relevant , absence of SliC did not render the cells more susceptible ( Fig 4C ) . We hypothesized that the lysozyme-blocking function of ACP may contribute to the observed phenomenon [9 , 26 , 27] . To dissect the possible functional relationship between SliC and ACP , we constructed an acp clean knockout and ΔacpΔsliC mutant in the parental FA1090 background . Subsequently , all mutant strains concurrently with the corresponding complemented strains were incubated with lactoferrin at 5 mg/mL and increasing concentrations of lysozyme . The single ΔsliC and Δacp mutants had similar sensitivity towards lysozyme as the wild type cells while the double mutant exhibited 70 , 370 and 67 , 000 fold reduction in survival when exposed to 5 , 10 and 20 mg/mL of lysozyme , respectively ( Fig 4C; green line ) . This sensitivity phenotype was restored to wild type resistance in the complemented strain carrying acp and sliC placed under the control of their own promoters . These studies suggested that SliC and ACP work interchangeably and each of the lysozyme inhibitors , in addition to the other mechanisms used by N . gonorrhoeae to resist killing by lysozyme e . g . lytic transglycosylases [9 , 26 , 27] , is sufficient to provide resistance to lysozyme in vitro . Exposure of ΔsliC and Δacp knockouts to additional cell envelope permeabilizing agents ( antimicrobial peptide polymyxin B , bile salts , or nonionic surfactant Tween 20 ) in combination with lysozyme and lactoferrin treatment resulted in statistically significant decrease in viability of ΔsliC bacteria solely in the presence of the non-physiologically-relevant Tween 20 ( Fig 4D ) . To test whether lack of the surface–exposed SliC , ACP or both proteins have effects on cell envelope integrity , LOS , or cell morphology , we carried out Etest antibiotic susceptibility experiments ( S2 Table ) , isolated and stained LOS , and performed scanning electron microscopy studies . No differences were noted in LOS migration and abundance ( Fig 4E ) , nor were the overall morphology or cell sizes altered in any gonococcal strain ( Fig 4F ) . In addition , solely the ΔacpΔsliC mutant showed a two-fold decrease in minimal inhibitory concentration ( MIC ) towards benzylpenicillin , suggesting modest disruption in cell envelope homeostasis ( S2 Table ) . Surface-exposed lipoproteins are linked to a wide range of roles including bacterial adhesion [22 , 28] . In Neisseria , the lipoprotein heparin-binding antigen ( NHBA ) facilitates adhesion to the surface of human epithelial cells [29] and ACP functions as both a lysozyme inhibitor and an adhesin in N . meningitidis [9 , 30] . Therefore , to examine whether SliC plays a similar dual function in N . gonorrhoeae , we performed time-course adhesion experiments and invasion assays with human cervical epidermal carcinoma ME180 cells in vitro . Similar numbers of adherent ( Fig 5A ) and internalized ( Fig 5B ) gonococci were recovered for wild type , ΔsliC , and both complemented strains at all timepoints , demonstrating that SliC does not contribute to bacterial association with and invasion to human cervical ME180 cells . To further assess SliC function in N . gonorrhoeae biology , expression of the protein was examined in vitro and in vivo during colonization of the female mouse genital tract ( Fig 6 ) . These studies showed that gonococci produce SliC throughout different growth phases in liquid media , similar to the ubiquitously-expressed MetQ [16] , which was used as a comparison ( Fig 6A and 6B ) . SliC was also expressed at comparable levels under different host-relevant conditions , such as iron deprivation and in the presence of normal human sera . Interestingly , an increase in the SliC cellular pool was observed during anaerobic growth ( Fig 6C ) . The E . coli MliC and Ivy proteins are both part of the Rcs extra-cytoplasmic stress regulon that responds to lysozyme insult by increased expression [31] . We wondered whether sliC induction could result from exposure to lysozyme . To test this hypothesis , wild type N . gonorrhoeae was incubated in the presence of lactoferrin and increasing concentrations of HL . No significant changes in the SliC cellular pool between untreated and lysozyme-treated cells were revealed by immunoblot analysis ( Fig 6D ) . Different members of the MliC/PliC family have been implicated in bacterial pathogenesis [8] . However , studies which demonstrate their expression in vivo are lacking to date . Therefore , we next sought to examine expression of SliC during N . gonorrhoeae colonization of the female mouse vaginal tract . Female BALB/c mice were infected with wild type FA1090 and vaginal washes were collected at days 1 , 3 , and 5 post-infection and pooled . Equal numbers of bacterial cells ( CFUs ) were separated by SDS-PAGE and probed with anti-SliC antisera . These studies showed that SliC was readily produced throughout the 5-day infection period ( Fig 6E ) . Densitometry analyses using SliC abundance on day 1 as a reference showed that the amounts of SliC lowered to 0 . 73 fold on day 3 postinfection and increased to 1 . 46 fold on day 5 when compared to the amount of SliC detected on day 1 . Cumulatively , these experiments demonstrated that N . gonorrhoeae stably produces SliC during different growth conditions in vitro and provide the first evidence for the expression of a MliC/PliC protein in vivo , suggesting that maintenance of this lysozyme c inhibitor is important during infection . To test the potential significance of SliC in gonococcal virulence , we performed preliminary in vivo competitive infection experiments with the ΔsliC mutant and the wild type parent strain in the female mouse model of gonococcal genital tract infection . In two independent experiments , the lack of SliC caused a dramatic 10 , 250 , and 167-fold attenuation of N . gonorrhoeae colonization on days 1 , 3 , and 5 post-infection , respectively ( S3 Fig ) . We next conducted competitive infection experiments in which the ΔsliC mutant and the ΔsliC/P::sliC and ΔsliC/P::sliC* complemented mutants were competed with the wild type strain in parallel . The competitive indices obtained for all strains did not differ during in vitro co-culture ( Fig 7A ) , whereas results from three independent coinfection experiments demonstrated a severe reduction in the relative number of CFUs recovered from vaginal swabs for the ΔsliC knockout and ΔsliC/P::sliC* mutant ( Fig 7B–7D , respectively ) . The recovery of the ΔsliC mutant was approximately 2- , 58- , and 34-fold ( geometric mean CIs ) lower than the wild type strain on days 1 , 3 , and 5 post-inoculation , respectively ( Fig 7B ) . The fitness disadvantage was exacerbated for the ΔsliC/P::sliC* bacteria , with averages of 34 , 270 , and 2 , 375-fold lower than for wild type gonococci on days 1 , 3 , and 5 post-infection , respectively ( Fig 7D ) . In contrast , the defect was fully restored by genetic complementation in the ΔsliC/P::sliC strain ( Fig 7C ) . When Kruskal-Wallis Dunn’s multiple comparison tests were performed to compare statistical significance of CIs between the ΔsliC/wild type and ΔsliC/P::sliC/wild type , the values were >0 . 99 , 0 . 027 , and 0 . 77 on days 1 , 3 , and 5 post-inoculation . The calculated p values for the ΔsliC/P::sliC*/wild type versus ΔsliC/P::sliC/wild type were all significant and reached 0 . 029 , 0 . 0058 , and 0 . 016 . To test the basis for the attenuation of ΔsliC bacteria relative to wild type , we next conducted competitive infections in mice that were genetically defective for lysozyme using C57BL/6J mice as the background controls . Data from two combined experiments showed that the relative recovery of the ΔsliC mutant from C57/BL6J mice was approximately 3- , 372- , and 198-fold lower ( geometric mean CIs ) than the wild type strain on days 1 , 3 , and 5 post-inoculation , respectively ( Fig 7E ) . In contrast , the ΔsliC mutant was recovered at the same level as the wild type strain in mice that lack lysozyme on days 1 and 3 , while on day 5 the geometric mean CI was 10-fold lower than the wild type strain ( Fig 7F ) . The difference in the CIs in C57BL/6J mice and lysozyme knockout mice was statistically different on day 3 ( p value of 0 . 03 ) . Only two reports to date examined the in vivo role of lysozyme inhibitors in bacterial virulence [13 , 14] but it has never been dissected whether the virulence defects observed were dependent on the lysozyme inhibition function . Together these studies demonstrated that SliC provides a significant fitness benefit to N . gonorrhoeae in the murine model and offered the first evidence that a proteinaceous lysozyme c inhibitor is important during in vivo infection solely due to its ability to bind and inactivate lysozyme .
To establish infection , microbes must evade a combination of host antimicrobial peptides and enzymes , including lysozymes , which are abundantly secreted by the epithelium and produced within professional phagocytes [32] . Described for the first time by Alexander Fleming in 1922 as a substance with the ability to “lyse” bacteria , lysozyme additionally acts as an antimicrobial peptide by interacting directly with cell membranes via its positively charged amino acids [33] . Pathogenic and commensal bacteria have developed many methods to evade lysozyme; however , until the recently reported ACP [9] , no proteinaceous lysozyme inhibitor has been described for the genus Neisseria . ACP is a surface-exposed meningococcal adhesin that induces cross-strain bactericidal antibodies [30] . Despite low primary sequence homology , the apo crystal structure of N . meningitidis ACP resembles an eight-stranded antiparallel ß-barrel similar in the overall fold to those of MliC/PliC proteins [9 , 11 , 12 , 19] . In vitro assays have demonstrated that ACP proteins act as inhibitors of HEWL and HL , contributing to lysozyme tolerance in the presence of the membrane-permeabilizing agent lactoferrin [9 , 30] . Nevertheless , the importance of lysozyme inhibition for host colonization has not been addressed in Neisseria and investigations in other bacterial species are also sparse [13 , 14] . In this report , we demonstrated for the first time that inhibition of lysozyme activity is pivotal for N . gonorrhoeae mucosal infection in the female genital tract . Through genetic , biochemical , in vitro and in vivo functional assays , we characterized a previously unrecognized surface-exposed protein , SliC , that plays a role in virulence as a weapon to counteract c-type lysozyme and contributes to N . gonorrhoeae colonization . We originally identified SliC ( NGO1063 ) as a potential vaccine/drug candidate in quantitative proteomic surveys of the N . gonorrhoeae cell envelopes and naturally released membrane vesicles [15 , 16] . We found that SliC belongs to the MliC/PliC family of lysozyme inhibitors ( Fig 1 ) and is actually closer to subgroup 1 MliCs , represented by E . coli MliC , than to subgroup 2 MliCs , represented by P . aeruginosa MliC ( Fig 1A ) . SliC possesses the FWSKG motif commonly present in subgroup 1 MliCs and does not have any of the conserved hydrophobic residues characteristic of subgroup 2 [10 , 14] . Our phylogenetic analysis , however , demonstrated that Neisseria SliC and ACP , despite their low primary amino acid sequence identity ( S1 Table ) , were more closely related to each other than either protein was to MliC/PliC lysozyme inhibitors from other bacteria ( Fig 1B ) . MliC proteins are found anchored to the periplasmic side of the outer membrane , whereas PliC family members are localized in the periplasm [8] . In our proteomic investigations , SliC was found in cell envelopes and naturally released membrane vesicles [15] . We confirmed these initial observations by performing sub-cellular fractionation experiments coupled with immunoblotting using antisera specific to SliC as well as control protein markers ( Fig 2D ) . In addition , we showed by two independent approaches–protease accessibility studies ( Fig 2E ) and immunodotting ( Fig 2F ) –that SliC is a surface-displayed lipoprotein , which we predicted earlier with in silico approaches [15] . Similarly , the Mycobacterium tuberculosis lysozyme inhibitor , LprI , is a surface-exposed lipoprotein [34] . Surface-localization of ACP and SliC appears to be another smart strategy employed by Neisseria to guard against the devastating action of lysozyme before it traverses the outer membrane , which is facilitated by the permeabilizing action of other host antimicrobial peptides . The conservation and expression of SliC throughout phases of growth , in conditions relevant to different micro-niches of the human host , during experimental infection of female mice , and among diverse gonococcal isolates , as well as in N . meningitidis , N . lactamica , and N . weaveri , further highlights the importance of maintaining this surface-exposed outer membrane protein in pathogenic , commensal and opportunistic Neisseria ( Figs 1C–1F , 6 and S1–S2 ) . To better understand the mechanism of lysozyme inhibition , different biochemical approaches were utilized for several members of the MliC/PliC family and for ACP [8 , 9 , 11 , 12 , 14 , 18 , 19 , 34] . Our studies with untagged rSliC demonstrated that SliC inhibits both HEWL and HL ( Fig 3A and 3B ) but is a more potent inhibitor of HL ( Fig 3C ) with a calculated IC50 of 1 . 85 μM . Similar observations have been made for N . meningitidis ACP [9] . Site-directed mutagenesis of the two residues in SliC corresponding to the sites of P . aeruginosa MliC interaction with HEWL [11] , S83 and K103 , led to complete abrogation of lysozyme hydrolytic activity ( Fig 3D ) , suggesting that SliC employs a similar mode of lysozyme inhibition as MliC proteins . To further evaluate the SliC-HL interaction , BLI was utilized , which yielded a calculated KD value of 9 . 2 ± 1 . 9 μM ( average ± SEM; Fig 3F ) . This result suggests a moderate binding of SliC to HL in vitro , which is very similar to the ACP-HL interaction , with a KD of 11 μM . Intriguingly , MliC from Brucella abortus shows a thousand-fold higher binding affinity [12] . It is likely that SliC and ACP lysozyme binding affinities are influenced by the presence of peptidoglycan , the lysozyme substrate . Our complementary biochemical approaches demonstrated that SliC is an inhibitor of c-type lysozyme ( Fig 3 ) ; however , exposure of N . gonorrhoeae to HL did not result in any significant differences in bacterial viability in comparison to the wild type strain ( Fig 4C ) . We concluded that other mechanisms employed by N . gonorrhoeae to resist killing by lysozyme , ACP and lytic transglycosylases , for example [9 , 26 , 27] , compensated for the lack of SliC and were sufficient to provide resistance in vitro . Deletion of acp did not render the bacteria more sensitive to lysozyme ( Fig 4C ) , which was in contrast to an earlier observation [9 , 26 , 27] perhaps due to differences between gonococcal isolates . It is plausible that a positive feedback loop exists between ACP and SliC that provides increased levels of ACP upon decreases in SliC expression . Supporting this conclusion , removal of both SliC and ACP rendered the bacteria remarkably sensitive to lysozyme , which was entirely complemented in the Δacp ΔsliC/P::acp P::sliC strain ( Fig 4C ) . Most importantly , SliC and its lysozyme inhibition function conferred a fitness advantage to N . gonorrhoeae in vivo , which we demonstrated in preliminary competition experiments with the ΔsliC mutant against wild type during co-infection in the murine female genital tract ( S3 Fig ) and in biological triplicate studies ( Fig 7B–7D ) , in which the ΔsliC/P::sliC* mutant strain , in addition to ΔsliC and the complemented strain , ΔsliC/P::sliC , were challenged against equal numbers of wild type bacteria . A drastic decrease in ΔsliC fitness in vivo was observed ( Fig 7B ) , which was further exacerbated for the ΔsliC/P::sliC* bacteria , which expresses a functionally inactivated SliC protein ( Fig 7D ) . This statistically significant drop in fitness was not caused by altered LOS ( Fig 4E ) , cell morphology ( Fig 4F ) , or other potential defects in the outer membrane integrity in cells lacking SliC , as no significant differences were observed between wild type and ΔsliC bacteria in sensitivity to seven antibacterial compounds with different modes of action ( S2 Table ) . Further , bacterial fitness could be fully restored by genetic complementation in the ΔsliC/P::sliC strain ( Fig 7C ) . Nor was the defect triggered by instability or affected transport of SliC* to the gonococcal cell surface , as the mutated protein was stably expressed from its native promoter ( Fig 2C ) and efficiently escorted to the gonococcal cell surface ( Fig 2F ) . No significant changes in the numbers of adherent and internalized bacteria were found in our in vitro studies ( Fig 5 ) eliminating SliC as an adhesin . Finally , our studies involving lysozyme defective mice ( Fig 7E and 7F ) further support SliC as a critical inhibitor of lysozyme during infection . Lysozyme is detected in the human cervico-vaginal fluid proteome and may be of either phagocytic or epithelial origin [35–38] . Taken together , our studies provide the first evidence that a novel lysozyme-binding lipoprotein of N . gonorrhoeae , SliC , plays a critical role during colonization of the host . Lysozyme is a major host defense factor against bacteria . Therefore , a detailed understanding of the mechanisms that bacteria have developed to fight lysozyme may allow development of new antimicrobial strategies against pathogens and also against certain microbiota during dysbiosis .
Neisseria utilized in this study include N . gonorrhoeae FA1090 [39] and 36 temporally and geographically diverse gonococcal clinical isolates with the 2016 WHO N . gonorrhoeae reference strains [16 , 40] , N . meningitidis MC58 [41] , N . lactamica NL1983/-01 , and N . weaveri 1032 [42] . Neisseria were removed from frozen stocks and plated on gonococcal base solid medium ( GCB , Difco ) and incubated in a 5% CO2 atmosphere at 37°C for 18–20 h . Transparent , non-piliated colonies were subcultured onto GCB . After passage on GCB , bacteria were cultured in gonococcal base liquid ( GCBL ) medium supplemented with Kellogg’s supplement I in 1:100 and 12 . 5 μM ferric nitrate [43 , 44] . To achieve iron-limited conditions , GCB without ferric nitrate and with deferoxamine mesylate salt ( Desferal , Sigma ) at 5 μM final concentration was utilized [45] . In addition , when stated in the text , GCB supplemented with 7 . 5% normal human serum [46] was used . For anaerobic growth , bacteria were plated on GCB with 1 . 2 μM nitrite as a terminal electron acceptor and harvested after 48 h [16] . Except where otherwise indicated , colonies were collected from plates with a polyester-tipped sterile applicator ( Puritan ) and suspended to an OD600 of 0 . 1 in GCBL supplemented with 0 . 042% sodium bicarbonate and supplements as above [43 , 44] . Liquid cultures were propagated at 37°C with shaking for 3 h , back-diluted to an OD600 of 0 . 1 in supplemented GCBL , and cultured in the same manner . Piliated colonies were used for transformation and adhesion/invasion assays while non-piliated variants were utilized in all other experiments . Escherichia coli strains were grown either on Luria-Bertani agar ( LBA , Difco ) or cultured in Luria-Bertani broth ( LB , Difco ) at 37°C . Antibiotics were used in the following concentrations: for N . gonorrhoeae: kanamycin 40 μg/mL , erythromycin 0 . 5 μg/mL , chloramphenicol 0 . 5 μg/mL , streptomycin 100 μg/mL; for E . coli: kanamycin 50 μg/mL , carbenicillin 50 μg/mL , chloramphenicol 34 μg/mL . Cloning procedures were performed in E . coli MC1061 [47] . N . gonorrhoeae FA1090 ( NC_002946 ) genomic DNA was isolated using the Wizard Genomic DNA Purification Kit ( Promega ) . Oligonucleotides were designed using SnapGene software version 2 . 8 ( GSL Biotech LLC ) and synthesized by Integrated DNA Technologies . Q5 High-Fidelity DNA polymerase , DNA ligase and NEBuilder HiFi DNA Assembly Master Mix were purchased from New England Biolabs ( NEB ) . Site-directed mutagenesis was performed using Q5 Site-Directed Mutagenesis Kit ( NEB ) . All obtained genetic constructs were verified by Sanger Sequencing at the Center for Genomic Research and Biocomputing at Oregon State University . Transformation of N . gonorrhoeae was performed as described previously [48] . To create an in-frame knockout of sliC , individual 500 bp fragments upstream ( sliC_up ) and downstream ( sliC_down ) of sliC were amplified using primer pairs 5’ACGTTGAGAATTCGCCGTCTGAGTCGGAATATGTCGGAGC3’ , 5’AGCGTACAGGTACCAGCGCGAAAAACCTGATA3’ , and 5’ACTCAATAGGATCCCGAAACTTCCTGCCGC3’ , 5’ACTCGGTCAAGCTTCATTGGATACCGACAATGAAAC3’ . The obtained sliC_up and sliC_down PCR products were digested with EcoRI/KpnI and BamHI/HindIII , respectively . Subsequently , sliC_up was cloned into similarly digested pUC18K , yielding pUC18K-upsliC and sliC_down was cloned into this genetic construct to yield pUC18K-ΔsliC . This plasmid was digested with ScaI and used for liquid transformation of N . gonorrhoeae FA1090 [48] . The sliC knockout was verified using primers 5'GGTTGGCGATGTAGAGGCT3’ and 5'GATTGCAGTTACAACGCGTGG3' and chromosomal DNA isolated from wild type FA1090 as controls , as well as immunoblotting analysis with anti-SliC antisera of the whole cell lysates obtained from wild type and ΔsliC . An in frame and markerless knockout of acp was achieved using Gibson assembly method [49] . Briefly , pNEB193 and DNA fragments comprising 1000 bp upstream and downstream from acp gene were amplified using primers: 5’GTTTAAACCTGCAGGCATGCAAG3’ , 5’TCTAGACTTAATTAAGGATCCGGCG3’ , 5’CGCCGGATCCTTAATTAAGTCTAGAACAAGCCGCTCAAAGAAGGTGACATTATCAACATC3’ , 5’ATTATTCAGACGGCATTTACGGCGGCCTGTCCGGTGT3’ , 5’GACAGGCCGCCGTAAATGCCGTCTGAATAATCAGGCAACAAAAAACAGCGTTTTCATTTC3’ , 5’AGCTTGCATGCCTGCAGGTTTAAACCGCCGATACCCGCAACCC5’ and Q5 polymerase . The obtained DNA fragments were purified and assembled using Hi-Fi Assembly Mix ( NEB ) to create pNEB-Δacp plasmid . This plasmid was linearized with ScaI and used for spot transformation of N . gonorrhoeae FA1090 . The FA1090 Δacp knockout was confirmed by PCR reaction using primers 5’GACCGGGATGAACCAGATAG3’ and 5’GGCTGATGCACCAATGCTTC3’ . The FA1090 Δacp knockout strain and pUC18K-ΔsliC were used to create the isogenic Δacp ΔsliC double mutant . Briefly , pUC18K-ΔsliC was cut with ScaI and used for spot transformation of N . gonorrhoeae FA1090 Δacp . The deletion of sliC in Δacp ΔsliC double knockout was confirmed by PCR with the primer pair: 5'GGTTGGCGATGTAGAGGCT3’ and 5'GATTGCAGTTACAACGCGTGG3' , as well as by probing whole cell lysates obtained from wild type and ΔacpΔsliC in immunoblotting analysis with anti-SliC antisera . Constructs for complementation of ΔsliC with the wild type sliC allele or sliC* ( carrying alanine substitutions in S83 and K103 ) and for production of recombinant proteins , rSliC and rSliC* , were designed as outlined in Fig 2A using the following steps . First , sliC was amplified with its native promoter using primers 5’GATCTTAATTAATTCAGACGGCATCGTCAGGC3’ and 5’GATCGTTTAAACTCAAACAGGCTGCCCCGT3’ . The resulting PCR product was digested with PacI/PmeI and cloned into similarly treated pNEB193 , yielding pNEB-sliC . To create sliC* , the S83A substitution was introduced with primers 5’CGTTGCCGCAGCTGGCGAACGCT3’ and 5’TCGGAAGAGAGGACGGCACG3’ using pNEB-sliC as template . The K103A mutation was generated in SliC S83A using primers 5’GTGGCACCAGGCTGGCGGCGAAG3’ and 5’TCGGTTCCGTTTCCGAAC3’ . To create constructs for complementation of ΔsliC , pNEB-sliC and pNEB-sliC* were treated with PacI/PmeI and DNA fragments corresponding to sliC and sliC* , respectively , were cloned into similarly digested pGCC5 [50] . The obtained plasmids , pGCC5-sliC and pGCC5-sliC* , were introduced to FA1090 ΔsliC . The presence of the integrated recombinant genes was confirmed in transformants by PCR , and the level of SliC expression was confirmed by immunoblotting with anti-SliC antisera . To complement the Δacp mutant , the acp gene with its native promoter was PCR amplified using primers 5’GATCGTCGACCCATATCGAATGCCTCGA3’ and 5’GATCTTAATTAAGGAACGGTCAAAAAACAGC3’ . The resulting DNA fragment was digested with PacI and cloned into PacI/PmeI digested pGCC5 plasmid to yield pGCC5-acp . This plasmid was introduced into FA1090 Δacp by transformation and the presence of the integrated acp was confirmed by PCR . To complement the FA1090 ΔacpΔsliC double knockout strain , pGCC5-acpsliC plasmid carrying both genes under control of their native promoters was constructed as follows: acp gene and 240 bp upstream from its start codon was amplified by PCR using the primer pair 5’GATCGTCGACCCATATCGAATGCCTCGA3’ and 5’GATCTTAATTAAGGAACGGTCAAAAAACAGC3’ . The resulting DNA fragment was digested with SalI/PacI and cloned into similarly cut pGCC5-sliC to yield pGCC5-acpsliC used for spot transformation of FA1090 ΔacpΔsliC . The presence of acp and sliC alleles in heterologous location was subsequently confirmed by PCR and their functionality was verified by immunoblotting and complementation studies . Constructs used for production of recombinant wild type SliC ( rSliC ) and SliC bearing S83A K103A substitutions ( rSliC* ) , both lacking a signal peptide and containing a C-terminal Tobacco Etch Virus ( TEV ) protease cleavage site followed by a 6×His-tag , were created by amplifying sliC and sliC* with primers 5’CGTAATGCCATGGTGCCGGAAGCGTATGATGGC3’ and 5’AGTCAGCAAGCTTACGGGCGCGGCAGG3’ using pNEB-sliC and pNEB-sliC* as templates , respectively . The obtained PCR products were digested with NcoI/HindIII and cloned into pRSF-NT [51] . Recombinant SliC and SliC* were purified from 3 L cultures of E . coli BL21 ( DE3 ) [52] carrying pRSF-rSliC and pRSF-rSliC* , respectively . Bacterial cultures were incubated at 37°C until OD600 of 0 . 7 and induced with 0 . 5 mM IPTG . The cultures were incubated overnight at 16°C and the cells were pelleted at 8 , 000 × g for 15 min . Cells were suspended in lysis buffer ( 20 mM HEPES pH 7 . 5 , 500 mM NaCl , 10 mM imidazole ) and lysed by passing four times through French Press at 1 , 500 psi . Cell debris was removed by centrifugation at 17 , 000 × g for 20 min and the cleared lysate was applied to a 5 mL IMAC column ( BioRad ) . Proteins were purified using a NGC Medium-Pressure Liquid Chromatography System ( BioRad ) using lysis buffer and elution buffer ( 20 mM HEPES pH 7 . 5 , 500 mM NaCl , 250 mM imidazole ) . Fractions containing either rSliC or rSliC* were pooled and the buffer was supplemented with DTT and EDTA to final concentrations of 1mM and 0 . 5 mM , respectively . The His-tag was removed by overnight incubation at 4°C with TEV protease in a 1:25 ratio . The samples were concentrated using a Vivaspin 20 centrifuge concentrator ( GE HealthCare ) and subjected to size exclusion chromatography using a HiLoad 16/600 Superdex 75 pg column ( GE HealthCare ) with phosphate buffered saline ( PBS ) as running buffer to separate SliC and SliC* from TEV protease . Protein purity was confirmed by SDS PAGE and the protein concentration was measured using the BioRad DC Protein Assay Kit . Polyclonal antisera against purified rSliC were prepared by Pacific Immunology Corp . using a 13-week antibody production protocol and two New Zealand White rabbits under Animal Protocol #1 approved by IACUC and the NIH Animal Welfare Assurance Program ( #A4182-01 ) in a certified animal facility ( USDA 93-R-283 ) . The EnzChek Lysozyme Assay kit ( ThermoFisher ) was used to determine the SliC-mediated inhibition of c-type lysozyme from chicken egg white ( HEWL , Sigma ) and human lysozyme ( HL , Sigma ) . Assays were conducted using 96 well plates and 100 μL volumes of 0 . 1 M sodium phosphate , 0 . 1 M NaCl , pH 7 . 5 and 2 mM sodium azide as buffer . Experimental samples containing 2 . 5 μM of either HEWL or HL were incubated with increasing concentrations of SliC ( 0–10 μM ) for 30 min at 37°C . The control wells contained HEWL or HL alone . After incubation , the reaction was initiated by addition of 50 μL of the 50 μg/mL DQ lysozyme substrate . The reaction was monitored for 1 h using a Synergy HT Microplate Reader ( BioTek ) at excitation and emission wavelengths of 485 nm and 530 nm , respectively . To calculate the IC50 , results were exported to GraphPad Prism 6 ( GraphPad Software ) . To test the inhibition of HL by SliC* , 2 . 5 μM of HL and 5 μM of the protein was used . Comparison of SliC inhibition of chicken and human lysozyme was assessed by incubation of 2 . 5 μM final lysozyme solution with increasing concentrations of SliC ( 0–10 μM ) . After mixing the proteins , the assay was performed as described above . The binding affinity of SliC to lysozyme was assessed by BLI on an OctetRed 96 ( ForteBio ) . The SliC protein was randomly biotinylated on surface-exposed lysine side chains with succinimidyl-6- ( biotinamido ) hexanoate , a biotin label with a medium chain length spacer arm , per manufacturer’s instructions ( EZ- Link NHS-LC-Biotin , Pierce ) . SliC was incubated with a 20-fold molar excess of biotin reagent 1 h at room temperature , and protein was subsequently applied to a PD10 column to remove excess biotin reagent . The SliC and HL samples were prepared in Kinetic Buffer ( 20 mM HEPES pH 8 , 15 mM NaCl , 0 . 002 Tween 20 , 0 . 1 mg/mL BSA ) . Streptavidin ( SA ) biosensors ( ForteBio ) were loaded with biotinylated SliC for 10 min at a 20 μg/mL concentration . Unloaded tips were used as a control . The baseline was established for 240 s and the association and dissociation steps were performed for 600 s . Experiments were performed in three biological replicates with curve fitting using a 2:1 ( Heterogeneous Ligand ) model . KD value calculations were completed using Octet Data Analysis ( version 9 ) . Subcellular fractionation procedures , dot-blots and protease shaving assays were conducted as described previously [16] . Briefly , extraction of proteins from the cytosolic , cell envelope , membrane vesicle , and soluble supernatant fractions was performed using 500 mL cultures of wild type N . gonorrhoeae FA1090 at mid-logarithmic phase of growth . Cell envelope proteins were separated from cytoplasmic fraction by a sodium carbonate extraction procedure and differential centrifugation , while culture supernatants were subjected to filtration and ultracentrifugation to separate naturally-released membrane vesicles from soluble proteins . After centrifugation , the soluble protein fraction was precipitated at 4°C for 1 h using 15% trichloroacetic acid . In experiments involving proteinase K shaving and immunodotting , intact bacterial cells were used [16] . For protease accessibility studies , N . gonorrhoeae FA1090 wild type was sub-cultured in GCBL for 3 h after collecting from solid media , diluted to OD600 of 0 . 1 and cultured until OD600 of ~ 1 . 0 was reached . Gonococci were gently centrifuged and suspended to OD600 of 2 . 0 in sterile PBS ( pH 8 . 0 ) supplemented with 5 mM MgCl2 . Bacterial suspensions ( 500 μL ) were incubated for 1 h at 37°C with proteinase K at final concentrations of 0 , 20 , or 40 μg/mL . To deactivate protease , 10 μL of 50 mM phenylmethylsulfonyl fluoride ( PMSF ) was added . Bacteria were then washed with GCBL , subjected to SDS-PAGE , and probed with polyclonal antisera to determine protease susceptibility . For immunodotting , N . gonorrhoeae strains were suspended in GCBL to an OD600 of 0 . 1 , cultured under standard growth conditions for 3 h , harvested , and spotted as 5 μL suspensions onto nitrocellulose membranes after adjusting the OD600 to 2 . 0 . The samples were dried at room temperature for 15 min and subjected to immunoblotting as described below . For assessing the expression of SliC during experimental murine infection , five mice were infected vaginally with strain FA1090 and vaginal washes were collected on days 1 , 3 and 5 post-bacterial inoculation by pipetting 40 μl of sterile PBS in and out of the vaginas . This procedure was repeated three times for each mouse , and the resultant fluids from each time point were pooled . Non piliated cells of wild type FA1090 , ΔsliC , ΔsliC complemented with sliC , Δacp , Δacp complemented with acp , ΔsliC Δacp , and ΔsliC Δacp complemented with sliC and acp were suspended in GCBL with Kellogg’s supplements and 0 . 042% sodium bicarbonate to OD600 of 0 . 1 and cultured at 37°C for 3 h . Bacteria were diluted to 1×104 CFU/mL and 170 μL of each culture was incubated for 8 h at 37°C with lactoferrin ( 5 mg/mL; InVitria ) and increasing concentrations of human lysozyme ( 0 , 1 , 5 , 10 and 20 mg/mL; BioVision ) in a final volume of 200 μL . Bacterial suspensions were plated for CFUs scoring . All experiments were performed in 5 biological replicates and means with corresponding SEMs are presented . To test if the presence of cell envelope permeating agents increases the lysozyme sensitivity of gonococci , Polymyxin B ( 100 U/mL ) , Tween 20 ( 0 . 001% ) , or bile salts ( 0 . 05% ) reconstituted in GCBL were added alone or in combination , as indicated , to bacterial cell suspensions ( 1×104 CFU/mL ) containing lactoferrin ( 5 mg/mL ) and lysozyme ( 5 mg/mL ) as described above . We established in preliminary studies that to obtain viable bacteria , cultures can be incubated at 37° in the presence of all these antimicrobials up to 4 h . Accordingly , after that time period mixtures were serially diluted and spotted on GCB to enumerate CFUs . All experiments were performed in 3 independent trials and means with corresponding SEMs are presented . The growth kinetics of wild type N . gonorrhoeae FA1090 and isogenic ΔsliC , ΔsliC/P::sliC , and ΔsliC/P::sliC* mutants were conducted in GCBL under standard growth conditions . Bacteria were collected from GCB and suspended in GCBL to an OD600 of 0 . 1 . Following 3 h of incubation at 37°C with aeration bacterial cultures were back-diluted to OD600 of 0 . 1 in fresh GCBL and cultured for additional 6 h . Samples were withdrawn for OD600 measurements and immunoblotting every hour ( n = 4; mean ±SEMs ) . To assess the viability of N . gonorrhoeae lacking sliC during host-relevant in vitro growth conditions , colonies of wild type FA1090 and ΔsliC were collected from GCB , suspended in GCBL to an OD600 of 0 . 1 , and cultured for 3 h at 37°C with aeration . Subsequently , the cultures were normalized to OD600 = 0 . 2 , serially diluted , and plated on solid media for standard growth conditions ( SGC ) , iron limiting conditions , NHS , and anaerobic conditions , as described above . The CFUs were scored after 22 and 48 h for aerobic and anaerobic conditions , respectively ( n = 3; mean ±SEMs ) . N . gonorrhoeae FA1090 wild type and isogenic ΔsliC , ΔsliC/P::sliC , Δacp , and ΔacpΔsliC were incubated in GCBL until strains reached mid-logarithmic growth . Cells were pelleted at 4 , 000 × g for 3 min and washed twice with PBS filtered through 0 . 1 μm filter . Bacteria were suspended in PBS and 2 . 5 μL of the cell suspensions were spotted onto 300 mesh copper grids . Cells were allowed to attach to the grid for 15 min and excess PBS was removed . Gonococci were negatively stained using phosphotungstic acid . Images were acquired using a FEI Helios NanoLab 650 electron microscope at the Oregon State University Electron Microscopy Facility . LOS was isolated from N . gonorrhoeae as described previously [53] . Bacteria were collected from GCB plates incubated at 37°C and 5% CO2 for 18 h and resuspended in GCBL to final OD600 of 0 . 2 . Cell suspensions ( 1 . 5 mL ) were collected and bacteria were lysed by addition of 50 μL of lysis buffer ( 2% SDS , 4% β-mercaptoethanol , 10% glycerol , 1M Tris-HCl pH 6 . 8 , and 0 . 01% bromophenol blue ) and incubation at 100°C for 10 min . Samples were cooled to room temperature and proteins were digested by addition of 25 μg proteinase K in 10 μL of lysis buffer for 1 h at 60°C . Isolated LOS was resolved on 18% SDS-PAGE gels and visualized by a silver staining procedure [54] . Human cervical epidermal carcinoma ( ME180 , ATCC HTB-33 ) cells were maintained in McCoy’s 5A Medium ( Iwakata and Grace Modified; Corning ) supplemented with 10% heat inactivated Fetal Bovine Serum ( FBS; Gemini Bio-Products ) . Epithelial cells were seeded in 24 well plates ( Greiner BioOne ) at 1×105 cells/well and incubated overnight at 37°C in 5% CO2 . N . gonorrhoeae strains were grown for 18 h . Bacteria were suspended in McCoy’s 5A Medium supplemented with 10% heat inactivated FBS and containing 0 . 5 × Kellogg’s supplements . Bacteria were added to the epithelial cells at a multiplicity of infection ( MOI ) of 10:1 for 1 , 2 , 3 , and 4 h . For adherence assays , cells were washed with PBS and then treated with 1% saponin for 15 min . Lysates were collected and aliquots plated on GCB . Results are expressed as the number of CFUs recovered ( n=6; mean ±SEMs ) . For invasion assays , cells were treated with 100 μg/mL of gentamicin for 2 h . After washing the cells extensively , the number of invasive gonococci was quantified by lysing the cells with 1% saponin for 15 min . Aliquots of the lysates were plated on GCB . Results are expressed as the number of CFUs recovered after gentamycin treatment ( n = 3; mean ±SEMs ) . Female BALB/c mice ( 6 to 8 weeks old; Charles River Laboratories Inc . , Wilmington , MA; NCI Frederick strain of inbred BALB/cAnNCr mice , strain code 555 ) were treated with water-soluble 17β-estradiol and antibiotics to increase susceptibility to N . gonorrhoeae [55] . Groups of mice were inoculated vaginally with similar numbers of wild type FA1090 and either isogenic ΔsliC , ΔsliC/P::sliC , or ΔsliC/P::sliC* bacteria ( total dose , 106 CFU N . gonorrhoeae; 7 mice/group ) . Vaginal swabs were collected on days 1 , 3 , and 5 post-inoculation and suspended in 100 μL GCBL . Vaginal swab suspensions and inocula were cultured quantitatively on GCB agar with streptomycin ( total number of CFUs ) and GCB with streptomycin and kanamycin ( ΔsliC , ΔsliC/P::sliC , or ΔsliC/P::sliC* CFU ) . Results were expressed as the competitive index ( CI ) using the equation CI = [mutant CFU ( output ) /wild-type CFU ( output ) ]/[mutant CFU ( input ) /wild-type CFU ( input ) ] . The limit of detection of 1 CFU was assigned for a strain that was not recovered from an infected mouse . A CI of <1 indicates that the mutant is less fit than the wild type strain . Preliminary experiments were performed in biological duplicates ( shown in S3 Fig ) . Final experiments involving competition assays with three pairs of strains , wild type and ΔsliC , wild type and ΔsliC/P::sliC , or wild type and ΔsliC/P::sliC* were conducted at the same time and on three separate occasions . Statistical analysis was performed using Kruskal-Wallis Dunn’s multiple comparison tests to compare significance of CIs between the ΔsliC/wild type and ΔsliC/P::sliC/wild type as well as ΔsliC/P::sliC*/wild type and ΔsliC/P::sliC/wild type . Competitive infections were also performed with mixtures of wild-type FA1090 and similar numbers of isogenic ΔsliC in C57BL/6J and B6 . 129P2-Lyz2tm1 ( cre ) Ifo/J 9 mice , which do not produce lysozyme ( Jackson Laboratories ) . Results from two experiments were combined . Statistical analysis was performed using the Mann-Whitney test to compare statistical significance of CIs between the ΔsliC/wild type in C57BL/6J and ΔsliC/wild type in lysozyme knockout mice . MICs for cefotaxime , azithromycin , tetracycline , polymyxin B , vancomycin , ampicillin , and benzylpenicillin were determined using E-tests ( Biomerieux ) according to the manufacturer’s recommendations . Each determination was performed on three separate occasions using fresh bacterial cultures and the consensus MICs obtained in at least two trials were reported . Samples of whole cell lysates , protein fractions , intact cells , purified proteins , or vaginal washes were normalized based on either OD600 , protein concentration , or CFUs as specified in the text . Protein concentration was measured using the DC Protein Assay . SDS-PAGE with loading controls for immunoblotting with differential protein abundances are provided in S4 Fig . Proteins were transferred onto nitrocellulose membranes using a Trans-blot Turbo ( Bio-Rad ) and detected by immunoblotting as described previously [16] . The immunoblotting analysis was performed using polyclonal rabbit antisera with the following dilutions: anti-SliC ( 1:20 , 000 ) , anti-BamA ( 1:20 , 000 ) [16] , anti-BamD ( 1:20 , 000 ) [16] , anti-SurA ( 1:10 , 000 ) or anti-Zwf ( 1:10 , 000 ) [56] and secondary goat anti-rabit HRP conjugated antibodies ( 1:10 , 000 ) ( Bio-Rad ) . SliC abundance during wild type FA1090 infection in female BALB/c mice was determined by densitometry using the volume tool built into Image Lab 5 . 0 software ( Bio-Rad ) as described in [42 , 57] . Rectangle tool , local background subtraction , and linear regression method were used in the calculations . The amount of SliC on day 1 post-infection was arbitrarily set to 1 and the protein abundance on days 3 and 5 is expressed relative to the SliC level detected on day 1 . Amino acid sequence identity of SliC homologs was assessed by aligning sequences downloaded from NCBI with the Clustal Omega online tool ( Clustal 2 . 1; https://www . ebi . ac . uk/Tools/msa/clustalo/ ) using the default alignment parameters . A subsequent phylogenetic analysis was performed in Molecular Evolutionary Genetics Analysis ( MEGA ) version 7 . 0 . 26 . A maximum likelihood tree was constructed using the Jones-Taylor-Thornton model [58] . Neighbor-Join and BioNJ algorithms were applied to a pairwise-distance matrix derived from the JTT model to obtain an initial tree for the heuristic search . The phylogenies were tested by 500 bootstrap iterations , and the tree with the highest log likelihood is presented . To analyze single nucleotide polymorphisms of sliC ( locus NEIP0196 ) , DNA sequences were compared between 44 , 028 isolates of all Neisseria spp . and 4990 isolates of N . gonorrhoeae deposited into the PubMLST database ( http://pubmlst . org/neisseria/ as of October 25 , 2017 and April 11 , 2018; respectively ) . GraphPad Prism’s build-in t-test was utilized to determine statistically significant differences between experimental results with the exception of animal studies . A confidence level of 95% was used for all analyses . Animal experiments were conducted at the Uniformed Services University of the Health Sciences ( USUHS ) according to the guidelines of the Association for the Assessment and Accreditation of Laboratory Animal Care under a protocol # MIC16-488 that was approved by the University’s Institutional Animal Care and Use Committee . The USUHS animal facilities meet the housing service and surgical standards set forth in the “Guide for the Care and Use of Laboratory Animals” NIH Publication No . 85–23 , and the USU Instruction No . 3203 , “Use and Care of Laboratory Animals” . Animals are maintained under the supervision of a full-time veterinarian . For all experiments , mice were humanely euthanized by the laboratory technician upon reaching the study endpoint using a compressed CO2 gas cylinder in LAM as per the Uniformed Services University ( USU ) euthanasia guidelines ( IACUC policy 13 ) , which follow those established by the 2013 American Veterinary Medical Association Panel on Euthanasia ( https://www . usuhs . edu/mps/facilities-resources ) .
|
Neisseria gonorrhoeae , the etiologic agent of gonorrhea , is a clinically important pathogen due to the emergence of multi-drug resistance and the lack of a vaccine ( s ) . During host colonization , pathogenic and commensal Neisseria inevitably encounter lysozyme , a major host innate defense factor that is abundantly present in epithelial secretions and phagocytic cells . Although Neisseria spp produce a c-type lysozyme inhibitor , the Adhesin Complex Protein , the significance of lysozyme inhibition for host colonization has not been addressed . Here we demonstrate the existence of a new c-type lysozyme inhibitor in Neisseria . We show that it is a surface-displayed lipoprotein in N . gonorrhoeae and , through its lysozyme-blocking function , plays a critical role in colonization of genital tract mucosae during infection in the female gonorrhea mouse model . We named the protein SliC as a surface-exposed lysozyme inhibitor of c-type lysozyme . Understanding the mechanisms underlying anti-lysozyme strategies may facilitate antimicrobial development .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"blood",
"serum",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"molecular",
"probe",
"techniques",
"enzymes",
"pathogens",
"microbiology",
"neisseria",
"gonorrhoeae",
"immunoblotting",
"enzymology",
"animal",
"models",
"membrane",
"proteins",
"model",
"organisms",
"outer",
"membrane",
"proteins",
"experimental",
"organism",
"systems",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"neisseria",
"neisseria",
"meningitidis",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"lipoproteins",
"mouse",
"models",
"molecular",
"biology",
"cell",
"membranes",
"biochemistry",
"blood",
"anatomy",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"proteases",
"immune",
"serum",
"organisms"
] |
2018
|
SliC is a surface-displayed lipoprotein that is required for the anti-lysozyme strategy during Neisseria gonorrhoeae infection
|
Specific members of complex microbiota can influence host phenotypes , depending on both the abiotic environment and the presence of other microorganisms . Therefore , it is challenging to define bacterial combinations that have predictable host phenotypic outputs . We demonstrate that plant–bacterium binary-association assays inform the design of small synthetic communities with predictable phenotypes in the host . Specifically , we constructed synthetic communities that modified phosphate accumulation in the shoot and induced phosphate starvation–responsive genes in a predictable fashion . We found that bacterial colonization of the plant is not a predictor of the plant phenotypes we analyzed . Finally , we demonstrated that characterizing a subset of all possible bacterial synthetic communities is sufficient to predict the outcome of untested bacterial consortia . Our results demonstrate that it is possible to infer causal relationships between microbiota membership and host phenotypes and to use these inferences to rationally design novel communities .
The composition of plant-associated microbial communities influences plant health and development [1][2] . This has raised interest in the use of microbes for biotechnology and agriculture [3][4] . However , it is challenging to measure the contribution of individual microbes from a complex microbiota to host health . Thus , a number of in vitro screening strategies are commonly applied to identify candidate plant-interacting microbes; however , none of the traits typically screened are correlated with a plant-beneficial outcome [5] . Another common prescreening strategy involves performing plant–bacterium binary-association assays [6] , but only a few have been successfully translated into agricultural settings [7][8][9] , suggesting that these assays also fail to capture critical aspects of nature’s complexity . Moreover , it is well established that microbial consortia can produce strong and unexpected effects on host health [10][11] , and such emergent properties are hard to predict , hindering the rational design of microbial consortia with desired host outputs . Previous strategies to address this conceptual problem included the exhaustive study of possible communities assembled from a small number of microbiota constituents in zebrafish [12] and the analysis of randomized combinations of bacteria in mice [13] . Other approaches often begin with an exhaustive evaluation of all combinations of , for example , the nutrients nitrate and ammonium and the hormones auxin , cytokinin , and abscisic acid on plant root growth and development [14] . Although exhaustive approaches can provide a complete picture of interactions within complex systems , they are unfeasible for systems with more than a handful of variables , given the astronomical number of possible factorial combinations . Even in the rare cases in which functional microbial consortia have been assembled , most studies focus on a single community that is considered a treatment , and rarely is an effort made to dissect the contribution of its constituents . This makes it impossible to establish predictable generalizations beyond the tested communities or conditions used . An instance that dissected the components of a consortium consisted of only 2 bacterial strains [15] . These findings reinforce the necessity for reduced complexity and modular model systems to associate microbial community composition with host phenotypes . Our approach is summarized in Fig 1 . In short , we first characterized the relationship between in vitro bacterial assays and plant–bacterium binary-association assays , and we used the latter to define functional bacterial blocks . These blocks are groups of bacteria that , by themselves , have a similar influence on a host phenotype . Then , we defined a subset of all the possible communities by constructing partially overlapping synthetic communities ( SynComs ) of 2 blocks each , tested the effect of these consortia on multiple plant phenotypes , and characterized the plant transcriptional response to these consortia . We evaluated the predictive performance of different statistical models on communities that the models had not seen before . We selected a neural network ( NN ) because it maximized predictive performance and used this model to design novel synthetic communities that maximized the change in 1 plant phenotype . Finally , we tested the model designs by constructing the novel communities it suggested and validated nearly all of the predicted host phenotypic outputs . We focus on bacterial manipulation of the plant response to phosphate ( Pi ) starvation , a commonly limiting nutrient for plant growth [16] . Pi is an essential macronutrient for plants and also for microbes [17][18] and is limited in soil [19] . Microbial communities living in the proximity of the plant take up Pi from the environment using a highly efficient Pi transport system [20][21] . Therefore , the available Pi in the close vicinity of plants is subject to direct and intense competition for uptake between microbes and plants [18] . Although the response of Arabidopsis thaliana seedlings in axenic conditions to phosphate starvation is well characterized [22] , the elucidation of the regulatory mechanisms of this response in the presence of the plant microbiome is only recently emerging [23] [24] . We systematically evaluated the performance of a large collection of root bacterial isolates using in vitro screening and binary plant–bacterium association assays as predictors for the effect of derived bacterial consortia on plant phenotypes in response to phosphate starvation . We confirm that bacterial in vitro assays have no correlation with bacterial effects on plant phenotypes . However , we found that plant–bacterium binary-association assays are informative for designing small synthetic communities . Surprisingly , the effects of bacterial consortia on host physiology were mostly additive and independent of bacterial abundances , suggesting that functional stacking within a microbial consortium can determine its effect on host phenotypic response . Finally , we successfully validated novel synthetic communities designed by an NN that led to predictable changes in plant shoot Pi content . Our results provide a useful road map from binary host–microbe assays to the design and testing of useful small consortia to predictably alter host phenotypes .
In response to Pi deficiency , plants change root exudate metabolite profiles and root architecture to explore Pi-rich soil patches [25] . This may lead to bacterial soil community shifts [26] . In order to learn how root exudate profiles change in response to Pi , we harvested root exudates from A . thaliana plants in response to 2 short and complementary nutritional transitions that mimic the dynamics of Pi stress [27] ( S1A Fig; Materials and methods 1a , 1b , 1d , 1e ) . We demonstrated that our Pi transitions were sufficient to induce a reconfiguration of plant exudate primary metabolic profiles ( S1C and S1D Fig , and S1 Table ) . We next tested whether these exudates modified the in vitro growth capacity of a collection of 440 bacterial strains isolated from the roots of Brassicaceae grown in soil that is not overtly Pi deficient ( nearly all from A . thaliana ) ( [28] , S1B Fig , and Materials and methods 1c ) . We identified a range of bacterial growth behaviors ( Materials and methods 1f , 1g ) and found that the bacterial growth differences between phosphate conditions are much weaker than the differences between strains ( S2 Fig ) . As expected , phylogeny explained most of the growth differences between strains ( S2A Fig and Materials and methods 1h ) . Most of the bacterial growth parameters provided the same information , so we selected the area under the growth curve ( optical density [OD] versus time ) ( AUC ) as a growth marker for subsequent analyses . Hierarchical clustering of AUC differences between in vitro conditions identified 10 groups of bacteria that represented different response patterns to exudates derived from roots grown in different Pi concentrations and media supplemented or not with Pi ( S2B Fig and S2 Table ) . We found that root exudates could enhance or inhibit bacterial growth and that this effect could be either general or specific to one type of exudate ( S2B Fig ) . Thus , consistent with previous findings [26] , plant-derived root exudates modulated the growth of bacterial root isolates depending on the plant’s Pi starvation status . We selected a subset ( n = 183 ) of the strains from the in vitro assays for determining whether they exerted a functional role on the plant under different phosphate conditions . We selected bacterial isolates that belonged to all of the different response patterns ( S2B Fig ) and that were most responsive to both Pi levels and the presence of exudates ( Materials and methods 1g , S2 Table ) . We measured the change in plant shoot Pi content , a direct marker of phosphate starvation responsiveness [22] , in response to the presence of each of 183 individual strains , when compared to axenically grown plants . We evaluated shoot Pi content under 4 Pi conditions that represented a 2 × 2 design matrix of 2 Pi levels used for plant germination ( full; 1 mM; and depleted , about 5 μM Pi ) and 2 Pi concentrations ( 30 μM Pi and 100 μM Pi ) to which seedlings were switched , concomitant with the application of each bacterial strain ( Fig 2A and Materials and methods 2 ) . The use of 2 germination conditions in the experimental design allowed us to evaluate the effect of the activation of the phosphate starvation response and the shoot Pi content on the plant–bacterium interaction under different Pi concentrations . On average , bacteria had a slightly negative effect on plant shoot Pi content , visualized as a small tail in the bacterial treatment graphs ( pink ) in Fig 2B . This effect was stronger when the environmental Pi concentration was lower ( Fig 2B and 2C and S3 Table ) . These findings are consistent with our previous results that a bacterial synthetic community drives a context-dependent competition with the plant for Pi [23] . Overall , we found that more strains had a negative than a positive effect on shoot Pi content ( Fig 2B , S3 Table , and Materials and methods 2e ) . Specifically , there were significantly more strains that had a stronger negative effect on plant shoot Pi content in the most limiting Pi conditions ( germination in Pi depleted , followed by transfer to 30 μM Pi ) ( Fig 2B and S3 Table ) , in which the phosphate starvation response should be active . Conversely , the least Pi-deprived condition ( germination in full Pi , followed by transfer to 100 μM Pi ) exhibited a significant enrichment of strains that positively affected shoot Pi content ( Fig 2B and S3 Table ) . These results are consistent with bacterial effects on Pi content in the shoot being modulated by the nutritional status of the plant . Importantly , germination conditions did not alter bacterial colonization ( Fig 2D ) , and the effect of individual strains on plant shoot Pi content was independent of the ability of root-inoculated bacteria to colonize the shoot and independent of bacterial titers in different plant organs ( Fig 2D; S3A Fig and Materials and methods 2c ) . We detected a weak phylogenetic signal in the ability of bacterial strains to modulate plant Pi content that was significant in only 2 of the 4 conditions ( Fig 2C and Materials and methods 2f ) . Accordingly , we found no correlation between the effect of individual bacterial isolates on shoot Pi content and their in vitro growth phenotype in response to switched Pi levels and root exudates ( S4 Fig ) . Overall , our survey of plant–bacteria binary associations and the resulting distribution of bacterial effects on shoot Pi content argue that the majority of plant–bacteria interactions are competitive , at least in the context of phosphate starvation response . We recently demonstrated that the A . thaliana phosphate starvation response is largely antagonistic to immune system function [23] . We therefore asked whether activation of the plant phosphate starvation response could modulate the outcome of binary bacteria–plant interactions . We analyzed shoot Pi content in plants pretreated with phosphite ( Phi ) ( KH2PO3 ) and then transferred to either 30 μM Pi or 100 μM Pi in the presence of each of 30 selected bacterial strains that either reduced , increased , or had no effect on the shoot Pi content ( 10 strains per class; S3E Fig ) . Phi is a nonmetabolizable analog of Pi and its accumulation delays the phosphate starvation response , resulting in low accumulation of Pi in the shoot [29] ( S3B and S3C Fig ) . We found that germinating plants on Phi ( low shoot Pi , phosphate starvation response off ) dramatically reduced the number of bacterial isolates that diminished shoot Pi content , compared to germination on low Pi ( low shoot Pi , phosphate starvation response on ) ( S3D and S3E Fig ) . Additionally , we observed that under Phi pretreatment , none of the strains significantly increased shoot Pi content compared to germination on high Pi ( high shoot Pi , phosphate starvation response off ) ( S3D and S3E Fig ) . Importantly , Phi treatment did not alter bacterial colonization ( Fig 2D , S3A Fig and Materials and methods 2c ) . These findings indicate that activation of the plant phosphate starvation response results in different modes of bacterial interactions with the plant that are independent of shoot phosphate content . These results indicate that an active phosphate starvation response can modulate the outcome of both positive and negative interactions with bacteria , likely mediated via coregulation of the plant immune system . An analogous mechanism has been described for the interaction between A . thaliana and a beneficial fungus [24] . We sought to establish whether the results from binary associations are indicative of bacterial effects when a more complex bacterial community is present . We used a microcosm reconstitution approach , in which we inoculated plants with defined complex bacterial synthetic communities ( Materials and methods 3a ) . A subset of 78 strains analyzed in the binary-association experiments was grouped into 3 functional groups consisting of positive ( P1-P3 ) , indifferent ( I1-I3 ) , and negative ( N1-N3 ) bacteria , depending on their effect on shoot Pi accumulation . For the positive and negative groups , we focused on strains that had a statistically significant effect on plant Pi accumulation , after correcting for multiple testing ( Materials and methods 3i ) . Each functional group was further divided into another 3 blocks of 8–9 bacterial strains , according to the magnitude of their individual effects ( Fig 3A , S3 and S4 Tables , and Materials and methods 3i ) . We then combined pairs of these blocks to define 14 partially overlapping bacterial synthetic communities ( Fig 3B ) . This scheme was designed to maximize the probability of observing extreme plant phenotypes by stacking functionally similar blocks and to gain information from combining the most extreme phenotypic blocks defined in the binary-association assays . We evaluated shoot Pi content , primary root elongation , shoot size , and total root network in A . thaliana plants grown in association with the 14 bacterial synthetic communities in the same growth conditions used for the binary-association analysis ( Fig 2A and Materials and methods 3b ) . We found that synthetic communities , like individual bacterial strains , were more likely to reduce plant shoot Pi content , and that synthetic communities made of negative blocks led to lower shoot Pi accumulation than those composed of positive blocks ( Fig 3C and 3D and S5 Fig ) . For example , in general , the estimated negative effect on shoot Pi accumulation for negative blocks is significantly larger than for positive or indifferent blocks ( Fig 3C and 3D , S5a Table , and Materials and methods 3j ) . At the synthetic community level , the effect of the negative strains was clearly dominant; only 2 communities containing negative blocks ( I3N1 and N2N3 ) showed a nonsignificant reduction in shoot Pi content and this in only 1 of the tested conditions . Importantly , the only significantly positive effect with respect to no bacteria involved 2 positive blocks and was weak ( P1P3 ) ( Fig 3C , S5b Table , and Materials and methods 3j ) . We also observed that the majority of the cases in which a synthetic community did not significantly reduce the shoot Pi accumulation occurred under the less Pi-restricted condition ( 100 μM ) ( Fig 3C , S5b Table , and Materials and methods 3j ) , consistent with the results from the individual strains ( Fig 2B ) . This trend was generally consistent for the other plant phenotypes analyzed ( Fig 3C and 3D and S5 Fig ) . Overall , the reduction in shoot Pi content associated with negative blocks correlated with less shoot area , shorter primary roots , and bigger root networks ( top and bottom rows in Fig 3C and 3D , and S5 Fig ) , morphological changes that match the canonical phosphate starvation response in axenic conditions [22][30] . In contrast , positive bacterial blocks caused less intense plant phosphate starvation response phenotypes . These effects were more obvious in plants grown at low environmental Pi concentration ( Fig 3C and 3D and S5 Fig ) . Thus , the binary-association assays were generally informative with regard to the behavior of bacteria in more complex biotic backgrounds . Interestingly , we observed that a number of synthetic communities , for example P2P3 and P1P2 , led to increased shoot area compared to axenically grown plants , despite exhibiting reduced shoot Pi content ( Fig 3C and S5 Fig ) . In contrast , plants treated with P1P3 in +Pi_100 μM Pi condition , had shoot Pi content similar to Pi-sufficient plants but unexpectedly exhibited a reduced shoot area ( Fig 3C and S5 Fig ) . Thus , bacterial consortia can decouple shoot Pi-content accumulation from the growth inhibition responses typically associated with the canonical phosphate starvation response [22][30][31] . We estimated the common ( additive ) effects of each block of strains across different bacterial backgrounds ( e . g . , in different synthetic communities ) ( Materials and methods 3j ) . Surprisingly , we found that additive contributions of the bacterial functional blocks are sufficient to explain most of the plant phenotypic variation observed ( Fig 4 ) . We found that synthetic community membership ( i . e . , ignoring bacterial relative abundances ) typically explained more than 50% of the plant phenotypic variance ( Fig 4 ) . This indicates that intrablock bacterial interactions contribute at least as much as interblock interactions to the plant phenotypes tested . Furthermore , the effect of bacterial blocks on the phenotypes analyzed is generally consistent across different synthetic communities , despite each strain’s relative abundance being dependent on the microbial context ( S6 and S7 Figs ) . We found that the bacterial abundances in either agar substrate or in the root endophytic compartment were poorly correlated with plant phenotypes ( Materials and methods 3e , 3k ) . Despite the consistent taxonomic profiles of the inoculum , we observed that bacterial communities of agar and root samples were dominated by variable bacterial taxa , depending on the specific combination of bacterial blocks present ( S6A and S6B Fig ) . This suggests that bacteria–bacteria interactions are important in shaping the final community . Furthermore , we found clear taxonomic differences between root and agar samples . Most notably , Streptomyces strains ( order Actinomycetales ) were particularly good root colonizers despite their limited success on agar , while Pseudomonadales strains were relatively more successful in agar than in root samples ( S6A and S6B Fig ) . These results recapitulate previous findings in natural soils , indicating that Actinobacteria are enriched in A . thaliana roots [32][33] . Phosphate concentrations in the media had only a minimal effect on the final community composition ( S6B Fig ) . We then quantified the information gained by incorporating relative abundance data ( S6 Fig ) into our additive model ( Materials and methods 3k ) . Surprisingly , in all cases ( 16/16 ) the plant phenotypic variance explained by microbiota composition decreased when we incorporated relative abundance ( S7 Fig ) . While in some cases , the differences might not be statistically significant , together , this result demonstrates significantly better performance by the model that ignores relative abundance ( p-value = 0 . 000481; 2-tailed Wilcoxon signed-rank test ) . Our results indicate that bacterial blocks disproportionately modulate shoot Pi content with respect to their strain abundances , an observation analogous to that seen in bacteria modulating zebrafish immune responses [12] . The synthetic communities differentially modulated plant phenotypes related to phosphate starvation response . Therefore , we examined the transcriptomes of plants growing with different synthetic communities . We first explored the expression of a literature-based core set of 193 phosphate starvation response transcriptional markers [23] . Plants did not exhibit induction of phosphate starvation response markers in axenic conditions , even when Pi was low ( Fig 5A ) [23] . However , some synthetic communities induced the canonical transcriptional response to Pi starvation in plants grown on 30 μM Pi ( Fig 5A ) . Plants that showed transcriptional activation of the phosphate starvation response displayed lower shoot Pi accumulation . However , we also observed that some synthetic community treatments lead to low shoot Pi content and no activation of the transcriptional phosphate starvation response ( S8 Fig ) . The effect of synthetic communities was in general dependent on the presence of negative bacterial strain blocks ( Fig 5A ) . In contrast , synthetic communities consisting of only positive blocks of bacteria did not induce the phosphate starvation response transcriptional signature in any condition analyzed ( Fig 5A ) . No induction of the phosphate starvation response genes was observed when the Pi stress was released ( following transfer to 100 μM Pi ) except for the bacterial combination P3N3 , which exhibited induction on 100 μM Pi ( Fig 5A ) . In accordance with the shoot Pi content data ( Figs 3C , 3D and 4 ) , we found that additive effects of bacterial blocks could explain the level of transcriptional induction ( Fig 5B ) . The specificity in the bacterial modulation of plant phenotypes suggests that the changes observed in the plant in response to the synthetic communities are linked to bacterial block activities . We next explored the overall plant genome-wide transcriptional response to bacteria consortia , Pi conditions , or both . Our design allowed us to test both the response to synthetic communities and to individual bacterial blocks between and within conditions ( Materials and methods 3h ) . As anticipated , plants growing with bacterial synthetic communities on low Pi generally induced phosphate starvation responsive genes and modified the expression of immune system–related genes ( S9 Fig , S6 and S7 Tables ) [23] . Overall , there was not a common response to bacterial presence , with only 45 and 35 genes being significantly up- or down-regulated by more than half of the bacterial blocks , respectively ( S6 Table ) . The number of genes differentially expressed in response to different bacterial blocks did not correspond with the strain composition of the blocks; blocks P2 , N3 , and I1 altered the expression of the most genes , and blocks I3 , N1 , and P3 influenced the least ( S6 Table ) . In particular , block I3 only altered the expression of 17 genes , despite being detected in plant roots and surrounding agar ( S6 Fig ) . At the functional level , most of the bacterial blocks induced the expression of the plant defense response , specifically up-regulating genes for salicylic acid biosynthesis ( S10 Fig ) , consistent with overall Bacteria versus No Bacteria comparisons ( S9C Fig ) . We also investigated differences between the genes induced by different bacterial blocks . Comparison of genes differentially expressed between positive and negative blocks across all conditions showed that positive blocks had higher expression of genes involved in energy production , while negative blocks specifically induced abiotic stress–responsive gene sets , specifically abscisic acid–related genes ( Fig 6A , S6 and S7 Tables ) . Negative blocks of bacteria also increased the expression of a specific sector of the jasmonic acid response involved in glucosinolate biosynthesis ( Fig 6A–6C , S6 and S7 Tables ) . The glucosinolate pathway modulates the interaction of A . thaliana with a beneficial fungus at low Pi [24] , and its expression is regulated by the master regulator of phosphate stress response , PHR1 ( PHOSPHATE STARVATION RESPONSE1 ) [23] . When the environmental Pi was low ( 30 μM Pi ) , we observed many more differentially expressed genes between positive and negative blocks ( Fig 6A ) , with negative blocks driving higher expression of genes of both the phosphate starvation and defense response . We then focused our analysis on the Pi-limiting conditions ( 30 μM ) . In this condition , synthetic communities containing negative blocks showed a strong induction of the phosphate starvation response ( Fig 5A ) . We asked whether the different negative blocks ( N1 , N2 , and N3 ) differed in their effects . There were almost no expression differences between the 2 most negative blocks ( N2 and N3 ) , but we identified 103 genes differentially regulated by bacterial blocks N1 and N3 . These genes were mostly stress-related genes , including general abiotic stress and defense response , the expression of which was comparatively reduced in the phenotypically more negative block N3 ( Fig 6D , S6 and S7 Tables ) . This result indicates that under phosphate starvation , all negative blocks activate a similar set of phosphate starvation response genes but differentially suppress other stress responses . We found that some genes were induced in response to specific block combinations . For example , we found that PHOSPHATE2 ( PHO2 ) , a ubiquitin-conjugating E2 enzyme in A . thaliana required for the degradation of Pi transporters at high Pi [34] , is highly expressed only in plants exposed to the synthetic community P3N3 in all Pi conditions analyzed ( Fig 6E ) . This finding may explain the strong transcriptional response to Pi starvation caused by this synthetic community ( Fig 5A ) . The auxin-regulated gene AUXIN-REGULATED GENE INVOLVED IN ORGAN SIZE ( ARGOS ) [35] showed a weak positive correlation with the induction of the phosphate starvation response , and it was induced in plants grown with the synthetic communities P3N3 and N2N3 ( Fig 6F ) . ARGOS controls organ size in A . thaliana and its transgenic expression results in enlarged aerial organs [35] . This could serve to counterbalance the negative effect on shoot size that low Pi typically causes . Our design of synthetic communities emphasized placing every bacterial functional block into at least 2 microbial backgrounds; therefore , we should be able to estimate bacterial effects that are independent of background . In principle , this estimation could be used to design novel synthetic communities with predictable outputs . We found that additivity of bacterial effects could explain most , but not all , of the host phenotypic variation . Therefore , we built 3 different quantitative predictive models capable of capturing different levels of complexity and evaluated their performance . We constructed a simple linear model ( LM ) , a linear model that included pairwise interactions between bacterial functional blocks ( INT ) , and a Neural Network model ( Fig 7A and Materials and methods 4 ) . We focused on shoot Pi content , which had the strongest signal-to-noise ratio ( SNR ) of all plant phenotypes tested ( S11 Fig , Materials and methods 4b ) . To evaluate the predictive performance of each model , we used a form of cross validation in which the data from each synthetic community were held out one at a time , and the remaining synthetic communities were used to train each of the 3 models; those trained models were then used to predict the plant phenotypic output of the held-out consortium . We found that the NN had the lowest cross-validated prediction error of the 3 models and that the difference was statistically significant ( p-value = 0 . 0073 ) ( Fig 7B ) . Neural Networks are popular predictive models because they can capture more complex and nuanced relationships that simpler ( linear ) models cannot; however , this can come at a cost of reduced interpretability . We performed a sensitivity analysis ( Materials and methods 4f ) by calculating the effect that changing each variable would have on shoot Pi content according to the NN and the 2 linear models ( LM and INT ) . We found a general agreement between the 3 models; for example , all models showed that Pi level in the media and the presence of negative bacterial blocks had the strongest effect on shoot Pi content , but the NN produces much more fine-grained results , because it is able to predict the change differentially across each condition ( Fig 7C ) . In order to validate the prediction accuracy of the NN , we chose the 25 bacterial block replacements that were predicted to result in the largest increase in shoot Pi content and experimentally tested whether an increase was produced by these synthetic communities ( Fig 7D , S8 Table , and Materials and methods 4g ) . There was a significant correlation ( ρ = 0 . 42 , p-value = 0 . 0375 ) between predicted and observed shoot Pi content changes caused by the bacterial block replacements ( Fig 7E ) . Strikingly , we found that 23 out of 25 bacterial block replacements increased shoot Pi content on average ( p-value = 0 . 004; 1 , 000 permutation tests with synthetic community labels randomly permuted ) ( S9 Table ) . Moreover , the improvement in shoot Pi content was statistically significant in 16 out of 25 bacterial block replacements ( p-value = 0 . 032; 1 , 000 permutation tests with synthetic community labels randomly permuted ) ( S9 Table ) . Only 1 out of 25 bacterial block replacements significantly decreased Pi content ( S9 Table ) . Again , we noted little correlation between bacterial abundances and their effect on Pi content ( S12 Fig and S9 Table ) . Compared to linear models ( LM and INT ) , the NN had significantly lower prediction errors ( p-value ≤ 4 . 65 × 10−7 ) ( Fig 7F ) . In summary , we were able to rationally design novel synthetic communities that lead to predictable plant phenotypic outputs .
While plant responses to stress have been shown to be influenced by associated microbial communities , causal relationships in plant–microbe interactions in a community context and measured phenotypes have proven difficult to establish . This limitation is , in fact , generally true across complex host–microbial interaction systems [12][13][36] . Here , we demonstrate that binary-association assays can inform the design of synthetic bacterial communities that lead to predictable plant phenotypes , an observation seen only once previously , in one animal system [12] . The host phenotypic output of the bacterial synthetic communities was consistent with the output expected from binary interactions . Validation of predictions from an NN confirmed that we could predictably alter certain plant phenotypes by changing the plant’s microbiota membership . Other phenotypes and host–microbiota systems can likely be studied with this approach . The only requirements are that a microcosm reconstitution system is available and that functional bacterial blocks can be defined , so that synthetic communities that maximize the expected range of phenotypic variance can be constructed ( Fig 3A and 3B ) . In practice , other aspects that are likely to influence the tractability of a system are the functional bacterial diversity and the SNR of the phenotypes being measured . In the case of plant phosphate starvation , we found that bacterial abundances provided no information , and while it is too early to say if this is a general feature , the only other work that directly manipulated a well-defined microbiota to establish its effect on a host phenotype reached a similar conclusion [12] . If this trend continues across other host–microbiota systems , then our approach has the added advantage that strains need not be distinguishable by a specific marker gene . While a simple additive model typically explained more than 50% of the host phenotypic variation , we found value in utilizing an NN approach that was able to capture more complex relationships but remained interpretable and significantly increased our prediction accuracy for novel communities . Our framework is based on empirical validation and thus remains flexible enough to allow for simpler or more complex models , depending on the case . We achieved high prediction accuracy across an untested set of synthetic communities , thus demonstrating that selecting a subset of the possible communities by partial overlap of bacterial functional block pairs is sufficient to characterize this system . This method requires no design of specific heuristics . Thus , this methodology provides an opportunity to expand the capacity for mechanistic understanding not only of biological networks that control plant phenotypes [37][14] but of other complex ecological systems [12][13][36] . Furthermore , by focusing on block replacements as testable hypotheses , we provide a simple outcome that can be extracted from both linear and nonlinear predictive models . This can guide the next set of experimental designs , thus providing nonlinear methods like deep learning a stronger empirical grounding , rendering them less of a “black-box . ” We demonstrate the utility of our approach by defining mechanistic aspects of the plant phosphate stress response in the presence of combinations of bacterial blocks . We observed that bacteria range in their effect on phosphate content in the plant between severely decreasing and moderately enhancing it . These data are consistent with our previous findings that bacterial interactions with the plant are controlled by negative regulation exerted by the phosphate starvation response on the plant immune system [23] . A similar mechanism was described for the interaction under phosphate-limiting conditions of A . thaliana with the beneficial fungus Colletotrichum tofieldiae [24] . Thus , our results provide additional evidence for mechanisms by which plants and bacteria compete in times of nutritional stress . The use of multiple bacterial synthetic communities led us to define interesting particular aspects of the phosphate stress response . We observed that certain synthetic communities , such as P2P3 and P1P2 , drive an increase in the shoot area compared to axenically grown plants , despite the low shoot Pi content that they engender . These data recapitulate a previous observation [31] on the effect of altering the activity of PHOSPHATE1 ( PHO1 ) , a gene required for Pi loading into the xylem [38] . These authors found that shoot Pi content could be uncoupled from the developmental responses typically linked to Pi scarcity . We corroborated that reduced shoot growth is not necessarily a direct consequence of Pi limitation . The observations that both bacterial activity and the modulation of PHO1 expression can uncouple plant phenotypes during the response to low Pi leads us to hypothesize that microbes could interdict PHO1 transport activity , thus modifying Pi translocation from roots to shoots . Additionally , we found that synthetic community P3N3 uniquely induced a strong transcriptional response to phosphate starvation in the majority of the conditions tested . Plants exposed to this bacterial combination showed a high-level induction of PHO2 , a ubiquitin-conjugating E2 enzyme required for the degradation of Pi transporters at high Pi [34] . This discovery may explain the intense transcriptional response to Pi starvation caused by this particular synthetic community . We observed much more variability in the bacterial colonization patterns than in their effects on plant phenotypes . Synthetic communities tended to be dominated by 1 block , but the identity of that block did not correlate with plant phenotype . On the other hand , synthetic communities had remarkably consistent effects on plant phenotypes , and synthetic community membership was sufficient to predict host phenotype . These observations suggest that bacteria–bacteria interactions are critical for microbial community assembly , which is probably a highly dynamic process in which the microbial background determines which bacteria perform well . On the other hand , the effect of bacteria on plant phenotypes is probably due to functional stacking , in which many phenotypically redundant strains with potentially different niches maximize the chance of attaining the desired host phenotypic output . This “lottery model” has been proposed as a major driving mechanism of host colonization by its microbiota at the taxonomic level [39] , and it would be interesting to test whether a similar process governs functional assembly . In conclusion , we provide a general method for the study of various biological host–microbiome systems through rational selection of a tractable subset of the possible combinations of bacteria from a defined collection . We demonstrate that complex relationships among host phenotypes , the microbiota , and the abiotic environment can be captured using deep learning techniques . By testing each block of bacterial strains in multiple synthetic communities , and successfully validating predictions derived from an NN , we demonstrated that it is possible to both infer causality and attain generality when it comes to predicting host phenotypes in this complex system . This approach contributes to the rational design and deployment of microbes to improve responses of hosts to biotic and nutritional stresses .
|
Symbiotic microbes influence host development and health , but predicting which microbes or groups of microbes will have a helpful or harmful effect is a major challenge in microbiome research . In this article , we describe a new method to design and predict bacterial communities that alter the plant host response to phosphate starvation . The method uses plant–bacterium binary-association assays to define groups of bacteria that elicit similar effects on the host plant . By constructing partially overlapping bacterial communities , we demonstrated that it is possible to modify phosphate accumulation in the plant shoot and the induction of plant phosphate starvation genes in a controlled manner . We found that bacterial colonization of the plant root does not predict the capacity to produce this phenotype . We evaluated the predictive performance of different statistical models and identified one best able to predict the behavior of untested communities . Our work demonstrates that studying a subset of all possible bacterial communities is sufficient to anticipate the outcome of novel bacterial combinations , and we establish that it is possible to deduce causality between microbiome composition and host phenotypes in complex systems .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion"
] |
[
"chemical",
"compounds",
"phosphates",
"plant",
"growth",
"and",
"development",
"brassica",
"developmental",
"biology",
"plant",
"science",
"model",
"organisms",
"metabolites",
"experimental",
"organism",
"systems",
"genome",
"analysis",
"seedlings",
"plants",
"bacteria",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"genomics",
"gene",
"expression",
"chemistry",
"gene",
"ontologies",
"biochemistry",
"eukaryota",
"plant",
"and",
"algal",
"models",
"root",
"growth",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"metabolism",
"organisms"
] |
2018
|
Design of synthetic bacterial communities for predictable plant phenotypes
|
Experimental evidence suggests that a tetramer of integrase ( IN ) is the protagonist of the concerted strand transfer reaction , whereby both ends of retroviral DNA are inserted into a host cell chromosome . Herein we present two crystal structures containing the N-terminal and the catalytic core domains of maedi-visna virus IN in complex with the IN binding domain of the common lentiviral integration co-factor LEDGF . The structures reveal that the dimer-of-dimers architecture of the IN tetramer is stabilized by swapping N-terminal domains between the inner pair of monomers poised to execute catalytic function . Comparison of four independent IN tetramers in our crystal structures elucidate the basis for the closure of the highly flexible dimer-dimer interface , allowing us to model how a pair of active sites become situated for concerted integration . Using a range of complementary approaches , we demonstrate that the dimer-dimer interface is essential for HIV-1 IN tetramerization , concerted integration in vitro , and virus infectivity . Our structures moreover highlight adaptable changes at the interfaces of individual IN dimers that allow divergent lentiviruses to utilize a highly-conserved , common integration co-factor .
To establish productive infection , a retrovirus must insert the reverse-transcribed form of its genome into a host cell chromosome . This process critically depends on two reactions , 3′-processing and strand transfer , catalyzed by the viral enzyme integrase ( IN ) ( reviewed in [1] ) . During 3′-procesing , IN endonucleolytically removes two or three nucleotides from the 3′-termini of viral DNA to expose 3′-OH groups of invariant CA dinucleotides . These are subsequently utilized in a pair of coordinated transesterification reactions , resulting in the insertion of both viral DNA termini across the major groove of chromosomal DNA . Integration is completed through the action of host DNA repair enzymes , which mediate the necessary joining of viral DNA 5′-ends , yielding a short duplication of target DNA sequence flanking the integrated provirus . Retroviral INs have a characteristic three-domain organization , all containing N-terminal , catalytic core and C-terminal domains ( NTD , CCD , CTD ) ( reviewed in [2] ) . The CCD contains the invariant D , D-35-E motif responsible for coordination of two Mg2+ ions within the active site and accounts for sequence-specific interactions with viral DNA [3] , [4] . The positively-charged CTD is also implicated in DNA binding , likely accounting for sequence-independent interactions [5] . All three domains contribute to IN multimerization [6]–[8] . CCDs of divergent retroviral INs invariably crystallize as dimers , with isomorphous dimer interfaces [9]–[11] . Structures of the NTD and CTD have been solved both alone and as part of two-domain constructs involving the CCD by respective use of NMR and crystallography [12]–[15] . The NTD forms a three-helical bundle stabilized through coordination of a Zn2+ ion by the invariant HHCC motif . The CTD consists of a five-stranded β-barrel similar to Src homolgy 3 domains . Although the structure of full-length retroviral IN remains elusive , its partial structures were instrumental in unraveling the mechanism of integration . The near-spherical CCD dimer cannot alone explain the concerted integration of two viral DNA ends . Indeed , the active sites , located on opposite sides of the dimeric CCD structure , are separated by ∼40 Å , while the distance between target scissile bonds in ideal B form DNA is close to 18 Å . A tetramer would be the minimal IN multimer to provide a pair of active sites with the expected spacing , and available experimental evidence suggests that the functional form of retroviral IN is indeed tetrameric [16]–[19] . An attractive model was derived from the crystal structure of a two-domain fragment of HIV-1 IN ( INNTD+CCD ) [15] . Although lacking the CTD , this construct crystallized in tetrameric form , best described as a dimer-of-dimers , with the dimers interacting with each other predominantly via NTD-CCD contacts . This model was inviting because it showed some structural similarity to the synaptic complex of the related Tn5 transposase [20] and , while the ∼29 Å separation of active sites was too far to accommodate concerted integration , it seemed plausible that flexibility along the dimer-dimer interface could provide the necessary geometry . For efficient integration , HIV-1 and other lentiviruses depend on lens epithelium derived growth factor ( LEDGF ) [21]–[23] ( reviewed in [24] ) , a cellular chromatin-associated protein implicated in transcription regulation and apoptosis [25] , [26] . LEDGF directly interacts with lentiviral IN proteins and is thought to tether the preintegration complex to chromatin for strand transfer [27]–[29] . The CCD of HIV IN is the main determinant for the interaction with LEDGF , although the NTD is required for high-affinity binding [28] , [30] . Reciprocally , a small alpha-helical domain within the C-terminal portion of LEDGF is necessary and sufficient for the interaction with IN [31] , [32] . Crystal structures of the integrase-binding domain ( IBD ) of LEDGF ( LEDGFIBD ) in complex with HIV-1 INCCD and HIV-2 INNTD+CCD have revealed molecular details of this interaction [30] , [33] . Herein we present two new crystal structures containing the NTD and the CCD of maedi-visna virus ( MVV ) IN in complex with LEDGFIBD . In both structures , this highly divergent lentiviral IN is present in tetrameric forms , stabilized by swapping pairs of NTDs between interacting dimers . Comparison of four independent IN tetramers observed in our structures reveals variability of the dimer-dimer interface , which affords juxtaposition of a pair of active sites for concerted integration . Using a range of complementary functional assays , we show that the tetramerization interface is essential for IN function , both in vitro and in the context of viral replication .
To ascertain protein-protein interfaces involved in retroviral integration , we sought to determine crystal structures of divergent lentiviral INs . MVV IN presented an appealing target because it shares less than 30% overall sequence identity with its HIV-1 counterpart ( Figure S1 ) . Opportunely , sequence analysis of LEDGF cDNA isolated from sheep , a natural MVV host , confirmed that the amino acid sequence of its IBD is identical to that of the human ortholog . Bacterial co-expression of MVV INNTD+CCD ( residues 1–219 ) with LEDGFIBD yielded monodisperse preparations of the protein-protein complex without introducing solubilizing point mutations into the IN construct . The protein complex crystallized in two forms , referred to as crystal form ( CF ) 1 and CF2 , and the resulting structures were refined to 3 . 28 and 2 . 64 Å , respectively ( Table 1 ) . The asymmetric unit ( ASU ) of CF1 contains three IN dimers ( chains A–F ) , each with a pair of associated LEDGF chains ( G–L ) . The dimers interact with each other to form three independent dimer-dimer interfaces , such that the EF dimer interacts with the AB and CD dimers , and the CD dimer with the A′B′ dimer from another ASU ( Figures S2A–S2C ) . The ASU of CF2 contains a pair of IN dimers that form a single tetramer with four associated LEDGF chains ( Figure S2D ) . Although in most IN chains the loops connecting NTDs and CCDs are disordered , clear electron density was seen in chain B of CF1 , allowing unambiguous assignment of all NTDs in this crystal form ( Figure S2C ) . In CF2 , where the NTD-CCD linkers are disordered for all monomers , unambiguous assignment of IN chain B and C NTDs ( cyan and yellow in Figure S2D ) was possible due to distance restraints: the shortest path to connect chain B Gln44 with chain C Ser55 , while avoiding clashes with the rest of the model , would be well over 50 Å , a distance that cannot be covered by 10 amino acid residues . Collectively , CF1 and CF2 reveal four independent IN tetramers ( Figure S2 ) . Within each tetramer a pair of NTDs ( henceforth referred to as inner NTDs ) mediate stable dimer-dimer interactions . The remaining ( outer ) NTDs do not share a conserved role or position within the tetramers ( Figure S2 ) . The salient details of higher-order dimer-dimer interaction are shown for three of the four tetramers ( CF1/IN chains CDEF , CF1/ABEF , and CF2/ABCD ) in Figure 1A–1C , with LEDGF chains and outer NTDs omitted for clarity . The interface within the CF1/CDA′B′ tetramer is very similar to that in ABEF , and will therefore not be discussed separately . Within tetramers , the positions of the inner NTDs relative to the opposing CCD dimers are maintained in all cases , and are identical to those seen in the earlier tetrameric HIV-1 ( Figure 1D ) and dimeric HIV-2 INNTD+CCD structures , although in the latter case the NTD-CCD interfaces were intramolecular [15] , [30] . The NTD-CCD interfaces , observed in the structures of divergent INs , share conserved features including a well-defined salt bridge between Glu11 and Lys188 ( Lys186 in HIV-1 IN; refer to Figure S1 for an MVV/HIV-1 IN sequence alignment ) and hydrophobic interactions involving Trp15 ( Tyr15 in HIV-1 IN ) and chain A Tyr134 as well as chain B Leu167 , Ile183 , Thr184 and Lys188 ( Trp132 , Val165 , Phe181 , Ile182 and Lys186 , respectively , in HIV-1 IN ) ( Figure 2A and 2B ) . An additional salt bridge is formed between Glu25 and Lys190 , and this is reproduced in the HIV-1 IN interface as Asp25:Lys188 . HIV-2 IN encodes Lys at position 25 , so it cannot form the same salt bridge; instead the related Arg188 forms a salt bridge with Glu21 ( Figure 2C ) . The conservation of the NTD-CCD interface and the resulting tetramers in crystal structures of divergent lentiviral INs strongly argues for their functional relevance . Although each IN tetramer is stabilized by identical intermolecular NTD-CCD interactions , there is remarkable variation in the relative positions and orientations of the interacting dimers ( Figure 1 , Figure S2 , Videos S1 and S2 ) . The plasticity of the dimer-dimer interface is sufficient to allow a pair of active sites from the opposing CCD dimers in CF2 to approach 14 . 9 Å separation ( measured as the distance between Cγ atoms of the active site Glu residues ) . For a comparison , the separation between the structurally-equivalent active sites in CF1/ABEF is 27 . 5 Å , while that in the HIV-1 INNTD+CCD structure [15] is ∼29 Å ( Figure 1 ) . In addition to the stable intermolecular NTD-CCD interactions , the tetramerization interface involves a loop connecting CCD helices α5 and α6 ( residues 188–196 and 186–195 in MVV and HIV-1 respectively , Figure S1 ) , termed finger [2] . Although rich in Gly residues , the loop adopts a constrained conformation stabilized by a network of hydrogen bonds , the aforementioned salt bridges with the NTD , and wields a hydrophobic residue at the tip ( Leu193 in MVV; Ile191 in HIV ) ( Figure 2D–2F ) . Examination of the dimer-dimer interfaces within individual tetramers reveals profound differences in relative orientations and contacts made by the fingers of opposing CCD dimers ( Figure 1 ) . Notably , the fingers switch positions between CF1/CDEF and CF2 structures , with CF1/ABEF representing an intermediate state ( Videos S1 and S2 ) . The most defined , symmetric and potentially relevant interactions involving this loop are observed in the CF2 structure , where side chains of Leu193 residues nucleate a hydrophobic core , engaging Ile200 , Phe203 and Thr195 from the finger of the opposing CCD dimer ( Figure 3A ) . The chain of hydrophobic contacts propagates to involve Leu24 and Val20 from the inner NTDs and Ile60 from the CCD of the same chain and is further stabilized by a well-defined salt bridge involving Arg58 and Asp18 side chains . These interactions effectively zip the two halves of the tetramer together , bringing a pair of active sites from the inner monomers into close proximity ( Videos S1 and S2 ) . A complementary interaction between the active sites involves a symmetric pair of hydrogen bonds formed by Gln150 residues of the inner monomers ( Figure 3B ) . Interestingly , the closure of the tetrameric structure also subtly modifies the internal configuration of the congregated active sites . Repulsive dipole-dipole interactions between realigned α4 helices , exacerbated by the close stacking of Arg155 side chains ( Figure 3B ) , result in a slight deformation of both helices , forcing Glu154 to shift towards Asp66 and Asp118 of the same active site . For example , the distance between the Cα atoms of Glu154 and Asp66 decreases from 10 . 4 Å in the open CF1/ABEF and CF1/CDEF conformations to 7 . 7 Å in CF2 . The active site separation in the closed tetramer observed in CF2 is compatible with the spacing between scissile phosphodiester bonds in B-form target DNA ( Figure 3C ) . Hence , CF2 represents an IN tetramer conformation committed for concerted integration . Predictably , the overall architecture of the MVV IN-LEDGF interaction is similar to that described for HIV-1 and HIV-2 INs [30] , [33]: it primarily involves the tip of the IBD , notably LEDGF residues Ile365 and Asp366 , and a cleft at the interface of the CCD dimer . The stoichiometry of MVV INNTD+CCD:LEDGFIBD complexes observed in both crystal forms is 1∶1 ( Figure S2 ) , similar to that in crystals of the HIV-1 INCCD:LEDGFIBD complex [33] . Thus , each MVV IN CCD dimer interacts with a pair of IBDs , bound at two equivalent positions . All ten CCD:IBD interfaces observed in CF1 and CF2 structures are very similar . LEDGF Ile365 forms hydrophobic interactions with Met104 , Leu131 and Tyr134 of one MVV IN chain and Met170 and Phe171 of the second IN chain ( Figure 4A ) . These interactions are related to those observed for HIV-1 , although the actual IN side-chains involved differ due to lack of sequence identity ( Figure S1 ) . As predicted [27] , LEDGF Asp366 duplicates the previously described bidentate hydrogen bond with backbone amides of MVV IN residues Asn172 and Ala173 ( Glu170 and His171 in HIV-1 ) . Lentiviral INs display surprisingly little sequence conservation at the positions directly involved in the interaction with LEDGF , itself a well-conserved protein [27] , [33] . Predictably , some details of the MVV IN-LEDGF interaction show marked differences with those elaborated for HIV-1 or HIV-2 INs [30] , [33] ( Figure 4 ) . One such difference occurs due to MVV encoding residues Arg100 and Leu131 in place of two Ala residues at HIV-1 IN equivalent positions 98 and 129 . The bulky side-chains pry MVV IN CCD helices α1 and α3 slightly apart , enlarging the cleft occupied by the protruding IBD loop . The extra space is filled by the insertion of LEDGF side chains Asn367 and Leu368 , which make hydrogen bonds with Gln97 and Arg100 and hydrophobic interactions with Leu130 , Leu131 and Tyr134 , respectively ( Figure 4A ) . The result of this alternate binding orientation is a ∼34° rotation of the IBD with respect to the HIV-1 structure , centered at the site of interaction with the CCD . Consequently , Phe406 and Val408 located on the second loop of the IBD make hydrophobic interactions with MVV IN Tyr134 . Such interactions would not be possible with HIV-1 IN due to an inevitable steric conflict with the side chain of Trp131; the equivalent position of MVV IN is occupied by Lys133 , whose flexible side chain makes way for incoming Phe406 and Val408 ( Figure 4 ) . The rotation also allows LEDGF Lys364 to form a hydrogen bond with the carbonyl group of MVV IN Pro169 ( Figure 4A ) . In the complex with HIV-1 IN , Lys364 forms a salt bridge with non-conserved IN residue Glu170 . Additional interactions involving the positive patch on one side of the IBD structure and carboxylates of HIV-1 and HIV-2 IN NTDs are important for high affinity interaction [30] . In CF2 , LEDGF residues Lys401 , Lys402 and Arg405 are sufficiently close for electrostatic interactions with MVV IN Asp41 , Glu10 and Glu9 , respectively ( not shown ) . However , the side chains of the interacting residues are not well defined in electron density maps . To test the relevance of the tetramerization interface observed in the crystal structures , we designed a series of HIV-1 IN mutants . The changes were introduced at the positions predicted to be important for tetramerization by the earlier HIV-1 INNTD+CCD [15] and current MVV structures . Multimerization properties of purified proteins were studied using analytical size exclusion chromatography ( SEC ) ( Figure 5 ) . All proteins displayed non-ideal behavior , such as temperature-dependent interaction with Superdex and silica matrices ( data not shown ) , and generated complex elution profiles , indicative of multiple multimeric forms . Nonetheless , in agreement with previous results [34] , the elution profile of WT HIV-1 IN was consistent with a predominantly tetrameric species ( Figure 5A ) . Preincubation of IN with an excess of LEDGFIBD prior to injection resulted in a slightly earlier elution of the major species ( Figure 5B ) . The peak shift of ∼0 . 15 ml was consistent with binding of four 10-kDa LEDGFIBD molecules per IN tetramer . Zinc binding is essential for folding of the NTD and promotes HIV-1 IN self-association [6] , [35]–[37] . Concordantly , disruption of zinc coordination by the NTD H12N mutation grossly affected the SEC elution profile ( Figure 5A ) . Under these experimental conditions , H12N IN behaved as a dimer or a dimer-monomer mixture . Remarkably , several mutations at the NTD-CCD interface affected HIV-1 IN self-association properties to a similar extent as the NTD-destabilizing H12N mutation . Thus , mutating Tyr15 , a residue involved in several hydrophobic interactions with the CCD ( Figure 2B ) , abolished multimerization ( Figure 5A ) . Similarly , disrupting the Glu11:Lys186 salt bridge with single point mutations E11K or K186E resulted in pronounced shifts to lower molecular weight species ( Figure 5A ) . Interestingly , less dramatic shifts were observed for D25K and K188D , suggesting lower importance of the Asp25:Lys188 interaction for multimerization . These results agree with an earlier report showing that the K186A change had a greater effect on tetramerization than did K188A [34] and are consistent with the crystal structures . Thus , in HIV-1 IN [15] , the ε-amino group of Lys188 is shared between the carboxylates of Asp25 and Glu198 , separated from either by ∼4 . 6 Å ( Figure 2B ) . In contrast , the ε-amino group of HIV-1 Lys186 is only ∼3 . 2 Å from the carboxylate of Glu11 , indicating strong bonding . In MVV IN , the Glu25:Lys190 salt bridge appears to be the stronger of the two , with the Glu11:Lys188 interaction weakened by interactions between Glu11 and Lys14 ( Figure 2A ) . Remarkably , combining the E11K and K186E mutations in one protein led to a significant recovery of the higher-multimeric HIV-1 species , as did mixing equimolar quantities of single mutants ( Figure 5A ) . Cross-linking with the homobifunctional reagent BS3 confirmed that WT HIV-1 IN existed as a predominantly tetrameric species , and that tetramerization was highly sensitive to the E11K or K186E mutation ( Figure S3 ) . Further corroborating results of the SEC experiments , partial recovery of tetramer formation was observed in equimolar mixtures of E11K and K186E mutants ( Figure S3 ) . These results demonstrate that ( i ) the contact between Glu11 and Lys186 is essential for the stability of higher-order HIV-1 IN multimers in vitro and ( ii ) the salt bridge between these residues can be formed intermolecularly , corroborating the NTD-CCD connectivity observed in the MVV structures . Deletion of residues 190Gly-Ile-Gly192 from the CCD finger abrogated multimerization ( Δ190-2 , Figure 5C ) , although the I191E point mutant multimerized as well as WT ( Figure 5C ) . Therefore , while the whole of the constrained loop structure is clearly essential for multimerization , the conserved aliphatic residue at its tip is not . LEDGF was shown to enhance HIV-1 IN tetramerization [34] , an effect likely dependent on the IBD-NTD interface [30] , [34] . Accordingly , preincubation with LEDGFIBD led to at least partial rescue of multimerization for all NTD-CCD interface mutants ( Figure 5B and 5D ) . These results are wholly consistent with the crystal structures ( Figure S2 ) , where LEDGF binding is expected to stabilize IN tetramers . Next , we tested the HIV-1 IN mutants for the ability to catalyze 3′-processing and DNA strand transfer using either a blunt-ended 500-bp ( Figure 6A ) , or blunt or pre-processed 23-bp mimic of the viral U5 DNA end ( Figure 6B and 6C ) . The assay with the longer viral DNA substrate distinguishes concerted strand transfer reaction products from those that result from the integration of a single donor DNA end into only one strand of target DNA , whereas the oligonucleotide-based assays do not . The Y15A and Δ190-2 mutants were almost devoid of 3′-processing activity ( Figure 6B ) , and did not produce strand transfer products in either assay format ( Figures 6A–6C ) . Interestingly , I191E IN , which multimerized as well as WT , was attenuated for both 3′-processing ( Figure 6B ) and strand transfer ( Figures 6A and 6C ) , suggesting that I191E tetramers might exist in a defective conformation . Mutants D25K and K188D functioned relatively well in 3′-processing ( Figure 6B ) and retained near WT strand transfer activity in the oligonucleotide assay ( Figure 6C ) . However D25K and , to a lesser degree , K188D , displayed a specific concerted integration defect , with D25K generating half-site products at near WT level ( Figure 6A ) . Mutations E11K and K186E , targeting the Glu11:Lys186 salt bridge , decreased 3′-processing and strand transfer activities ( Figures 6B and 6C ) while completely eliminating concerted integration ( Figure 6A , lanes 8–13 ) . The importance of the salt bridge was further illustrated by the recovery of concerted integration activity to almost WT levels with the double E11K/K186E mutant ( Figure 6A , lanes 14–16 ) . This result also confirmed that the mutations do not affect the intrinsic catalytic properties of the enzyme , or its functional association with donor or target DNA . Likewise , mixing the two individual mutants ( E11K+K186E ) , each incapable of forming intramolecular NTD-CCD interactions , recuperated concerted integration ( lanes 17–19 ) . Consistent with the observation that LEDGF binding aids IN multimerization ( Figures 5B and 5D , see also [34] ) , the concerted integration activities of E11K , D25K , K188D , and , to a lesser extent , K186E , were rescued in the presence of the host factor ( Figure 6D ) . IN mutations were next introduced into the single round HIV-Luc vector , and infectivity was assessed 2 days post-infection . Based on the results with purified enzymes , E11K , K186E , and E11K/K186E mutants were initially compared to D64N/D116N ( N/N ) active site mutant virus . N/N supported 0 . 25±0 . 06% ( n = 6 ) residual HIV-Luc infectivity , whereas E11K , K186E , and E11K/K186E faired less well , each scoring near the assay detection limit ( <0 . 025% of HIV-Luc ) . This suggested that E11K , K186E , and E11K/K186E might exert class II mutant behavior: certain mutants , like N/N , are referred to as class I because they are specifically blocked at integration and accordingly support residual levels of gene expression from unintegrated DNA , whereas the majority of mutant viruses , class II , display additional reverse transcription and/or virus assembly defects [38] . The preliminary assignment of class II mutant behavior is consistent with the previously reported K186Q reverse transcription defect [39] , [40] . The activities of class II mutant viral enzymes can be analyzed during infection via trans-incorporation of Vpr-IN fusion proteins into assembling virus particles [40] , [41] . Various mutant proteins were therefore compared to Vpr-INWT for their ability to stimulate N/N-Luc infectivity . Vpr-INWT enhanced N/N-Luc infection approximately 6- to 16-fold , yielding overall infectivities that ranged from 1 . 4% ( Figure 7 ) to 6 . 8% ( data not shown ) of HIV-Luc . Vpr-INE11K and Vpr-INK186E displayed partial activities , yielding 39±5 . 8% and 33±1 . 6% of Vpr-INWT function in repeat ( n = 5 ) experiments ( Figure 7 and data not shown ) . Akin to the result with purified enzymes , the Vpr-INE11K/K186E double mutant was significantly more active than either single mutant , actually outshining Vpr-INWT to restore 21 . 5% of HIV-Luc activity ( Figure 7 ) . Trans incorporation of separate Vpr-INE11K and Vpr-INK186E single mutants also significantly stimulated N/N-Luc , yielding 15 . 7% of overall HIV-Luc infectivity . Importantly , incorporating the D116A active site mutation into either Vpr-INE11K or Vpr-INK186E counteracted the stimulatory affect of the mixture ( Figure 7 ) . Immunoblotting revealed similar levels of functional and non-functional Vpr-IN protein incorporation into virions ( Figure 7 ) .
Retroviral INs function as multimers [16]–[19] , [41]–[43] . Due to obvious structural restraints , such as distances between active sites in their dimeric CCDs , minimally a tetramer of IN would be required to carry out concerted integration of both viral DNA ends . Because a structure of a full-length IN has remained elusive , much effort is being expended to model a full-length IN tetramer based on the available partial crystal structures [15] , [44]–[46] . In this work we present two crystal structures containing a two-domain construct of a divergent lentiviral IN in complex with the isolated IBD of its natural host cofactor LEDGF . Together with earlier results [15] , [30] , these structures elucidate the mechanism for IN tetramerization , indicate the dramatic flexibility of the IN tetramerization interface ( Videos S1 and S2 ) and for the first time reveal a tetramer conformation that is compatible with concerted integration ( Figure 3 ) . It is important to note that the CTD , which is also involved in IN multimerization [7] , [47] , is not present in our structures . Nonetheless , we were able to validate the tetramerization interface observed in the crystals using a range of functional assays with mutants of full-length HIV-1 IN . Herein we demonstrated that the main proponent of IN tetramerization is the conserved NTD-CCD interface brought about by swapping a pair of NTDs between participating IN dimers . We recently showed that within an IN dimer , the NTDs fold back onto their own CCDs [30] . In contrast , in the context of a tetramer , interacting IN dimers swap a pair of NTDs ( Figure 1 ) . Although similar connectivity was postulated earlier [15] , hitherto direct evidence for NTD swapping was not available . The absence of structured NTD-CCD linkers and the open conformation of the HIV-1 INNTD+CCD tetramer described by Wang et al . [15] allow various alternative NTD-CCD connectivities ( for more discussion see [30] and [2] ) . Detailed analyses of the NTD-CCD interfaces in the current MVV as well as earlier HIV-1 and HIV-2 IN structures [15] , [30] revealed a network of conserved interactions ( Figure 2 ) that are essential for multimerization ( Figure 5 ) . The key interaction involves a conserved salt bridge , which in HIV-1 IN is mediated by Glu11 and Lys186 , and the latter residue has been shown to be important for HIV-1 IN multimerization [34] , [48] . Herein we demonstrate that the Glu11:Lys186 salt bridge is functionally reversible , allowing us to significantly extend prior observations . Thus , while individual mutations of both residues abrogated tetramerization and concerted integration , mixing HIV-1 IN E11K and K186E single mutants partially recovered tetramerization ( Figure 5 and S3 ) , rescued concerted integration in vitro ( Figure 6 ) , and moreover robustly stimulated N/N-Luc infection ( Figure 7 ) . These results imply that the intermolecular NTD-CCD interface is functional . The behavior of the E11K+K186E mixture in the virus complementation assay highlights this functionality . A significant fraction of inner monomers from the N/N+Vpr-INWT mixture will contain inactivating D64N/D116N mutations , poisoning tetramer function . In the N/N+Vpr-INE11K+Vpr-INK186E case , N/N IN would only be allowed to assume the role of the outer monomers to accommodate the reversible salt bridge between inner INE11K+INK186E pairs . Hence the activity of the Vpr-INE11K+Vpr-INK186E mixture outshines that of Vpr-INWT in this assay ( Figure 7 ) . Furthermore , because the double E11K/K186E mutant is functional , we can conclude that the mutations do not affect the intrinsic catalytic properties of the enzyme or its interactions with DNA . Not only did the double mutant E11K/K186E recover concerted integration activity and HIV-1 infection , it also supported greater levels of 3′-processing and half-site integration activities over the individual mutant proteins . This indicates that while it could be possible for a dimer of IN to catalyze 3′-processing and half-site integration , both reactions are more efficiently catalyzed by a tetramer ( or possibly a larger aggregate of IN dimers ) . A similar conclusion was made based on kinetic studies utilizing a mutant of an alpharetroviral IN that was unable to form tetramers [49] . Furthermore , this finding is in agreement with Li and Craigie [50] , who observed that 3′-processing and concerted HIV-1 integration are functionally coupled . We speculate that tetramerization could play a role in the correct organization of the active site . Indeed , closure of the tetramerization interface leads to a slight compression of the MVV IN active site , with active site residue Glu154 relocating closer to its Asp66 and Asp118 mates . In addition , IN tetramerization and engagement of the viral DNA termini are likely to be co-dependent . Intriguing questions remain as to the nature of the class II phenotype of HIV-1 IN mutants [38] . Although E11K/K186E HIV-1 IN was fully competent to carry out concerted integration starting with blunt ended substrate ( Figure 6 ) , the virus carrying these mutations was not infectious . It is possible that Glu11 and/or Lys186 impact important noncatalytic IN function ( s ) at a step prior to integration , such as reverse transcription [51] . Alternatively , the mutations might disrupt interaction with a host factor that would engage the outer IN monomers of the tetramer during integration . It is important to note that the IN tetramer structure contains two structurally and functionally-distinct pairs of IN subunits , with the inner pair ( painted cyan and yellow in Figure 1 ) swapping their NTDs and providing the active sites , and the other pair ( green and orange ) playing a supporting role . Therefore , many residues in the IN sequence likely have two distinct functions . The current MVV and the earlier HIV-1 IN [15] structures ( Figure 1 ) , as well as our analyses of the Δ190-2 mutant , clearly indicate that the CCD finger is involved in multimerization . Similarly , alterations within the CCD finger structure impaired tetramerization of alpharetroviral IN [48] . Truncation of the constrained loop structure is expected to affect salt bridges involving HIV-1 Lys186 and Lys188 side chains , and thus the crucial intermolecular NTD-CCD interface . The significance of the aliphatic residue at the tip of the finger structure ( Ile191 in HIV-1 or Leu193 in MVV ) is highlighted by its conservation in all lentiviruses . A substitution of HIV-1 IN Ile191 for Glu produced a protein that was able to multimerize ( Figure 5 ) , but was essentially devoid of enzymatic activity ( Figure 6 ) . These results are consistent with the importance of the aliphatic residue for the formation of the closed tetramer conformation , represented by the CF2 structure , where a pair of Leu193 residues from opposing CCD fingers nucleate a hydrophobic core at the dimer-dimer interface ( Figures 1 and 3A ) . Superposing partial HIV-1 IN structures onto the CF2 MVV structure results in a plausible full-length tetrameric model devoid of significant steric conflicts ( Figure S4 ) . Although the majority of the residues involved in the closure of the dimer-dimer interface are not conserved between MVV and HIV-1 INs ( Figure S1 ) , the model suggests a potential role of HIV-1 IN residue Tyr194 in formation of the closed structure via hydrophobic interactions with Ile191 from the opposing dimer . The conformational variability of the dimer-dimer interface described here suggests that the committed IN tetramer is likely stabilized via IN-DNA interactions . It is noteworthy that the synaptic Tn5 transposase:DNA complex is primarily stabilized via protein-DNA interactions [20] . An earlier model based on the open conformation of HIV-1 IN tetramer suggested that target DNA would bind into the cleft between widely separated active sites [15] , [44] . This implies that the active sites would approach target DNA duplex from opposing sides , a configuration not easy to reconcile with the size of target DNA duplications flanking integrated proviruses . On the other hand , the closed tetramer conformation would preclude target DNA access to the interior of the dimer-dimer interface . We speculate that the target duplex binds roughly along the vector connecting the active sites , affording them direct access to the scissile phosphodiester bonds located across the major groove ( Figures 3C and S4 ) . This binding mode is supported by findings of Katzman and colleagues , who demonstrated that HIV-1 IN residue Ser119 , located within CCD α2 , is involved in target DNA capture [52] , [53] . More recent results from this laboratory further confirm a target DNA binding platform extending along this direction [54] . The locations of the CTDs in the current model ( Figure S4 ) are compatible with a role in binding viral DNA termini . It is noteworthy that although the CCD-CTD linker adopted alpha helical conformation in the structure of the HIV-1 INCCD+CTD fragment [13] , similar studies with INs from Rous sarcoma and simian immunodeficiency viruses [55] , [56] highlighted significant flexibility of this region . DNA binding moreover induced considerable structural rearrangements within the CCD-CTD linker of HIV-1 IN [57] . Hence positions and orientations of the CTDs within the tetramer cannot be directly inferred from the available partial structures . Because the current MVV ( Figure S2 ) and earlier HIV-1 IN [15] tetrameric structures disagree on the locations of the outer NTDs , their roles remain uncertain . In particular , the NTD-NTD interfaces observed in MVV CF1 tetramers ( Figure S2 ) differ both from each other and from those observed in HIV-1 INNTD+CCD or the isolated HIV-1 NTD dimer in solution [12] . These interfaces likely represent packing artifacts in crystal structures , which contain continuous chains of dimers linked by tetramerization interfaces , with the outer NTDs in one tetramer assuming roles of inner NTDs in another ( not shown ) . In contrast , the tetramer in CF2 is isolated and does not have NTD:NTD contacts , with the outer NTDs folding back to lock onto the connected CCDs ( Figure S2D ) . We expect that the outer NTDs would reveal their role in a tetramer of full-length retroviral IN or within its complex with DNA .
The plasmid pCDF-MVV-INNTD+CCD , used for bacterial expression of non-tagged MVV INNTD+CCD , was made by ligating a PCR fragment encoding residues 1–219 of IN from molecular clone KV1772 [58] between NcoI and XhoI sites of pCDF-Duet1 ( Novagen ) . The MVV INNTD+CCD:LEDGFIBD complex , used for crystallography , was produced and purified essentially as described previously for HIV-2 INNTD+CCD:LEDGFIBD [30] . Briefly , MVV INNTD+CCD was co-expressed with His6-SUMO-tagged LEDGFIBD in Escherichia coli PC2 cells [27] transformed with pCDF-MVV-INNTD+CCD and pES-IBD-3C7 [30] . The protein complex , enriched by absorption to NiNTA agarose ( Qiagen ) , was treated with SUMO and human rhinovirus ( HRV ) 14 3C proteases to release LEDGFIBD from the N-terminal His6-SUMO tag and the C-terminal flexible tail , respectively . The complex , purified by SEC on a Superdex-200 column in 1 M NaCl , 50 mM Tris HCl , pH 7 . 4 , was supplemented with 5 mM DTT , concentrated to 12–15 mg/ml and stored on ice . For purification of isolated LEDGFIBD , E . coli PC2 cells transformed with pES-IBD-3C7 [30] and grown in LB medium to an A600 of 0 . 8–1 . 0 were induced with 0 . 25 mM isopropyl-thio-β-D-galactopyranoside at room temperature for 3–4 h . Bacteria were lysed by sonication in 500 mM NaCl , 0 . 5 mM PMSF , 20 mM imidazole , 50 mM Tris HCl , pH 7 . 4 , and the pre-cleared lysate was incubated with NiNTA agarose ( Qiagen ) . The resin was extensively washed with 20 mM imidazole , 500 mM NaCl , 50 mM Tris HCl , pH 7 . 4 . The protein , eluted in 200 mM imidazole , 500 mM NaCl , 50 mM Tris HCl , pH 7 . 4 , was supplemented with 5 mM DTT and SUMO protease ( 20 mg protease per mg protein ) [30] , [59] and dialyzed overnight against cold 250 mM NaCl , 25 mM Tris HCl pH 7 . 4 , 5 mM DTT , 40 mM imidazole . The protease and the released His6-SUMO tag were depleted by passing the sample through a 5-ml HisTrap column ( GE Healthcare ) . To remove the disordered C-terminal tail ( residues 436–471 ) [60] , the protein was digested with HRV14 3C protease ( 20 mg protease per mg protein ) at 7°C in the presence of 10 mM DTT . Minimal LEDGFIBD was then purified by chromatography through a HiLoad 16/60 Superdex-200 column ( GE Healthcare ) . To obtain HIV-1 IN mutants , the corresponding changes were introduced into pCPH6P-HIV1-IN [30] using quick-change procedure ( Stratagene ) . Full-length LEDGF , HIV-1 IN and the mutant proteins were produced in bacteria and purified as previously described [27] , [30] . All proteins used in activity assays and analytical chromatography experiments were tag-free . Hanging drop vapor diffusion crystallization experiments were conducted at 18°C , mixing 1 µl MVV INNTD+CCD:LEDGFIBD complex ( 5 mg/ml in 400 mM NaCl , 2 mM DTT , 20 mM Tris HCl , pH 7 . 4 ) with 1 µl of a reservoir solution . CF1 was obtained using a reservoir solution of 25–30% ( w/v ) Jeffamine M600 ( Hampton Research ) in 100 mM Bis-Tris propane-HCl , pH 6 . 6 . The crystals , grown over 5–10 days to a size of ∼50×50×30 µm , were cryoprotected in the reservoir solution supplemented with 20% ( v/v ) glycerol and frozen by immersion in liquid nitrogen . CF1 belonged to space group P21 with unit cell constants a = 91 . 1 Å , b = 148 . 9 Å , c = 91 . 1 Å , α = γ = 90° , β = 113 . 4° . A dataset , collected at 100 K on beamline I04 of the Diamond Light Source ( Oxford , UK ) , was integrated and scaled in XDS [61] to 3 . 28 Å ( Table 1 ) . The structure was solved by molecular replacement using Molrep [62] with three search models: HIV-1 IN CCD dimer ( residues 50–212 , from 2b4j ) , followed by LEDGF IBD ( residues 347–426 , 2b4j ) , and finally HIV-1 IN NTD ( residues 1–43 , 1k6y ) . The resulting model containing six IN and six LEDGF chains was refined using rigid body , maximum likelihood and simulated annealing routines as implemented in Phenix [63] with manual building in Coot [64] . Group isotropic B factors ( one per residue ) and 6-fold non-crystallographic symmetry ( NCS ) were applied throughout; translation , libration and screw-rotation ( TLS ) displacements [65] were accounted for towards the end of the refinement . The final refined model has good geometry and Rwork/Rfree of 21 . 3/25 . 5% ( Table 1 ) . CF2 was obtained using a reservoir solution containing 0 . 7–0 . 9 M ( NH4 ) 2HPO4 , 2 . 5% Jeffamine M600 and 100 mM Bis-Tris propane-HCl , pH 7 . 0 . Crystals , cryoprotected in the reservoir solution supplemented with 20% glycerol , were frozen by immersion in liquid nitrogen , and the data were acquired at 100 K on the Diamond Light Source beamline I02 . CF2 belongs to space group P21 with unit cell constants a = 102 . 7 Å , b = 83 . 0 Å , c = 115 . 3 Å , α = γ = 90° , β = 101 . 8° . Diffraction intensity data were corrected for the observed lattice translocation defect [66]; full details of the detwinning procedure will be reported elsewhere ( S . H . , P . C . , J . W . , submitted for publication ) . The structure was solved by molecular replacement , using Molrep with the MVV IN CCD dimer ( from CF1 ) as a search model , followed by IBD ( from 2b4j ) and MVV IN NTD . Two CCD dimers were found to form a tetramer with four associated NTDs and IBDs . Following additional cycles of building , TLS and restrained refinement in Refmac [67] the final model had Rwork/Rfree of 22 . 6/25 . 5% and good geometry ( Table 1 ) . Weighted 2Fo-Fc electron density maps for chain B of CF1 ( showing the ordered NTD-CCD linker ) and for three parts of the CF2 structure ( NTD:CCD and IBD:CCD interfaces , as well as the chain B active site with an associated phosphate ion ) are shown in Figure S5 . Transition states between observed conformations of the MVV IN tetramer ( Videos S1 and S2 ) were simulated using Yale Morph Server [68] . Protein structure images and animations were generated using PyMOL software ( DeLano , W . L . , http://www . pymol . org ) . The coordinates and structure factors for CF1 and CF2 have been deposited in the Protein Data Bank with pdb IDs 3hpg and 3hph , respectively . Raw diffraction images are available upon request . SEC was carried out using a 4 . 3-ml KW403-4F column ( Shodex ) attached to an ÄKTA Purifier system ( GE Healthcare ) . The column was immersed in ice and operated at 0 . 275 ml/min in 750 mM NaCl , 10 mM MgCl2 and 20 mM HEPES-NaOH , pH 7 . 0 . Thirty-five µl IN ( WT or mutant ) diluted to 0 . 6 mg/ml in gel filtration buffer supplemented with 25 µM ZnCl2 and 2 . 8 mM CHAPS was injected into the column . Where indicated , 0 . 3 mg/ml LEDGFIBD was pre-incubated with IN on ice for 5 min prior to injection . For cross-linking , 6 µl WT , E11K or K186E IN , or an equimolar IN mutant mixture ( 0 . 54 mg/ml protein in 1 M NaCl , 5 mM DTT , 7 . 5 mM CHAPS , 25 mM Hepes-NaOH , pH 7 . 5 ) was diluted with 21 µl reaction buffer ( 0 . 75 M NaCl , 2 mM MgSO4 , 25 µM ZnCl2 , 25 mM Hepes-NaOH , pH 7 . 5 ) . Cross-linking was initiated by addition of 4 µl BS3 ( Pierce; fresh 15–1 . 7 mM stock in water ) . Where indicated , reactions were supplemented with 0 . 3% SDS prior to addition of the cross-linking reagent . Reactions , allowed to proceed for 30 min at 18°C , were stopped by addition of Laemmli SDS PAGE sample buffer . The products were separated in Novex 10–20% Tricine SDS PAGE gels ( Invitrogen ) and detected by staining with Sypro Orange ( Invitrogen ) . Oligonucleotide-based 3′-processing assays were carried out as previously described [40] . Briefly , blunt 23-bp DNA substrate was obtained by annealing 5′-end labeled 5′-CAGTGTGGAAAATCTCTAGCAGT with 5′-ACTGCTAGAGATTTTCCACACTG . Reactions ( 20 µl ) contained 0 . 1–0 . 4 µM IN , 25 nM substrate DNA in 20 mM NaCl , 7 . 5 mM MnCl2 , 10% glycerol , 10 mM β-mercaptoethanol , 0 . 1 mg/ml BSA and 25 mM MOPS-NaOH , pH 7 . 2 . Reactions , initiated by addition of 0 . 5 µl IN in 750 mM NaCl , 5 mM DTT and 10 mM Tris-HCl , pH 7 . 4 ( DB ) , were allowed to proceed for 1 h and were stopped by addition of 15 mM ethylenediaminetetraacetic acid ( EDTA ) and 0 . 3% sodium dodecyl sulfate ( SDS ) . Products , separated on denaturing 17% polyacrylamide gels , were visualized and quantified by phosphor autoradiography using a Storm 860 imager . Strand transfer reactions using pre-processed donor DNA were carried out under the same conditions , except the 5′-CAGTGTGGAAAATCTCTAGCA oligonucleotide was radiolabeled . The concerted integration assay [50] , [69] used pGEM-9Zf ( - ) as target and 5′- end labeled 500-bp HIV-1 RU5 fragment [30] as donor . Reactions ( 25 µl ) contained 50–200 nM IN , 15 nM donor DNA and 11 nM pGEM in 100 mM NaCl , 10 mM MgSO4 , 5 mM DTT , 20 µM ZnCl2 , 5% dimethyl sulfoxide ( DMSO ) , 12% polyethylene glycol ( PEG ) 6000 and 20 mM HEPES-NaOH , pH 7 . 5 . Reactions were started with the sequential addition of donor DNA , target DNA , 1 µl IN in DB and 1 . 25 µl DMSO , followed by a 2–4 min pre-incubation at room temperature before addition of 6 µl 50% PEG6000 . Reactions , incubated for 1 h at 37°C , were stopped by addition of 15 mM EDTA and 0 . 3% SDS . The products , deproteinized by digestion with proteinase K and precipitation with ethanol , were analyzed by electrophoresis through 1 . 5% agarose gels in Tris-acetate buffer . Products were visualized in dried gels using a Storm 860 imager ( GE Healthcare ) . The LEDGF-dependent concerted integration assay [30] used blunt 32-bp donor DNA substrate , obtained by annealing oligonucleotides 5′-CCTTTTAGTCAGTGTGGAAAATCTCTAGCAGT and 5′-ACTGCTAGAGA TTTTCCACACTGACTAAAAGG , and supercoiled pGEM target . Reactions ( 40 µl ) contained 1 µM IN , 0 . 6 µM LEDGF , 0 . 6 µM donor DNA and 34 nM pGEM in 20 mM Hepes-NaOH pH 7 . 4 , 10 mM DTT , 110 mM NaCl , 5 mM MgSO4 and 4 µM ZnCl2 . Reactions were initiated by the addition of 2 µl IN in DB , followed by a 10-min incubation at room temperature , before addition of 2 µl LEDGF in DB . Reactions were allowed to proceed for 30 minutes at 37°C and stopped by addition of 25 mM EDTA and 0 . 5% SDS . DNAs recovered by ethanol precipitation following deproteinization with 40 µg proteinase K for 1 h at 37°C were resolved by electrophoresis through 1 . 5% agarose gels and detected by staining with ethidium bromide . Single-round HIV-1 strain NLX . Luc . R- carrying luciferase in place of nef ( HIV-Luc ) and either WT or D64N/D116N ( N/N ) active site mutant IN was pseudotyped with vesicular stomatitis virus G envelope glycoprotein as described [22] , [30] , [40] . WT or mutant IN protein was incorporated in trans during virus assembly by co-transfecting pRL2P-Vpr-IN plasmids [40] . Resulting cell-free virus titers were determined by reverse transcriptase incorporation of [α-32P]TTP . HeLa-T4 cells [70] ( 40 , 000 in 12 well plates ) infected in duplicate with 106 RT-cpm in 0 . 8 ml for 8 h were washed , lysed at 44 h post-infection , and luciferase activities were normalized to total protein content . Levels of virion-associated IN and capsid proteins were compared using western blotting as described [71] , [72] . GenBank entries EE831415 and EE774051 , identified using translated BLAST to span portions of Ovis aries LEDGF/p75 cDNA , were used to design oligonucleotide primers to isolate its entire coding region . To this end , total RNA prepared from phytohemagglutinin-stimulated sheep peripheral blood mononuclear cells was reverse-transcribed using Superscript III ( Invitrogen ) and gene-specific primer 5′-CTATCAATTACACATTAACATACACAC . A fragment spanning the entire coding region of sheep LEDGF cDNA was PCR-amplified using EasyA DNA polymerase ( Stratagene ) and primers 5′-CCTGAAACATGACTCGCGACTTCAAACC , 5′-ACTTCTCAAATGTTCTTTATATTCCAGG . The sequence determined using a pool of products from four independent amplification reactions was deposited with GenBank with the accession number FJ497048 ( RefSeq: NM_001143892 ) .
|
Integrase is the viral enzyme that orchestrates insertion of both ends of retroviral DNA into a host cell chromosome . This process , thought to require a tetramer of integrase , involves two concerted cutting/joining ( transesterification ) reactions that target a pair of phosphodiester bonds in chromosomal DNA , separated by ∼18 Å . Until now , the architecture of the integrase tetramer responsible for concerted integration has remained a mystery . We now report two crystal structures containing the N-terminal and catalytic core domains from a lentiviral integrase in complex with its co-factor LEDGF . Comparison of the structural arrangements observed in our crystals elucidates the details of the integrase tetramerization interface , reveals its dramatic flexibility and the mechanism by which a pair of active sites can be brought into close proximity . Taking advantage of the structural data , we generated a series of HIV-1 integrase mutants designed to disrupt or re-create its tetramerization interface . Biochemical and virus replication studies with these mutants strongly support the functional significance of the tetrameric architecture observed in the crystal structures . Our results provide important novel insights into the assembly of the functional integrase tetramer and will be invaluable for the ongoing efforts to model the retroviral pre-integration complex .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology/immunodeficiency",
"viruses",
"molecular",
"biology",
"biochemistry/protein",
"chemistry",
"virology/viral",
"replication",
"and",
"gene",
"regulation"
] |
2009
|
Structural Basis for Functional Tetramerization of Lentiviral Integrase
|
Phagocytosis by hemocytes , Drosophila macrophages , is essential for resistance to Streptococcus pneumoniae in adult flies . Activated macrophages require an increased supply of energy and we show here that a systemic metabolic switch , involving the release of glucose from glycogen , is required for effective resistance to S . pneumoniae . This metabolic switch is mediated by extracellular adenosine , as evidenced by the fact that blocking adenosine signaling in the adoR mutant suppresses the systemic metabolic switch and decreases resistance to infection , while enhancing adenosine effects by lowering adenosine deaminase ADGF-A increases resistance to S . pneumoniae . Further , that ADGF-A is later expressed by immune cells during infection to regulate these effects of adenosine on the systemic metabolism and immune response . Such regulation proved to be important during chronic infection caused by Listeria monocytogenes . Lowering ADGF-A specifically in immune cells prolonged the systemic metabolic effects , leading to lower glycogen stores , and increased the intracellular load of L . monocytogenes , possibly by feeding the bacteria . An adenosine-mediated systemic metabolic switch is thus essential for effective resistance but must be regulated by ADGF-A expression from immune cells to prevent the loss of energy reserves and possibly to avoid the exploitation of energy by the pathogen .
Pro-inflammatory state of various immune cells in mammals , as neutrophils , dendritic cells and M1 macrophages and some adaptive-immunity cells ( e . g . Th17 ) , is associated with a metabolic switch including increased glycolysis and glucose consumption [1] . Therefore , the immune response is an energy-demanding process [2] and poor nutrition often leads to reduced immune resistance [3] . However , the energy supply to the immune system must be tightly regulated to protect it from exploitation by the pathogen [4] and also properly allocate limited energy reserves , especially because the immune response is known to be associated with anorexia [5 , 6] . For example , mycobacterial infection may lead to wasting due to the consumption of energy reserves [7] . The mechanisms for the systemic regulation of the metabolism during an immune response have only recently begun to be intensively studied . Systemic insulin resistance , caused by pro-inflammatory cytokines ( e . g . TNF-α or Il-6 ) , is believed to be an important mechanism to ensure the energy supply to the immune system [8] . Although this is likely to be beneficial during an acute response , chronic inflammation may lead to various metabolic disturbances [9] . Experimental studies of the effects of metabolic regulations on host-pathogen interactions are practically challenging , especially in higher experimental models such as mice [6 , 10] . Simpler model organisms such as the fruit fly Drosophila melanogaster may provide a more streamlined platform for such studies . We have recently shown that extracellular adenosine ( e-Ado ) mediates a systemic metabolic switch upon infection of Drosophila larva by a parasitoid wasp [11] . This switch redirected energy normally devoted to developmental processes towards the immune system . Such a systemic switch is crucial for an effective immune response because blocking adenosine signaling drastically reduces resistance . We have also shown that immune cells produce this signal , thus usurping energy from the rest of the organism . Such a privileged behavior of the immune system was recently proposed to be important for an effective immune response [8] . Adenosine ( Ado ) is an important intracellular metabolite of purine metabolism . It can also appear extracellularly under certain conditions , thus becoming an important signaling molecule . For example , when tissue is damaged , adenosine triphosphate leaks out and is converted by ecto-enzymes into e-Ado [12] . Alternatively , when intracellular adenosine triphosphate levels decrease due to metabolic stress , increased adenosine monophosphate is converted to Ado , which is subsequently released into the extracellular space where it binds to adenosine receptors informing other tissues of this metabolic stress . During extensive exercise , muscles release Ado to induce fatigue to allow the initiation of recovery to this type of stress [13] . In another example , hypoxia induces adenosine to be released from the affected tissue to increase nutrient-rich blood flow [14] . This putative stress signaling by adenosine is conserved in evolution from primitive unicellular organisms all the way up to humans . For example , adenosine is released upon starvation by the social bacteria Myxococcus xanthus to induce the formation of fruiting bodies that produce spores for surviving harsh conditions [15] . Adenosine therefore represents a very ancient and universal system for signaling metabolic stress . Here we investigate the regulatory role of e-Ado during the immune response and the effects of this regulation on host-pathogen interactions . Taking two complementary approaches , we block adenosine signaling by elimination of the adenosine receptor AdoR and enhance adenosine signaling by down-regulation of adenosine deaminase ADGF-A . We analyze the effects of these manipulations on host response to two types of bacterial infections: acute caused by Streptococcus pneumoniae , and chronic by Listeria monocytogenes . We show that adenosine regulates energy supply during bacterial infections in adult flies and that adenosine signaling is crucial for host defense , extending our previous results from research on parasitoid wasp infection [11] . In addition , we also demonstrate an important regulatory role of adenosine deaminase ADGF-A during the immune response . Enhancing adenosine effects may be beneficial for host resistance during the acute phase of the infection , but negative feedback regulation becomes important to prevent exhausting energy reserves and to prevent host nutrients being exploited by the pathogen .
When our control w flies were injected with 20 000 CFUs of S . pneumoniae , bacteria grew to 200–300 thousands within 24 hours ( Fig 1A and 1B ) . This load persisted in the majority of flies for a few further days ( acute phase ) during which many flies died ( Fig 1C ) . The surviving flies cleared the infection by the 5th-6th day ( Fig 1A ) , but some lethality still occurred after this , probably because they either did not clear the infection or they did not recover ( Fig 1C ) . It has been previously demonstrated by injecting the flies with latex beads , which jams their hemocytes , making them unable to phagocytose , that phagocytosis was crucial for the clearance of S . pneumoniae [16] . We confirmed this result here when the injection of latex beads 24 hours before infection drastically increased pathogen load ( to a million bacteria per fly; Fig 1B ) and led to the rapid death of the flies within 2 days ( Fig 1C ) . The effect of blocking phagocytosis on resistance was independent of the used genotypes ( Fig 1B and 1C; see further detailed description of genotypes ) . Fig 1D shows the survival of PBS-injected ( i . e . non-infected ) flies under our experimental conditions . We had previously shown that a systemic metabolic switch , which redirected energy devoted to developmental processes to the immune system , was required for the resistance of Drosophila larvae to parasitoid wasp infection [11] . Therefore , we analyzed here the systemic metabolism during S . pneumoniae infection and found a similar systemic metabolic switch , in which circulating glucose in control flies rose upon infection , peaking at 12 hpi ( Fig 2A ) , while glycogen levels were reduced to half within the first 24 hpi ( Fig 2B ) . In agreement with this glycogen breakdown during infection , the expression of the glycogen phosphorylase rose during S . pneumoniae infection ( Fig 2C ) . To test the importance of energy release from glycogen for the immune response , we tested the effect of silencing glycogen phosphorylase ( GlyP ) during S . pneumoniae infection . We used Gal4/UAS induced RNA interference ( RNAi ) with thermosensitive Gal80 to induce RNAi just prior to infection to avoid any developmental effects of silencing GlyP; both control flies and flies with RNAi showed similar levels of glycogen at the beginning of infection ( Fig 2B ) . Knocking down GlyP specifically in the fat body ( ppl>GlyP[RNAi]; Fig 2C ) significantly reduced glycogen breakdown and suppressed hyperglycemia upon infection ( Fig 2A and 2B ) . This led to increased pathogen load ( Fig 2D ) and decreased survival ( Fig 2E ) . Similar effects on pathogen load and survival were achieved by systemically induced RNAi ( Act>GlyP[RNAi] ) ( Fig 2D and 2F ) . In our previous work , mentioned above [11] , we showed that the systemic metabolic switch was mediated by adenosine signaling . Therefore , we used here three different genetic manipulations to investigate the role of adenosine on the immune response against the bacterial infection of adult flies . First , we blocked systemic adenosine signaling by the adoR null mutation of the adenosine receptor AdoR . To enhance the effects of adenosine , we lowered an expression of adenosine deaminase ADGF-A first by a heterozygous null mutation in the ADGF-A gene ( adgf-a[kar]/+ ) , since the homozygous mutation causes larval lethality [17] . Second , we lowered the expression of ADGF-A by targeted RNAi in hemocytes ( Drosophila immune cells ) , using Gal4/UAS system [18] when the expression of double stranded RNA for ADGF-A was driven by the hemocyte-specific Hemolectin-Gal4 driver ( Hml>ADGF-A[RNAi] ) . The metabolic switch , induced by S . pneumoniae infection and expressed by hyperglycemia associated with glycogen breakdown ( Fig 2A and 2B ) , was clearly delayed in the adoR mutant , in which the glucose and the glycogen profiles were delayed 6–9 hours compared to the control ( Fig 3A and 3B ) . In contrast , lowering ADGF-A in adgf-a/+ and Hml>ADGF-A[RNAi] flies caused the continuation of hyperglycemia beyond the 9-hpi control peak ( Fig 3C and 3E ) , leading to a deeper depletion of glycogen stores ( Fig 3D and 3F ) , thus causing an expected opposite effect compared to the adoR mutation . The role of adenosine signaling in glycogen breakdown is further supported by the expression of the glycogen phosphorylase and glycogen synthase . There was a lower expression of the glycogen phosphorylase and a higher expression of the glycogen synthase in the adoR mutant ( Fig 3G and 3H ) . In contrast to the situation in the control flies , where the glycogen phosphorylase rose and the glycogen synthase dropped during infection , their expression did not change in adoR ( Fig 3G and 3H ) , supporting the role of adenosine signaling in the regulation of glycogen synthesis/breakdown . Since the immune response was dependent on the systemic metabolic switch and the results above showed that the switch was mediated by adenosine , we tested the resistance of the flies with manipulated adenosine signaling . Injecting a sublethal dose ( 15 000 CFUs or less ) allowed most of the control flies to survive the acute phase of infection ( the first 7–8 days; Fig 4A ) while a lethal dose ( 20 000 CFUs or more ) killed 50% of control flies by day 8 and led to less than 25% survival overall ( Fig 4B ) . The adoR mutation decreased survival compared to the w control when injected with both sublethal and lethal doses ( Fig 4A and 4B ) . While the lethal dose killed most of the control flies , adgf-a/+ and Hml>ADGF-A[RNAi] flies improved survival during the acute phase ( Fig 4B and 4C ) . The observed effects were not caused by a difference in vigor [19] among these genotypes since the injection of PBS buffer led to comparable survival under the experimental conditions ( Fig 1D ) . Fig 4D shows that S . pneumoniae load increased mostly within the first 24 hpi , reaching up to three hundred thousand CFUs per fly , and was eventually cleared within 5–6 days . We monitored the pathogen load during this 6-day span and detected differences among the tested genotypes , especially during the first 24–48 hpi ( Fig 4D ) when most flies were still alive ( Fig 4A–4C ) . At later time points , differences in pathogen load in flies , which were still alive and available for pathogen load measurements , virtually disappeared between genotypes ( Fig 4D and S1 Fig ) . Therefore , we monitored in detail the growth of S . pneumoniae during the first 24 hpi . In agreement with the relative survival rates , the number of CFUs was significantly higher in adoR and lower in both adgf-a/+ and Hml>ADGF-A[RNAi] compared to the controls ( Fig 4E and 4F and S2 Fig ) . Although the adgf-a/+ and Hml>ADGF-A[RNAi] flies had lower pathogen loads and better survived the acute phase of S . pneumoniae infection , many of them died after the acute phase and thus their overall survival was eventually comparable to the control ( Fig 4B and 4C ) . These results suggest that adgf-a/+ and Hml>ADGF-A[RNAi] better controlled the pathogen load , leading to their better survival of the acute infection phase , but eventually have problems with recovery from infection; the reason for the recovery problem is unknown . It is important to stress that the use of the heterozygous adgf-a knock out mutant as a model led to a similar result as that obtained by depleting ADGF-A in hemocytes by RNAi ( Hml>ADGF-A[RNAi] ) and to an opposite result to that obtained with the adoR mutation , thus connecting the phenotypes specifically to the genetic manipulations of extracellular adenosine . We were not able to detect clear and reproducible changes in the extracellular adenosine levels , having problems rapidly collecting a reasonable amount of hemolymph from adult flies and due to the short half-life of this molecule upon sample collection . Therefore , to determine whether the observed effects of lowering ADGF-A expression are indeed due to adenosine signaling , we tested resistance to S . pneumoniae in a double mutant , heterozygous for adgf-a and homozygous for adoR ( adgf-a adoR/adoR ) , lowering ADGF-A expression and at the same time removing adenosine signaling . Fig 4B and S3 Fig demonstrate that the effect of adgf-a/+ is completely suppressed by the adoR mutation in the adgf-a adoR/adoR double mutant when both survival and pathogen loads are similar to the adoR single mutant . This demonstrates that the observed effects in adgf-a/+ are dependent on adenosine signaling . As shown above , resistance to S . pneumoniae is dependent on phagocytosis; blocking phagocytosis with latex beads erased differences between genotypes ( Fig 1B and 1C ) demonstrating that the differences are connected to phagocytosis . The effectivity of the phagocytic response could be influenced by the number of hemocytes , which in turn is dependent on genetic background [20] . A comparable number of active phagocytes , and even adult hemocytes in general , among w , adoR and adgf-a/+ genotypes ( with unified genetic backgrounds ) showed that it was not the different number of phagocytes ( Fig 5A and 5B ) causing the observed effects of these genotypes on resistance to S . pneumoniae . We detected a significantly lower number of cells labeled by pHrodo , a marker of active phagocytosis , in the adoR mutant challenged by S . pneumoniae 6 hours prior to a pHrodo injection when the total number of the Hml>GFP-labeled hemocytes was the same as in the control ( Fig 5C ) . This lower phagocytosis in challenged adoR was significantly increased by a carbohydrate-rich , 10%-glucose diet ( Fig 5C ) . So , although there are comparable numbers of phagocytes with comparable basal phagocytic capacity ( without challenge ) , their activity ( but not their numbers ) is lower in the adoR mutant when the immune response is activated by the challenge . Next , we tested a different type of infection , triggered by Listeria monocytogenes , which causes a chronic intracellular infection leading eventually to the death of the host [21] . Intracellular infection is established by phagocytosis followed by the escape of L . monocytogenes from the phagosome to the cytoplasm . Similarly to S . pneumoniae infection , the L . monocytogenes infection also led to a systemic metabolic switch , characterized by hyperglycemia ( Fig 6A ) ; glycogen stores were accordingly reduced ( Fig 6B ) . The adoR mutation reduced hyperglycemia ( Fig 6A ) and increased glycogen storage compared to the control ( Fig 6B ) , the two effects persisting even during the chronic phase ( Fig 6C ) . Since there was no peak of hyperglycemia during the acute phase of L . monocytogenes infection , we did not detect a difference in glucose level in the adgf-a/+ flies ( S4 Fig ) , however both the adgf-a/+ and Hml>ADGF-A[RNAi] flies had significantly lower glycogen stores during the chronic phase ( Fig 6C ) . The length of host survival depends on the injected dose [19] and the genetic background that determines , for example , the number of phagocytes [20] . We injected 1000 bacteria ( S5A Fig ) leading to a median time to death of 17 days in our control flies ( Fig 7A ) . Both the adoR mutation and lowering ADGF-A by adgf-a/+ and Hml>ADGF-A[RNAi] shortened survival during L . monocytogenes infection to a median time to death of only 8 days ( Fig 7A–7C ) . The observed shorter survival durations were associated with increased pathogen load ( Fig 7D–7F ) , ostensibly suggesting a lower resistance to L . monocytogenes for all three genotypes . However , distinguishing the intracellular and extracellular L . monocytogenes populations revealed a more complicated picture . We used gentamycin , an antibiotic that is unable to cross cellular membranes , to determine the total number of bacteria ( without gentamycin treatment ) and the number of intracellular bacteria ( after gentamycin treatment ) . This gentamycin-chase assay described in [5] showed that the increased pathogen load in the adoR mutant was mainly due to an increased extracellular population ( Fig 7D and S5 Fig ) . On the other hand , the increased pathogen load in adgf-a/+ and Hml>ADGF-A[RNAi] flies was almost solely caused by an increase in the intracellular L . monocytogenes population ( Fig 7E and 7F and S5A Fig ) ; the extracellular population disappeared faster in adgf-a/+ and Hml>ADGF-A[RNAi] flies compared to the control suggesting more effective phagocytosis upon lowering ADGF-A . The increased intracellular pathogen load in adgf-a/+ and Hml>ADGF-A[RNAi] flies persisted even during the chronic phase ( day 7 , Fig 8A ) , when the phagocytosis likely no longer played a role because all bacteria were intracellular ( i . e . no difference in total and intracellular load in w control in Fig 8A ) . The adoR mutation also increased the total pathogen load at day 7 , although this represented a surge in the extracellular population , as the intracellular population was similar to the control ( Fig 8A ) . Melanization was previously shown to be important in controlling extracellular L . monocytogenes loads when lowering the melanization response led to an increased extracellular bacteria population [22] . The increased presence of extracellular bacteria in adoR flies most likely stimulated an increase in disseminated melanization [22] at day 7 ( S6 Fig ) , when only 37% of w showed melanization ( with extensive melanization in 12% of flies; n = 40 ) compared to 63% of adoR flies ( with extensive melanization in 30% fies; n = 24; S6 Fig ) . In our previous work , we had been able to partially rescue the effect of adoR on the metabolic switch and immune response by providing more glucose in the fly diet [11] . Here we show that a sugar-enriched , 10%-glucose diet significantly increased the survival of adoR flies ( Fig 7A and S7 Fig ) and lowered the pathogen load to control levels during both S . pneumoniae and L . monocytogenes infections ( Fig 8B and 8C ) . The sugar-enriched diet did not influence the survival or pathogen load of adgf-a/+ or Hml>ADGF-A[RNAi] flies ( Figs 7B , 7C and 8C ) . The results above demonstrated the effects of adenosine manipulation on immune defenses associated with phagocytosis and metabolism . Host immunity may also be mediated by the production of antimicrobial peptides ( AMPs ) and , therefore , we analyzed the expression of four selected AMPs ( S8 Fig ) . Both S . pneumoniae and L . monocytogenes induced the expression of Defensin , Diptericin , Drosocin and Metchnikowin , albeit to different extents . As shown in S8 Fig , during S . pneumoniae infection , the adoR mutant showed a lower expression of Defensin , Diptericin , and Drosocin at 6 hpi but not at 18 hpi . In the case of L . monocytogenes infection , only the Metchnikowin expression was strongly reduced at 6 hpi in the adoR mutant . The lower expression of three out of four AMPs in adgf-a/+ or Hml>ADGF-A[RNAi] flies infected with both S . pneumoniae and L . monocytogenes ( S8 Fig ) rather reflects the lower extracellular-bacteria load in these flies than being the reason for more effective clearance of this extracellular population . Adenosine is regulated by the adenosine deaminase ADGF-A [17] . Gene expression analysis showed that ADGF-A expression increased both upon L . monocytogenes and S . pneumoniae challenges ( Fig 9A and 9B ) . Fig 9A also demonstrates that there is lower ADGF-A mRNA expressed in the adgf-a/+ heterozygous mutant; some of this mRNA might possess a premature stop codon ( adgf-a mutation ) leading to aberrant protein production , which is not distinguishable by the used q-PCR ( for a description of this mutation , see [23] ) . The Hml-induced RNA interference of ADGF-A specifically in hemocytes effectively silenced the ADGF-A expression , demonstrating the functionality of the Hml>ADGF-A[RNAi] knockdown ( Fig 9A ) . It is important to note that the expression was measured on the whole-organism level , suggesting that the expression of ADGF-A in hemocytes , where Hml-Gal4 driver is expressed , represents a majority of the whole organism expression of this gene . A hemocyte-specific expression analysis ( Fig 9C ) confirmed this speculation when hemocytes showed a one order of magnitude higher expression of ADGF-A compared to the expression measured in whole flies and this expression rose four times during infection demonstrating that hemocytes are the primary producers of ADGF-A . The time-course expression during S . pneumoniae infection ( Fig 9B and 9C ) also showed that the ADGF-A expression rose after 9 hpi , which coincided with the down-regulation of hyperglycemia during S . pneumoniae infection ( Fig 3A ) .
In this work , we used two types of bacterial infection of adult D . melanogaster , one caused by the facultative intracellular bacterium L . monocytogenes and the other triggered by the extracellular pathogen S . pneumoniae , to test the effects of the genetic manipulation of adenosine signaling on host-pathogen interactions . To block systemic adenosine signaling , we used a null adoR mutation in the adenosine receptor and to enhance the effects of adenosine , we either removed one copy of the adenosine deaminase ADGF-A gene or lowered its expression by RNAi . We show that both infections are associated with a systemic metabolic switch manifested by hyperglycemia at the expanse of the glycogen stores . Manipulating e-Ado signaling influences this metabolic switch and , at the same time , affects host resistance . The activated immune system requires an increased supply of energy/nutrients [24] . Therefore , we propose that the observed e-Ado-mediated systemic metabolic switch supplies the immune system with the required nutrition , and thus is important for the effectivity of the immune response . Resistance to S . pneumoniae was shown to be dependent on effective phagocytosis by hemocytes [16] . We confirm here the crucial role of phagocytosis by injecting latex beads which jam hemocytes , making the hemocytes unable to phagocytose . Blocking the phagocytosis made flies extremely sensitive to S . pneumoniae infection and eliminated the differences in responses between the control and mutants used in this work . The observed effects of Ado manipulation on immunity are not due to an altered phagocyte number [20] since all the strains used in this work have comparable numbers of hemocytes , including active phagocytes . We show here that S . pneumoniae infection is associated with systemic metabolic switch manifested by hyperglycemia ( peaking at 9–12 hpi ) at the expanse of the glycogen stores , when glycogen drops to less than one third within 24 hpi . This could merely be a pathological consequence of the infection , however phagocytic cells are known to increase glycolysis and glucose consumption [25] and thus the systemic metabolic switch may also be a reflection of an increased need for energy by the hemocytes phagocytosing S . pneumoniae . The latter possibility is strongly supported by our experiment with knocking down GlyP , an enzyme responsible for glycogen breakdown . Knocking down GlyP in the fat body immediately prior to infection decreased resistance to S . pneumoniae , demonstrating that the glucose , liberated from the glycogen stores , is required for effective phagocytosis . Therefore , the observed systemic metabolic switch is most likely an active process ensuring adequate energy supply to the activated immune system . Our previous work [11] demonstrated that the immune cells of Drosophila larva release Ado during parasitoid wasp infection to mediate a systemic metabolic switch and thus to supply the immune system with the required nutrients . Here we show that blocking adenosine signaling by adoR mutation prevents the metabolic switch during L . monocytogenes infection and postpones the hyperglycemic peak and glycogen use during S . pneumoniae infection . This notion is further supported by transcriptional data showing that the glycogen synthesis/breakdown is under the AdoR control . Furthermore , hyperglycemia during S . pneumoniae infection peaks at 9 hpi , which coincides with a rise in ADGF-A expression . When ADGF-A action is lowered by the heterozygous adgf-a mutation or RNAi , hyperglycemia continues at the increased expense of the glycogen stores . All these results together demonstrate that the systemic carbohydrate metabolism is regulated by e-Ado in adult flies during infection . The AdoR signaling causes a release of energy , i . e . hyperglycemia associated with decreased glycogen stores and the e-Ado-mediated release of energy is regulated by ADGF-A . The lower resistance of the adoR mutant to S . pneumoniae is then in agreement with the role of Ado signaling in the systemic metabolic switch and with the importance of this switch for the effective immune response . Therefore , we propose that adenosine signaling mediates the supply of energy to the activated immune system . This is supported by our experiments with a carbohydrate-rich diet compensating for the missing switch in systemic metabolism in the adoR mutant when this diet increases phagocytosis , normalizes the pathogen load and ultimately increases survival in the adoR mutant when compared to the carbohydrate-poor diet . These observations are similar to our previous work showing that a carbohydrate-rich diet rescued the effect of adoR on the production of immune cells during parasitoid wasp infection [11] . The role of e-Ado in supplying the energy to immune cells is further supported by the opposite effect achieved by removing one copy of ADGF-A or lowering its expression by RNAi . These genetic manipulations have the opposite effect on carbohydrate metabolism compared to adoR and at the same time lower the pathogen load during S . pneumoniae infection , demonstrating more effective phagocytosis in these flies . The complete suppression of this increased resistance by simultaneously mutating adoR demonstrates that the effect of lowering ADGF-A is indeed due to adenosine signaling . We can then conclude that e-Ado mediates a systemic metabolic switch which is important for supplying energy to immune cells , and e-Ado is thus crucial for effective phagocytosis and host resistance to S . pneumoniae . S . pneumoniae infection provides a simpler model of host-pathogen interaction in which the clearance is crucially dependent on phagocytosis and the host either clears the bacteria or the pathogen outgrows and kills the host . The other pathogen , L . monocytogenes , is a facultative intracellular bacterium causing a chronic and ultimately lethal infection in flies . Phagocytosis is actually a way for this pathogen to colonize the host [21]; the intracellular population is eventually established following escape from the phagosome . Infection by this pathogen causes a systemic metabolic switch similar to the one caused by S . pneumoniae , i . e . hyperglycemia associated with the loss of glycogen stores . Here , as is the case in S . pneumoniae , adenosine mediates this switch because adoR decreases the infection induced-hyperglycemia associated with a lower loss of glycogen while lowering ADGF-A leads to a greater loss of glycogen . The adoR mutation increases the extracellular load of L . monocytogenes suggesting that , similarly to the case of S . pneumoniae infection , this mutation decreases the effectivity of phagocytosis via the suppression of the adenosine-mediated metabolic switch . This is supported by evidence of rescue with an increase of glucose in the diet , which normalizes pathogen load and increases the survival of the adoR mutant . In agreement with this , lowering ADGF-A leads to the opposite effect , in which this manipulation leads to a faster clearance of the extracellular L . monocytogenes population associated with the increased intracellular population , suggesting more effective phagocytosis . Thus far , the results obtained with L . monocytogenes , focused on the metabolism and early response associated with phagocytosis , are similar to those obtained with S . pneumoniae . Phagocytosis is not , of course , the only defense mechanism available to the flies , two other mechanisms , melanization and antimicrobial peptides , are discussed further . Unlike to S . pneumoniae infection , both blocked and enhanced Ado signaling decreased the survival of flies upon L . monocytogenes infection . This decreased survival is associated with increased total pathogen loads in all examined adoR , adgf-a/+ , and Hml>ADGF-A[RNAi] flies . In the adoR mutant , the larger pathogen load is mainly due to an increased extracellular population , as mentioned above . In contrast , the increased pathogen load in adgf-a/+ and Hml>ADGF-A[RNAi] flies is almost completely caused by an increase in the intracellular L . monocytogenes population . The increased intracellular load can be due to less effective intracellular defense mechanisms , which remains mostly unexplored in flies [26] , or due to an increased carrying capacity for the pathogen at the expanse of the host energy reserves as suggested by a greater loss of glycogen reserves in flies with lowered ADGF-A . An increased bacteria population obviously requires more nutrients . Among the important virulence factors of L . monocytogenes is the bacterial homolog of glucose- 6-phosphate translocase , which allows the pathogen to exploit hexose phosphates from the host cell as a carbon source [27] . In addition , L . monocytogenes hijacks host cell actin polymerization for its propagation [28] . Since this process requires energy , an increased supply to an infected cell may potentially further promote the propagation of this intracellular pathogen . The question is whether the nutrient supply to infected cells is the limiting factor for pathogen proliferation and propagation , however , there is evidence that the proliferation capacity of intracellular pathogens is strongly influenced by host metabolism [4] . If so , as a reaction to the increased release of energy from the host stores caused by lowering ADGF-A , the carrying capacity could increase , and thus could lead to the observed increase in the intracellular pathogen load . However , we cannot exclude the possibility that lowering ADGF-A somehow decreases the host intracellular defense and the associated wasting is just a consequence of increased pathogen load . Carbohydrate-rich diet normalizes the pathogen load in the adoR mutant , decreasing the extracellular population of L . monocytogenes , which is present in this mutant on a carbohydrate-poor diet . This in turn leads to the longer survival of this mutant , comparable to control flies . A carbohydrate-rich diet can rescue the glycogen loss in flies with lowered ADGF-A but the pathogen load , being intracellular , is still increased and survival is as short as on a carbohydrate-poor diet . It seems that increased dietary glucose can compensate for the glycogen loss detected on a carbohydrate-poor diet , which is potentially associated with the nutrient exploitation by the pathogen in these flies , as discussed above . However , the higher pathogen burden still kills the flies faster , suggesting that the survival is primarily determined by the pathogen number . Although we focus on phagocytosis in this work , there are other immunity mechanisms which may influence the host resistance and physiology . While melanization was not observed with S . pneumoniae infection , it was shown to play an important role in controlling the extracellular population of L . monocytogenes [22] . Therefore , the increased extracellular L . monocytogenes load detected in adoR could also be due to the lower induction of melanization response . However , we detected a rather stronger disseminated melanization [22] in the adoR mutant suggesting that the induction of melanization is not lowered by the lack of adenosine signaling and may rather reflect the reaction provoked to a greater extent by the increased extracellular population of L . monocytogenes in this mutant . Nevertheless , the role of adenosine signaling in the melanization arm of immunity requires further work . Host immunity is also mediated by the production of antimicrobial peptides ( AMPs ) . We did not detect clear and consistent differences in the expression of AMPs in the mutants that would be in agreement with the observed effects on resistance , i . e . lower expression of AMPs in adoR and higher expression in adgf-a/+ . The expression of AMPs was shown to be dependent on pathogen load , at least with L . monocytogenes [19] . We did not test the expression with different loads and therefore we can directly compare the expressions between mutants and control at only 6 hpi when the pathogen loads are similar among the genotypes both for S . pneumoniae and L . monocytogenes . In the case of Defensin , Diptericin , and Drosocin , but not Metchnikowin , the adoR mutant showed lower expression during S . pneumoniae infection , which would be consistent with its lower resistance . However , such a difference was not detected at 18 hpi and in the case of L . monocytogenes for neither time point . In case of adgf-a/+ , the expression of AMPs is mostly lower compared to control , including cases when the pathogen loads are comparable , which is in contrast to the observed higher resistance . While we cannot exclude the role of AMPs in the observed effects on resistance , this arm of immunity does not seem to play as important a role as phagocytosis . Our work demonstrates the crucial role of e-Ado in mediating the systemic metabolic switch , which is required for effective immune response . Although we did not directly measure e-Ado levels , the opposite effects of the adoR mutation , which blocks adenosine signaling on one hand , and lowering ADGF-A , which degrades adenosine , on the other , leave little doubt that the observed effects are indeed associated with the e-Ado action . Being able to mount an adequate immune response is vital for the organism ( as demonstrated by the lower resistance of adoR ) but its regulation is as important as the response itself; reducing regulation by lowering ADGF-A may prolong the switch associated with a greater loss of the host energy reserves and may potentially lead to the exploitation of released nutrients by the pathogen . The regulatory role of ADGF-A is demonstrated by a hyperglycemic peak and consecutive decrease after 9–12 hpi during S . pneumoniae infection which coincides with a rise in ADGF-A expression; when one copy of ADGF-A gene is removed or ADGF-A is knocked down by RNAi , the hyperglycemia continues past this time point and glycogen continues to fall . It is important to note that ADGF-A was knocked down specifically only in hemocytes by the Hml-Gal4 driver; also , our hemocyte-specific expression analysis demonstrated that hemocytes are the primary source of ADGF-A . Therefore , the immune cells are the regulators of e-Ado actions during the immune response . This is in agreement with our previous results showing a specific expression of ADGF-A in specialized larval immune cells , lamellocytes that encapsulate parasitoid wasp eggs [29] , which represent a later phase of immune response . Taken together with immune cells first releasing adenosine to usurp energy from the rest of the organism during a parasitoid wasp attack [11] , and later producing ADGF-A to downregulate adenosine , we can say that the immune system is able to regulate its privileged access to nutrients by producing a regulator of the signal mediating the systemic metabolic switch . In summary ( Fig 10 ) , we demonstrate here that bacterial infections of the adult Drosophila flies are associated with a systemic metabolic switch , manifested by a hyperglycemia at the expense of glycogen stores . This switch supplies the immune system with the energy required for an effective response and is mediated by e-Ado signaling . Blocking e-Ado signaling demonstrates its crucial role for an effective immune response , in this case phagocytosis . However , the proper regulation of e-Ado by adenosine deaminase ADGF-A is also important . Although lowering such regulation may increase host resistance to some infections , it may also lead to an excessive loss of energy reserves during chronic infection . An increased release of energy allocated for the immune system may also be exploited by the pathogen leading to decreased host survival . Here we build on our previous work showing that the privileged behavior of the immune system is crucial for host resistance by revealing a mechanism in which the same immune system limits its own privilege to ultimately protect the whole organism .
The fly strains used for manipulating adenosine were first backcrossed 10 times into the same w1118 strain with a genetic background based on CantonS , which was used as a control ( referred to simply as w in text ) . To block adenosine signaling , we used a homozygous adoR1 null mutation ( FBal0191589 ) of the adenosine receptor AdoR ( CG9753; FBgn0039747 ) . To enhance adenosine effects , we lowered the fly adenosine deaminase ADGF-A ( CG5992; FBgn0036752 ) by heterozygous null adgf-akar/+ mutation [23] , referred to as adgf-a/+ in the text , and by RNA interference for the ADGF-A gene P{GD17237} ( VDRC-50426; FBtp0028959 ) using the HmlΔ-Gal4 driver ( FBti0128549 ) specific to hemocytes in combination with a thermosensitive GAL80 construct P{tubP-GAL80ts}2 ( FBti0027797 ) . To induce RNAi , we crossed w1118; HmlΔ-Gal4; P{tubP-GAL80ts} flies to w1118; ADGF-ARNAi-P{GD17237} flies and the resulting progeny ( referred to as Hml>ADGF-A[RNAi] ) lowered ADGF-A mRNA to 20% or less of control levels ( Fig 9A ) . As a control for RNAi , we crossed w1118; ADGF-ARNAi-P{GD17237} to w1118 , with the resulting progeny referred to as ADGF-A[RNAi]/+ . To induce RNAi for glycogen phosphorylase , we crossed y , w1118; P{KK100434}VIE-260B line ( VDRC-109596; FBtp0066083 ) to either P{ppl-GAL4 . P} ( FBti0163688 ) or Act-Gal4 in combination with a thermosensitive GAL80 construct P{tubP-GAL80ts} to obtain P{KK100434}VIE-260B/P{ppl-GAL4 . P}; P{tubP-GAL80ts}/+ ( labeled as ppl>GlyP[RNAi] ) or P{KK100434}VIE-260B/P{tubP-GAL80ts}; Act-Gal4/+ ( labeled as Act>GlyP[RNAi] ) flies . KK control line y , w1118; P{attP , y[+] , w[3`]} ( VDRC-60100 ) was crossed to the Gal4 lines to obtain control flies for the GlyP RNAi with the same genetic background ( labeled as either ppl x KK control or Act x KK control ) . For hemocyte counting , we used the HmlΔ-Gal4 UAS-EGFP marker on chromosome II in each of the w , adoR , and adgf-a/+ backgrounds . Flies were grown on cornmeal medium ( 8% cornmeal , 5% glucose , 4% yeast , 1% agar ) in 6-oz square bottom plastic bottles ( 20 females per bottle laid eggs for 24 h only to prevent crowding ) . The w , adoR and adgf-a/+ flies were raised in bottles at 25°C , 70% humidity with 12/12 hours light/dark cycle; the Hml>ADGF-A[RNAi] , ppl>GlyP[RNAi] and Act>GlyP[RNAi] and ADGF[RNAi]/+ and KK control flies were raised at 18°C to suppress RNAi during development and transferred to 25°C 24 h before infection to induce RNAi . Two-day-old male progeny flies were anesthetized with carbon dioxide and collected in plastic vials ( 20 flies per vial ) with either carbohydrate-poor , 0%-glucose ( 8% cornmeal , 4% yeast , 1% agar , no additional sugar ) or carbohydrate-rich , 10%-glucose ( 8% cornmeal , 4% yeast , 10% glucose , 1% agar ) diets and transferred every other day to a fresh meal . The Listeria monocytogenes strain 10403S was stored at -80°C in brain and heart infusion ( BHI ) broth containing 25% glycerol . For the experiments bacteria were streaked onto Luria Bertani ( LB ) agar plates containing 100 μg/mL streptomycin and incubated at 37°C overnight; plates were stored at 4°C and used for inoculation for a period of two weeks . Single colonies were used to inoculate 3 mL of BHI and incubated overnight at 37°C without shaking to obtain a morning 600 nm optical density ( OD600 ) of approx . 0 . 4 . Further , L . monocytogenes cultures were diluted to OD600 0 . 01 in phosphate buffered saline ( PBS ) and stored on ice prior to loading into an injection needle . The Streptococcus pneumoniae strain EJ1 ( D39 streptomycin-resistant derivative; [30] ) was stored at -80°C in Tryptic Soy Broth ( TSB ) media containing 10% glycerol . For the experiments , bacteria were streaked onto blood-containing TSB agar plates containing 100 μg/mL streptomycin and incubated at 37°C overnight; a fresh plate was prepared for each experiment . Single colonies were used to inoculate 3 mL of TSB with 100 000 units of catalase ( Sigma C40 ) and incubated overnight at 37°C + 5% CO2 without shaking . Morning cultures were 2x diluted in TSB with fresh catalase and were grown for an additional 4 hours , reaching an approximate 0 . 4 OD600 . Final cultures were concentrated by centrifugation and re-suspended in phosphate buffered saline ( PBS ) so that the concentration corresponded to OD600 2 . 4 and stored on ice prior to injection needle loading . For sublethal doses , we used approximately 12000–15000 CFUs , for lethal doses 20 000 CFUs; the EC50 was between 15000–20000 CFU . Seven-day-old male flies were anaesthetized with carbon dioxide . The Eppendorf Femtojet microinjector and a drawn glass needle was used to inject precisely 50 nl of bacteria or mock buffer into the fly at the cuticle on the ventrolateral side of the abdomen . Infectious doses were determined for each experiment by plating a subset of flies at time zero . 50 nl of 10% 0 . 5-μm carboxylate-modified polystyrene ( latex ) beads ( Sigma L5530 ) in PBS or pure PBS as control were injected 24 hour before infection into the adult fly body cavity to block phagocytosis . Single flies were homogenized in PBS using a motorized plastic pestle in 1 . 5 ml tubes . Bacteria were plated in spots onto LB ( L . monocytogenes ) or TSB ( S . pneumoniae ) agar plates containing streptomycin in serial dilutions and incubated overnight at 37°C before manual counting . To determine intracellular L . monocytogenes loads , flies were injected with 50 nl of gentamycin solution ( 1 mg/ml in PBS ) 3 h prior to fly homogenization . Pathogen loads of 16 flies were determined for each genotype and treatment in each experiment; at least three independent infection experiments were conducted and the results were combined into one graph ( in all presented cases , individual experiments showed comparable results ) . Values were transformed to logarithmic values , since they followed the lognormal distribution , and compared using unpaired t-tests corrected for multiple comparisons using the Holm-Sidak method in the Graphpad Prism software . A total of 200 to 300 flies were injected for each genotype and treatment in one experiment; at least three independent infection experiments were repeated and combined into one survival curve ( in all presented cases , individual experiments showed comparable results ) . Injected flies were placed into vials with 20 flies per vial , transferred to a fresh vial every other day and checked daily to determine mortality . Flies infected by L . monocytogenes were kept at 25°C and flies infected with S . pneumoniae were kept at 29°C . Survival curves were generated by Graphpad Prism software and analyzed by Log-rank and Gehan-Breslow-Wilcoxon ( more weight to deaths at early time points ) tests , as specified in the appropriate figure legends . The number of hemocytes in adult flies were determined by counting Hml>GFP-positive cells visualized by confocal microscopy . The number of phagocytic cells was determined by injection of the marker pHrodo Red S . aureus Bioparticles ( ThermoFisher Scientific ) 40 min prior to fixing flies . Flies were fixed by 4% paraformaldehyde in PBS and imaged using confocal microscopy with maximal projection from five different layers; the same setting of the Z- stack range as well as the intensity of lasers were used for all animals . The cells were observed in whole flies to detect possible gross changes but no obvious differences were observed . The exact number of cells were counted within a selected thorax region as depicted in Fig 2A using Fiji software and compared by student t-tests using the Graphpad Prism software . A fluorescent activated cell sorter was used for an isolation of HmlΔ-Gal4 UAS-EGFP-labeled hemocytes from the adult flies . Approximately 12 000 living cells were seperated from the homogenate of 100 fly males . The males used for this analysis were anesthetized with CO2 , washed several times with PBS and then homogenized by sterile pestle in 800 μl of PBS . Cell homogenate was then filtered through a 70-μm cell strainer ( Corning ) and washed three times with ice cold PBS followed each time by centrifugation at 5000 RPM for 3 min at 4°C . Samples were filtered once more through a 40-μm cell strainer immediately before sorting . S3E cell sorter ( BioRad ) was used for sorting . GFP-specific cells constituted approximately 1% of the total cell number . Sorted hemocytes were verified by fluorescent microscopy and with DIC . Gene expression was analyzed by quantitative real-time PCR . Whole flies or sorted HmlΔ-Gal4 UAS-EGFP hemocytes were homogenized and total RNA was isolated by Trizol reagent ( Ambion ) according to manufacturer’s protocol . DNA contamination was removed by using a Turbo DNAse free kit ( Ambion ) according to the protocol ( 37°C 30 min ) with subsequent inactivation of DNAse by DNAse inactivation reagent ( 5 min at RT , spin 13000 RPM at RT ) . Reverse transcription was done by Superscript III reverse transcriptase ( Invitrogen ) and amounts of mRNA of particular genes were quantified using the IQ Sybr Green Supermix Mastermix ( BioRAd ) on a CFX 1000 Touch Real time cycler ( BioRad ) . Expressions were analyzed using double delta Ct analysis , normalized to the expression of Ribosomal protein 49 ( Rp49 ) in the same sample . Relative values ( fold change ) to control ( specified in each graph ) were compared and shown in the graphs . Primer sequences can be found in S1 Table . Samples were collected from three independent infection experiments with three technical replicates for each experiment and compared by unpaired t-tests using Graphpad Prism software . Glucose and glycogen were measured by approaches published in [31] . 3 flies were homogenized in 1x PBST ( PBS with 0 . 3% Tween ) and large tissue fragments were pelleted by centrifugation ( 800xg , 5 min , 4°C ) ; half of the sample was used for protein quantification and the remainder was denatured by heating at 75°C for 10 min and stored at -80°C . Glucose was determined using a GAGO-20 kit ( Sigma ) according to the supplier’s protocol using spectrophotometric measurement at 540 nm . Glucose measurements probably also contained trehalose since this carbohydrate is usually present in flies , but we were not able to distinguish between glucose and trehalose in our measurements , most likely due to endogenous trehalase activity in homogenized samples . Glycogen samples were first treated with amyloglucosidase enzyme ( Sigma ) for 30 min . Protein concentration was analyzed by Bradford measurement . Samples were homogenized and proteins were dissolved in 1x PBS . A 100-μl volume of protein sample was mixed with 10 μl of Bradford solution ( 10 mg of Brilliant blue , 5 ml of 96% Ethanol , and 10 ml of 85% Phosphoric acid in 100 ml of solution ) . The concentration of proteins was derived from absorbance of the reaction solution at 595 nm . Values were compared by multiple unpaired t-tests using the Graphpad Prism software .
|
The immune response is an energy-demanding process and a sufficient energy supply is important for resistance to pathogens . However , the systemic metabolism must be tightly regulated during an immune response since nutrients may also be exploited by the pathogen and host energy reserves are limited . Here we present how host-pathogen interaction can be influenced by extracellular adenosine . We show that adenosine regulates the allocation of energy during bacterial infections in flies and that its signal is crucial for host immunity . Furthermore , enhancing its effect may even boost host immunity during the acute phase . However , the removal of adenosine by adenosine deaminase and thus down-regulation of its effect on the energy metabolism might prevent unintended feeding of the pathogen at the expense of host energy reserves . Therefore , our work demonstrates on the one hand that immune cells usurp energy from the rest of the organism , which is crucial for the effectivity of the immune response but , on the other hand , that immune cells also regulate adenosine to prevent the negative consequences of the excessive release of energy .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"glycosylamines",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pneumococcus",
"intracellular",
"pathogens",
"pathogens",
"cell",
"processes",
"microbiology",
"diet",
"nutrition",
"bacteria",
"bacterial",
"pathogens",
"adenosine",
"medical",
"microbiology",
"streptococcus",
"microbial",
"pathogens",
"phagocytosis",
"listeria",
"monocytogenes",
"biochemistry",
"cell",
"biology",
"glycogens",
"nucleosides",
"biology",
"and",
"life",
"sciences",
"glycobiology",
"organisms"
] |
2018
|
Extracellular adenosine modulates host-pathogen interactions through regulation of systemic metabolism during immune response in Drosophila
|
The effects of disease mutations on protein structure and function have been extensively investigated , and many predictors of the functional impact of single amino acid substitutions are publicly available . The majority of these predictors are based on protein structure and evolutionary conservation , following the assumption that disease mutations predominantly affect folded and conserved protein regions . However , the prevalence of the intrinsically disordered proteins ( IDPs ) and regions ( IDRs ) in the human proteome together with their lack of fixed structure and low sequence conservation raise a question about the impact of disease mutations in IDRs . Here , we investigate annotated missense disease mutations and show that 21 . 7% of them are located within such intrinsically disordered regions . We further demonstrate that 20% of disease mutations in IDRs cause local disorder-to-order transitions , which represents a 1 . 7–2 . 7 fold increase compared to annotated polymorphisms and neutral evolutionary substitutions , respectively . Secondary structure predictions show elevated rates of transition from helices and strands into loops and vice versa in the disease mutations dataset . Disease disorder-to-order mutations also influence predicted molecular recognition features ( MoRFs ) more often than the control mutations . The repertoire of disorder-to-order transition mutations is limited , with five most frequent mutations ( R→W , R→C , E→K , R→H , R→Q ) collectively accounting for 44% of all deleterious disorder-to-order transitions . As a proof of concept , we performed accelerated molecular dynamics simulations on a deleterious disorder-to-order transition mutation of tumor protein p63 and , in agreement with our predictions , observed an increased α-helical propensity of the region harboring the mutation . Our findings highlight the importance of mutations in IDRs and refine the traditional structure-centric view of disease mutations . The results of this study offer a new perspective on the role of mutations in disease , with implications for improving predictors of the functional impact of missense mutations .
Recent years have seen significant advancements in cataloging the genetic variation in humans and relating it to disease susceptibility . In particular , missense mutations , which introduce changes in the amino acid sequence of proteins , have been the subject of considerable attention due to the large number of ongoing exome sequencing studies . As a result , numerous computational models that classify amino acid substitutions as damaging or benign are currently available ( reviewed in [1] , [2] , [3] ) . The majority of these methods rely on the information from solved or modeled protein structures [4] , [5] , [6] , [7] , [8] , [9] and/or are based on evolutionary conservation , following the assumption that functionally important residues of proteins are conserved [10] , [11] , [12] , [13] . This choice of features limits the usefulness of current methods for classifying mutations in proteins that lack a fixed structure or have low sequence conservation , both of which are hallmarks of the intrinsically disordered proteins ( IDPs ) . Underestimating the impact of missense mutations in intrinsically disordered regions ( IDRs ) leads to a decrease in overall sensitivity of the existing methods . For example , it has recently been observed that SIFT predictions have more false negatives on annotated disease mutations in disordered , solvent accessible and non-conserved regions [14] . Intrinsically disordered proteins were first identified as a distinct class of proteins more than a decade ago [15] , [16] , [17] , [18] . It has since been clearly demonstrated that IDPs are prevalent in eukaryotic proteomes [19] , are involved in signaling and regulation [20] , [21] , carry sites of posttranslational modifications [22] , [23] , and serve as hubs in protein interaction networks [24] , [25] , [26] . Despite their important functional roles [27] , [28] , [29] , [30] , [31] , IDRs generally have low sequence conservation [32] , with the exception of IDRs involved in chaperone activity and RNA binding [33] . IDPs have been implicated in many human diseases , including cancer , diabetes , cardiovascular and neurodegenerative disorders [20] , [34] . Due to their signaling and regulatory roles , IDPs tend to be tightly regulated , and disruptions in regulation of IDPs have been linked to disease [35] . Despite the functional importance and disease relevance of IDPs , the prevalence of disease-associated missense mutations in disordered regions and their impact on disordered conformations have not been investigated so far . Here , we offer a new perspective on disease mutations that accounts for mutations in disordered regions . We investigate disease-associated mutations located in ordered and disordered regions , and compare them to missense mutations from two control datasets , single amino acid polymorphisms and neutral evolutionary substitutions . We demonstrate that deleterious missense mutations may affect disordered regions , thereby disrupting the disorder-based type of structure . Our results suggest that disease mutations in ordered regions ( ORs ) and IDRs differ substantially in frequency , properties , and functional impact . We find that disease mutations in disordered regions more frequently cause predicted disorder-to-order transitions and influence predicted disordered binding regions ( MoRFs ) compared to mutations from the control datasets . IDR mutations are also enriched in DNA-binding and transmembrane domains , and in sites of posttranslational modifications . Accelerated molecular dynamics simulations performed on a deleterious disorder-to-order transition mutation that affects the DNA-binding domain of tumor protein p63 support our disorder predictions . We further show that two widely used predictors of functional impact of single nucleotide variants , PolyPhen-2 and SIFT , exhibit a >10% decrease in sensitivity when predicting the effect of annotated disease mutations located in IDRs compared to ORs mutations . Our findings have broad implications for improving predictors of the functional impact of missense mutations and therefore may significantly influence the interpretation of novel variants identified in large genome sequencing projects .
We examined the frequency of annotated disease mutations ( DM ) from the UniProt database in predicted ordered and disordered regions and compared them to the distributions of putatively functionally neutral mutations from two control datasets , annotated polymorphisms from UniProt ( Poly ) and neutral evolutionary substitutions ( NES ) ( Materials and Methods ) . We observed that disease mutations preferentially affect ordered regions , with 78 . 3% of them mapped to the predicted ordered regions and 21 . 7% mapped to the predicted disordered regions ( Table 1 ) . Neutral evolutionary substitutions are more evenly distributed , with 55 . 3% observed in ORs and 44 . 7% in IDRs ( Table 1 ) . The annotated polymorphisms show somewhat intermediate distribution , with 59 . 6% in ORs and 40 . 4% in IDRs . Enrichment of disease mutations in ordered regions agrees with previous observations that disease mutations frequently affect protein structure , activity and stability [4] , [7] . Our results were consistent across three disorder predictors , VLXT [36] , VSL2B [37] and IUPRED [38] ( Table S1 ) . The enrichment of disease mutations in ORs cannot be explained by the overall lower disorder content of the proteins containing these mutations . Although proteins that carry disease-associated mutations are on average slightly less disordered than proteins from the Poly dataset ( mean±SD 32 . 7±17 . 9% vs 35 . 3±19 . 5% , respectively; also see Figure S1 ) , this difference is not sufficient to explain the 3 . 6 fold enrichment of disease mutations in ORs . Furthermore , despite the fact that the NES dataset was constructed from the same set of proteins as DM ( Materials and Methods ) , only a 1 . 2 fold enrichment of mutations in ORs compared to IDRs is observed in this dataset ( Table 1 ) , which lends further support to enrichment of disease mutations in ORs . Finally , we compared mutation rates ( number of amino acid changes per ordered and per disordered residue ) in ORs and IDRs in all three datasets , and only in the DM dataset the mutation rate in ORs was higher than the mutation rate in IDRs ( Table 2 ) . Despite the prevalence of disease mutations in ordered regions , 21 . 7% of DMs are mapped to the predicted disordered regions . We have investigated these mutations in greater detail , as discussed below , and mutations in IDRs form the main focus of the remainder of this study . Based on the predicted disorder probability score , a residue can be classified as ordered or disordered depending on whether its score is below or above a threshold of 0 . 5 . When analyzed from an order/disorder perspective , any missense mutation can have two different outcomes: ( i ) it can change the prediction score sufficiently to cross the 0 . 5 threshold , which would result in a conversion of the prediction from disorder to order , or from order to disorder; or ( ii ) it can preserve the order/disorder assignment . Thus , the effect of missense mutations can be classified as D→D ( disorder-to-disorder ) or O→O ( order-to-order ) when disorder and order assignments do not change; and as D→O ( disorder-to-order ) or O→D ( order-to-disorder ) transitions when predicted disorder and order classes switch . Disease mutations mapped to disordered regions cause D→O transitions significantly more frequently than neutral evolutionary substitutions or polymorphisms ( Table 2 ) . We observed that 20% of the disease mutations in disordered regions result in a D→O transition , compared to only 11 . 5% and 7 . 3% in the Poly and NES control sets ( Fisher's exact P = 1 . 06·10−32 and 5 . 47·10−105 , respectively ) . In contrast , the rates of O→D transition show no change or a slight depletion in DM compared to Poly and NES , respectively ( Table 2 ) . Similar results were obtained using three different disorder predictors ( Table S3 ) . These observations suggest that disease mutations in disordered regions are more likely to cause a significant structural perturbation , and possibly disrupt functions that necessitate protein disorder . Below , we examine the structural and functional implications of disease mutations in greater detail . To better understand how disease mutations influence protein secondary structure , we applied the secondary structure predictor PHD [39] to both the disease and control datasets . In each dataset , we calculated the frequencies of secondary structure elements ( helices , strands and loops ) and transitions between them upon a mutation . Overall , we observed that disease mutations affect helices and strands more frequently than control mutations ( Table S4 ) . We also observed that although most mutations do not cause a change in the assignment to a predicted helix , strand or loop , there is nevertheless a statistically significant increase in transitions between secondary structure elements caused by disease mutations compared to the control datasets ( Table S5 ) . This increase was most pronounced for transitions from helices and strands into loops , and to a lesser extent for transitions from loops into helices and strands ( Figure 1 ) . There was no significant difference between disease and control mutations for transitions from helix into strand and vice versa ( Figure 1 ) . Although similar trends are observed for loops predicted by PHD and disordered regions predicted by VLXT , VSL2B and IUPred ( see Figure 1 , Table 2 and Table S3 ) , it is important to note that predicted regions of disorder and loops do not necessarily overlap [40] , [41] , and that many secondary structure elements predicted by PHD are found within experimentally verified disordered regions [40] , [42] , [43] . Despite the lack of stable secondary and tertiary structure in disordered regions , the dynamic behavior of IDRs does not preclude formation of short transient secondary structure elements . These short transient elements , or Molecular Recognition Features ( MoRFs ) [44] , frequently mediate interactions of IDRs with their physiological binding partners [44] , [45] , [46] . Below , we investigated the influence of missense mutations on MoRFs . Molecular recognition features ( MoRFs ) are short order-prone segments within longer disordered regions that fold upon binding to their interaction partners [47] . α-MoRFs specifically form α-helices upon binding . We predicted the presence of α-MoRFs at the position of the residue both before and after it was mutated , and classified the mutation as falling into one of the three categories: ( i ) “predicted MoRF lost” - an α-MoRF was predicted to overlap the position of the mutated residue in the wild-type sequence but not in the mutant sequence; ( ii ) “predicted MoRF gained” - an α-MoRF was predicted not to overlap the position of the mutated residues in the wild-type sequence but was predicted to overlap the position of the mutated residue in the mutant sequence; ( iii ) “MoRF present , no change” - an α-MoRF was predicted to overlap the mutated position in both the wild-type and the mutant sequences . Mutations where an α-MoRF was absent from both wild-type and mutant sequences were not taken into account . Amino acid substitutions were placed into IDR and OR categories based on the wild-type disorder score . Details of MoRF predictions are provided in the Materials and Methods and in the Supplementary Text S1 . IDR mutations lead to gain or loss of predicted α-MoRFs 2 . 2 to 5 . 1 times more frequently than OR mutations , independent of the dataset used ( Figure S2 ) . Disease mutations in IDRs lead to a loss of predicted α-MoRFs 1 . 39 times more frequently than Poly and 1 . 36 times more frequently than NES ( Fisher's exact P = 0 . 0012 and 7 . 9·10−4 , respectively ) . Disease mutations in ORs have an opposite effect – they lead to a gain of predicted α-MoRFs 1 . 5-fold more frequently than Poly and 1 . 8-fold more frequently than NES ( P = 0 . 0020 and 1 . 65·10−5 ) . A follow up investigation showed that D→O and O→D mutations significantly contribute to the observed effect ( Figure 2 ) . Disease D→O mutations lead to a loss of predicted α-MoRFs 2 . 1-fold more frequently than Poly and NES ( P = 1 . 11·10−4 and 5 . 68·10−5 ) , and similarly disease O→D mutations lead to a gain of predicted α-MoRFs 1 . 7-fold more frequently than Poly and 2 . 0-fold more frequently than NES ( P = 0 . 025 and 0 . 0012 ) . We also examined the influence of disease and control mutations on Eukaryotic Linear Motifs ( ELMs ) , short ( 3 to 11 residues ) conserved sequence motifs that play roles in mediating cell signaling , controlling protein turnover and directing protein localization [48] . ELMs were previously shown to be enriched in IDRs [49] . We mapped mutations from the three datasets onto 1040 annotated ELM instances from human proteome ( see http://elm . eu . org/elms/browse_instances . html ) and found that only 99 mutations overlap an ELM . Although disease D→O mutations were slightly enriched in ELMs in comparison to control D→O mutations ( Table S6 ) , this difference reached statistical significance only for DM vs NES ( P = 0 . 012 ) , but not for DM vs Poly ( P = 0 . 22 ) , likely due to a limited number of observations . We did not observe any differences for other classes of mutations . Although a decisive conclusion about enrichment of D→O disease mutations within ELMs could not be made at this point , we believe that the trend towards such enrichment warrants further investigation when larger numbers of ELMs and annotated mutations become available . To characterize the functional impact of missense mutations , we examined UniProt region/residue feature annotations associated with each mutation ( Materials and Methods ) . A number of functional annotations for disease mutations in IDRs and ORs show significant differences in fold enrichment ( Figure 3 ) . Disease mutations in disordered regions are enriched in domains and functions associated with DNA binding motifs ( homeobox , zinc finger , basic motif ) , transmembrane domains , sites of post-translational modifications , disulfide bond formation , and triple helical regions , which are often found in cytoskeletal and coiled-coil proteins . Some of these functional categories were previously strongly associated with disordered regions [28] , [30] , and many DNA-binding domains are known to be either entirely or partially disordered when not associated with DNA [50] , [51] , [52] . Further investigation of keywords associated with D→O transitions shows an enrichment of functions similar to IDR , while O→D transition mutations show enrichment in ABC transporter and ATP-binding regions ( Tables S7 and S8 ) . In order to investigate mutations that contribute to the observed D→O and O→D transitions , we calculated the “wild-type residue→mutant residue” transition matrices in all three datasets and compared the differences in frequencies of D→O ( Figure 4 , first row ) and O→D ( Figure 4 , second row ) mutations between DM and Poly ( Figures 4A and 4C ) , and DM and NES ( Figures 4B and 4D ) . We observe that certain residue-into-residue substitutions are enriched ( red ) , while others are depleted ( green ) in disease . Arginine ( R ) is the most frequently mutated residue in the D→O dataset , and leucine ( L ) is most frequently mutated in the O→D dataset . The overall results do not depend on the choice of the control dataset ( Poly or NES ) . The heat plots in Figure 4 point to specific mutations that are highly enriched in disease . The most frequent disease mutation that causes a disorder-to-order transition is R→W ( Figure 4C ) . Other D→O transition mutations significantly enriched in the DM dataset include most notably R→C , R→H , E→K , R→Q ( Figure 4E , left section ) . Several other types of disorder-to-order transition mutations , such as R→K , E→D , L→F , S→T , are significantly depleted in the DM dataset ( Figure 4E , right section ) , which demonstrates that distinct types of mutations preferentially occur within disease and control categories . To verify that this result is not an artifact of our analysis , for example due to general enrichment of R→W mutations in disordered regions , or the choice of control datasets , we have compared the frequencies of R→W substitutions from this study to the matrices constructed based on the alignments of completely disordered sequences [53] . This comparison showed that in general R→W substitutions occur extremely rarely in disordered regions ( with 0 . 11% in D85 matrix and 0 . 03% in D40 matrix ) , whereas we find R→W substitution with much higher frequency in our datasets ( 11 . 69% in DM , 6 . 52% in Poly , and 0 . 95% in NES ) . This result suggests that the R→W mutation is truly enriched among disease mutations . Another category of amino-acid substitutions in DM , albeit not significantly enriched as a group , involve order-to-disorder mutations , such as L→P , C→R , G→R , W→R and others ( Figure 4F ) . Some of the enriched order-to-disorder mutations are inverses of the enriched disorder-to-order mutations , such as W→R , C→R , L→R , whereas some are shared between D→O and O→D , such as G→E . This shared category points to the fact that there is no strong preference for glycine and glutamic acid to be located in either ordered or disordered regions , as reflected by the presence of both residues in the middle of the TOP-IDP scale of residue disorder propensities [54] . In summary , our analysis shows that a limited set of mutations accounts for a large fraction of all D→O and O→D transitions in the DM dataset . The top five disorder-to-order transition mutations ( R→W , R→C , E→K , R→H and R→Q ) collectively account for 44 . 0% of all D→O disease mutations , and the top five order-to-disorder transition mutations ( L→P , C→R , G→R , W→R and F→S ) collectively account for 32 . 2% of all O→D disease mutations ( Figure 4C and 4F ) . Specific knowledge of the mutations responsible for such transitions may help the development of new classifiers to better predict the effects of mutations in IDRs . We next compared the frequencies of wild-type and mutant residues in all datasets to the frequencies of typical human proteins from the UniProt database ( Figure S3 ) . Mutations of arginine and glycine are most dominant in DM and account for 28 . 5% of all disease mutations , 18 . 6% of all Poly and only 11 . 1% of NES mutations ( Figure S3B ) . After normalizing by the baseline residue frequency [55] ( Figure S3A ) , mutations of cysteine and tryptophane stood out , reflecting that in DM these two resides are mutated significantly above what is expected based on their frequency of occurrence in the human proteome . Interestingly , tryptophane and cysteine , and to a lesser degree histidine , are the residues into which other residues most frequently mutate , with a more pronounced effect in IDRs than in ORs ( Figure S4 ) . High mutability of arginine , also observed in earlier studies [56] , [57] , together with the high propensity of arginine mutations to cause disorder-to-order transitions suggest an underlying mechanism which predisposes arginine to be a frequent target for disease mutations . Arginine is encoded by 6 distinct codons , 4 of which contain the CG dinucleotide ( CGG , CGT , CGC and CGA ) . DNA methylation often involves CpG dinucleotides and due to spontaneous deamination 5-methylcytosine is more prone to mutating into T . Upon a C-to-T transition , the first three arginine codons would become codons for W ( TGG ) or C ( TGT , TGC ) , and the last one would create a stop codon ( TGA ) . The observed high frequency of R→W and R→C in DM and low frequency in control datasets ( Figure S5 ) argues in favor of negative selection against these amino acid substitutions , which frequently cause predicted disorder-to-order transitions in proteins . A recent study demonstrated that SIFT has a higher error rate when predicting the impact of SNVs in solvent accessible and disordered protein regions [14] . In order to rigorously evaluate this statement , SIFT [10] and PolyPhen-2 [58] were applied to all mutations in DM , Poly and NES datasets , and the prediction accuracies on mutations in different order/disorder categories were compared ( Figure 5 and Table S9 ) . Both SIFT and PolyPhen-2 predict significantly less disease mutations as deleterious in IDRs than in ORs ( SIFT “damaging” 64 . 3% vs 74 . 4% , χ2 P = 4 . 19·10−28; PolyPhen-2 “probably damaging” 60 . 8% vs 74 . 9% , P = 8 . 05·10−74 ) . SIFT and PolyPhen-2 both predict significantly more polymorphisms to be benign in IDRs than in ORs ( SIFT “tolerated” 78 . 7% vs 73 . 5% , P = 1 . 86 ·10−18; PolyPhen-2 “benign” 55 . 5% vs 53 . 7% , P = 3 . 74 ·10−64 ) , and likewise for neutral evolutionary substitutions ( SIFT 91 . 6% vs 87 . 9% , P = 7 . 38·10−45 , PolyPhen-2 80 . 2% vs 75 . 6% , P = 2 . 84·10−215 ) . IDR mutations seem to be more difficult to handle for the PolyPhen-2 model in general , and in all three datasets more IDR than OR mutations are returned as “unknown” ( DM 4 . 4% vs 1 . 2% , Poly 7 . 5% vs 3 . 4% , NES 8 . 7% vs 2 . 4% ) . Upon closer examination , we determined that among DM mutations , the D→D transition category was the most difficult to predict correctly for both predictors , while the D→O category was most often correctly predicted as deleterious . However , in the case of D→O mutations , higher sensitivity comes at the expense of lower specificity , and significantly more mutations from Poly and NES are predicted as deleterious in D→O transitions than in any other category ( Table S9 ) . Similar results were obtained by analyzing raw PolyPhen-2 and SIFT scores ( Figures S6 and S7 ) . Notably , the DM dataset investigated here overlaps with the predictors' training sets , and the reported accuracies are likely to be lower when applied to out-of-training set examples . In summary , our findings underscore the need for incorporating features of IDRs into predictive disease mutation models . We observed 670 mutations in UniProt predicted to cause D→O transitions , and 590 mutations predicted to cause O→D transitions ( Tables S10 and S11 ) . We note that the number of such examples would be higher if extensively studied proteins with an excessively large number of mutations ( such as p53 , androgen receptor , etc . ) were included in the analysis ( Materials and Methods and Figure S8 ) . In addition to disease mutations mapped to predicted disordered regions , we elsewhere summarized D→O disease mutations found in the experimentally ascertained disordered regions from the DisProt database [59] . Below , we show an example of a protein carrying predicted D→O disease mutation ( Figure 6 ) . Tumor protein p63 ( TP63 ) is a transcription factor involved in development and morphogenesis of epithelial tissues [60] , [61] . The sequence , structure and domain organization of p63 are highly similar to tumor suppressor protein p53 , with the exception of two additional domains at p63 C-terminus , which are alternatively spliced in some p63 isoforms . More than 30 distinct missense mutations have been identified in p63 and associated with several malformation genetic syndromes such as ectrodactyly ectodermal dysplasia-cleft syndrome 3 ( EEC3 , MIM: 604292 ) , split hand/foot malformation-4 ( SHFM4 , MIM: 605289 ) , and nonsyndromic cleft lip ( NSCL , MIM: 129400 ) . Most of the mutations that cause EEC3 occur within the DNA-binding domain of p63 [62] . One of these mutations , R243W , is predicted to cause a D→O transition , shown in Figure 6A as a sharp drop in disorder score of the 235–245 region ( red dotted line ) after R243 has been in silico mutated to W . Since R243 is not directly involved in binding to DNA , the mutations affecting this residue are predicted to destabilize the protein as a result of hydrogen bond loss and overpacking [63] . DNA-binding domains of transcription factors tend to be predicted as fully or partially disordered [64] , [65] , and binding to DNA typically induces a D→O transition [66] . In agreement with these observations , only a single NMR structure of p63 DBD without DNA ( PBD: 2RMN ) is available , while all X-ray structures of p63 DBD found in PDB ( PDB: 3US0 , 3US1 , 3US2 , 3QYM and 3QYN ) have been crystallized in complex with DNA . Residue R243 is located in the modeled turn region of the NMR structure , adjacent to a short α-helix . We investigated the effects of the R243W mutation on p63 DBD conformation using an extensive set of accelerated molecular dynamics ( AMD ) simulations [67] , [68] on both the wild-type p63 ( wt-p63 ) and its R243W mutant . AMD is an efficient and versatile enhanced conformational space sampling algorithm that has previously been successfully applied to the study of the conformational behavior of IDPs [69] , [70] . A comparative analysis of a series of AMD trajectories for wt-p63 and its R243W mutant revealed no significant differences in the global structural dynamics of the p63 DBD . However , marked differences in the conformational behavior of residues adjacent to R243W were observed ( Figure 6B ) . The introduction of R243W mutation caused a significant increase in the free energy weighted φ/ψ propensity of the α-helical/frustrated α-helical conformation of these residues , resulting α-helical population statistics of 70–90% and 30%–50% in the R243W mutant for residues 236–240 and 241–243 respectively , compared to 20%–60% and 20–25% in the wild-type system ( Table S12 ) . The formation of an ostensibly exclusive ( frustrated ) α-helical coil in this region in the presence of the R243W mutation is fully consistent with the predicted D→O transition ( Figure 6A ) . It is interesting to note that in both the experimental NMR structure and the AMD simulations for wt-p63 the side-chain of R243 forms a strong salt-bridge with E252 . One may postulate that in the wild-type system the strong electrostatic interaction between R243 and E252 introduces tensile stress in the extended loop region K232-R243 , which exhibits conformational exchange on slow time-scales between local extended β-sheet/PPII and α-helical constructs . By contrast , the introduction of the R243W mutation removes the tensile strain from the loop facilitating the formation of a stable α-helix .
The widely accepted structure-centric view of deleterious mutations asserts that a disease may be caused by mutations disrupting protein activity , stability , oligomerization and other structure-based properties . Here , we further extend this concept by introducing a disorder-centric view of disease mutations , according to which a disease may arise due to a disruption of the disorder-based protein properties [59] . We have demonstrated that a substantial fraction of disease-associated mutations are located within the intrinsically disordered protein regions , and that disease mutations in IDRs have a significant functional impact despite the fact that IDRs lack fixed structure and have fewer evolutionary constraints than ORs [32] . The analysis of mutations in IDRs shows that disorder-to-order transition mutations may be especially relevant to disease due to their enrichment compared to control datasets . In addition , our analysis suggests that several types of disease mutations may have particularly critical impact on disordered structure . There are many ways in which mutations in IDR may increase disease risk or cause a disease . For example , D→O mutations have a potential to alter interactions with DNA , RNA , proteins or ligands . Both , our results and those of a recent study by Dan et al . [71] , which examined transitions between disorder and secondary structure in proteins with solved 3D structures , converged on the observation that disorder-to-order ( i . e . disorder-to-secondary structure in [71] ) transitions are significantly enriched in DNA binding proteins . In addition , mutations in IDR could influence posttranslational modifications , assembly of macromolecular complexes , as well as signaling and regulatory processes that depend on disorder . Adding support to this hypothesis is an observation that disease mutations often disrupt anchoring of flexible loops of the catalytic domains in protein kinases , and that mutated residues are frequently involved in substrate binding and regulation [72] . This also suggests a potential downstream effect of mutations in IDR via dysregulation of cellular pathways which could lead to disease [59] . Our results show that across all three datasets , mutations in IDR are more likely to cause a predicted D→O transition than mutations in ORs are to cause a predicted O→D transition ( Table 2 ) . This is in agreement with a recent study by Schaefer et al . [73] , which showed that disordered regions are more sensitive to mutations than protein regions with defined secondary structure , with a caveat that “order” and “helix or strand” cannot be fully equated . Despite a significant enrichment of D→O mutations in disease , the majority of disease mutations in IDR do not result in a disorder-to-order transition ( as defined in this paper ) but they nonetheless sufficiently disrupt the disordered conformation to affect disorder-mediated functions . It is likely however that many other mutations that do not reach the disorder-to-order transition threshold may still disrupt the structure and consequently function of the disordered regions . Our findings have wide implications for large genome sequencing projects that aim to provide a better understanding of human genetic variation and its relevance to complex diseases [1] . Because the sheer volume of the observed variants precludes systematic functional follow-up studies on each one individually , newly identified SNVs are short-listed and prioritized using predictors of the functional impact of SNVs , such as SIFT , PolyPhen-2 and others [2] . The majority of the currently used predictors are structure- and/or conservation-based , and therefore less accurate on variants in unstructured and non-conserved protein regions . Disorder predictions could be either integrated into current approaches , or new approaches , which analyze the features of mutations in ORs and IDRs separately , could be developed . In addition , in this study we demonstrate that specific types of mutations ( such as R→W , R→C , etc . ) account for almost one half of all D→O transitions ( Figure 4 ) . This additional information may be important to include as a training feature when developing new predictors for the effects of D→O SNVs . A broader issue raised by our results is that caution should be exercised when interpreting the relationship between structure , function and conservation . A study by Yue and Moult found that human disease-relevant mutations in some cases could correspond to the wild-type variants in the mouse [11] . Compensatory mutations [74] illustrate that function cannot be fully equated with the “first order conservation” , and that sometimes co-evolution of amino acids constrained by protein structure necessitates looking into the “second order conservation” between pairs of residues . Our results are consistent with the fact that IDRs are less conserved at any individual position , but rather show a conservation of disorder propensity within a region [75] , with D→O transition mutations – detrimental to conservation of disorder – being particularly enriched in disease . Choosing an appropriate control for the analysis of disease mutations is an issue which deserves close attention [76] . One of our control datasets , polymorphisms from UniProt ( Poly ) , is likely to contain a fraction of as yet unannotated disease mutations , because it was assembled by translating missense single nucleotide variants ( currently without any disease associations ) into single amino acid changes [76] , [77] . This is further supported by the predictive result that between 20% [7] and 25% [11] of non-synonymous SNPs are likely to be associated with diseases . Nonetheless , Poly controls for an important previously identified confounder: because disease missense mutations are translations of a single nucleotide variation within a DNA codon , a genetically appropriate control has to be analogously constrained by the genetic code , that is , assembled from amino acid changes which are translations of functionally neutral SNVs [57] , [76] . In the protein space , another concern is that length distribution and amino acid compositions of proteins from DM and Poly datasets differ , which may influence their baseline biochemical properties , including disorder content ( Figure S1 ) . In order to address this potential confounder , the second control ( NES ) was generated starting with the sequences of proteins from the DM dataset . The downside of this approach is that the set of disease mutations spans within-population differences , while changes in the orthologs span larger , inter-species distances . In practice , this means that in DM and Poly the mutation probability matrix is dominated by the effects of the genetic code , while in DM and NES it is dominated by effects of physico-chemical similarity between amino acids . Nonetheless , variants fixed between species are likely to be non-deleterious ( even though about 9% of interspecies substitutions have been estimated to be damaging [7] ) , and therefore they provide a useful additional control that takes into account sequence conservation . In the light of advantages and shortcomings of different control datasets , it is reassuring to see that when using either Poly or NES , protein disorder-related properties ( Tables 1 and 2 ) and WT-to-mutant amino acid changes ( Figure 4 ) are consistent and independent of the control dataset used . In addition , the preponderance of annotated mutations within OR might show some degree of ascertainment bias since some disease mutations were annotated as “disease” because they were mapped to protein structured domains . We hypothesize that an unbiased sample would contain a higher proportion of disease mutations that map to IDRs . In summary , our results refine the traditional structure-centric view of disease mutations , and suggest new avenues for research in the area of protein disorder . With the recent explosion of exome and whole genome sequencing efforts , interpretation of the identified variants will require highly accurate predictors for the functional impact of SNVs in order to make reliable conclusions about their health risks . Our results offer help in narrowing down the gamut of disease mutations that dramatically influence protein structure and disorder . We hope that it will also facilitate predictions of the influence of mutations on protein function , which is currently a formidable task . The importance of mutations in disordered regions should not be overlooked in an attempt to construct better predictors .
A list of single amino acid substitutions annotated with the keyword “disease” was extracted from the UniProt/SwissProt database [77] . This manually curated catalog contains missense mutations associated with both Mendelian and complex diseases , but no nonsense nor frame shift mutations , and no products of alternative splicing . The initial set of mutations was filtered as follows: proteins that carry disease mutations and have ≥40% pairwise sequence identity were clustered using hierarchical clustering with single linkage , and one representative protein was selected at random from each cluster . We further removed four proteins with an unusually high number of annotated disease mutations ( Figure S8A ) : tumor suppressor p53 ( P04637 ) , coagulation factor VIII ( P00451 ) , androgen receptor ( P10275 ) , and Stargardt disease protein ( P78363 ) . Taken together , these four proteins account for a total of 12 . 4% of all disease mutations found in the non-redundant set of proteins . All mutations from the removed proteins were discarded . We assembled two control datasets: ( 1 ) annotated single amino acid polymorphisms from UniProt ( Poly ) [77] and ( 2 ) a set of pseudo-mutations based on amino acid variation in mammalian orthologous proteins ( neutral evolutionary substitutions , NES ) . The first control dataset ( Poly ) was filtered analogously to disease mutations , and redundant proteins and titin ( with unusually high number of polymorphisms ) were removed ( Figure S8B ) . The second control dataset ( NES ) ( Figure S8C ) was constructed following the approach of Sunyaev et al . [78] . Proteins that carry disease mutations which also passed our filtering criteria were aligned by the use of multiple sequence alignment program MUSCLE [79] against their InParanoid [80] orthologs from 10 mammalian species ( P . troglodytes , P . pygmaeus abelii , M . musculus , M . mulatta , C . familiaris , E . caballus , R . norvegicus , C . porcellus , B . taurus , and M . domestica ) , using the BLOSUM85 matrix . The set of neutral evolutionary substitutions ( NES ) was assembled from all single amino acid differences in orthologous proteins that had ≥95% sequence identity with the human disease protein . Finally , all annotated disease mutations were filtered out from the NES dataset . The numbers of proteins and mutations in the three datasets are summarized in Table 1 . Protein disorder was predicted using VLXT [36] , VSL2B [37] and IUPRED [38] . Disorder predictions were carried out on full length wild-type ( WT ) and mutated protein sequences , generated by changing only one residue at a time . Disorder score <0 . 5 signified predicted order and ≥0 . 5 signified predicted disorder . We defined the effect of a mutation as a disorder-to-order ( D→O ) transition if the prediction score for a residue to be mutated was ≥0 . 5 in the WT protein , and <0 . 5 after the mutation . Order-to-disorder ( O→D ) transitions were analogously defined . The enrichment/depletion trends for D→O and O→D transitions are consistent across all three predictors ( Tables S1 and S3 ) . As a second comparison of disorder predictors , we examined the distributions of the difference between disorder prediction scores on WT and mutated sequences , defined as Δps = ps ( WT residue ) −ps ( mutated ) . The three predictors have different observed dynamic ranges for Δps: [−0 . 91 , 0 . 85] for VLXT , [−0 . 34 , 0 . 39] for VSL2B and [−0 . 28 , 0 . 27] for IUPRED , consistent with the fact that VLXT is more sensitive to small changes in amino acid sequence . Distribution of Δps is more platykurtic in DM compared to Poly and NES for all three predictors ( higher % of disease-associated mutations in the tails ) , indicating that disease mutations tend to cause stronger differences in prediction scores . Secondary structure was predicted from sequence using PHDsec [81] . We used only reliable predictions , defined as having both a “from” and “to” secondary structure assignment score ≥4 . We note , however , that the trend was the same when all secondary structure predictions were used without thresholding on the reliability score . α-MoRFs were predicted from sequence using a two stage stacked prediction method [44] . The first stage identified potential α-MoRF regions from PONDR VLXT [36] predictions by scanning for short predicted ordered regions flanked by predicted disordered regions . The second stage classified potential α-MoRF regions as either α-MoRFs or non-α-MoRFs using a quadratic discrimination model [44] . Further details of α-MoRF predictions are provided in the Supplementary Text S1 . Residues were functionally annotated using the UniProt/SwissProt feature table ( FT ) at two levels of granularity , the FT keywords only ( level 1 ) and concatenations of the FT keyword and description ( level 2 ) . Features marked as “Potential” , ”Probable” or “By similarity” were removed . The “Description” field was normalized by removing prefixes such as “For” , “Required for” , “Sufficient for” , “Essential for” , “Essential to” , “Important for” , “Critical for” , “Necessary for” , “Involved in” , “Mediates” , etc . Finally , all features that occurred <5 times in DM were removed . After this process , 22 level 1 and 782 level 2 features remained . We removed all disease keywords from this analysis , since they would be trivially enriched in the DM dataset . Standard classical and accelerated molecular dynamics simulations were performed on both wild-type and R243W p63 mutant using an in-house modified version of the AMBER-10 simulation suite [82] . The reader is referred to the Supplementary Information ( Supplementary Text S2 ) for a description of the accelerated molecular dynamics method and computational details .
|
Intrinsically unstructured or disordered proteins have been implicated in the etiology of a wide spectrum of diseases . However , the molecular mechanisms that relate mutations in intrinsically disordered regions ( IDRs ) to disease pathogenesis have not been investigated . Disordered proteins do not conform to the prevailing view of deleterious mutations which equates function , structure and evolutionary conservation – intrinsically disordered regions are functional , but lack a fixed three-dimensional structure and in general have low sequence conservation . Here we demonstrate that >20% of disease-associated missense mutations affect IDRs and interfere with their functions . We further show that 20% of deleterious mutations in IDRs induce predicted disorder-to-order transitions . Our predictions are supported by accelerated molecular dynamics simulations that show an increase in helical propensity of the region harboring a disease disorder-to-order transition mutation of tumor protein p63 . Our results refine the traditional structure-centric view of disease mutations and offer a new perspective on the role of non-synonymous mutations in disease . Our findings have broad implications for improving predictors of the functional impact of missense mutations , and for interpretation of novel variants identified in large genome sequencing projects that aim to provide a better understanding of human genetic variation and its relevance to common diseases .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"sequence",
"analysis",
"protein",
"folding",
"biology",
"computational",
"biology",
"biophysics"
] |
2012
|
Disease-Associated Mutations Disrupt Functionally Important Regions of Intrinsic Protein Disorder
|
The homologous p10 fusion-associated small transmembrane ( FAST ) proteins of the avian ( ARV ) and Nelson Bay ( NBV ) reoviruses are the smallest known viral membrane fusion proteins , and are virulence determinants of the fusogenic reoviruses . The small size of FAST proteins is incompatible with the paradigmatic membrane fusion pathway proposed for enveloped viral fusion proteins . Understanding how these diminutive viral fusogens mediate the complex process of membrane fusion is therefore of considerable interest , from both the pathogenesis and mechanism-of-action perspectives . Using chimeric ARV/NBV p10 constructs , the 36–40-residue ectodomain was identified as the major determinant of the differing fusion efficiencies of these homologous p10 proteins . Extensive mutagenic analysis determined the ectodomain comprises two distinct , essential functional motifs . Syncytiogenesis assays , thiol-specific surface biotinylation , and liposome lipid mixing assays identified an ∼25-residue , N-terminal motif that dictates formation of a cystine loop fusion peptide in both ARV and NBV p10 . Surface immunofluorescence staining , FRET analysis and cholesterol depletion/repletion studies determined the cystine loop motif is connected through a two-residue linker to a 13-residue membrane-proximal ectodomain region ( MPER ) . The MPER constitutes a second , independent motif governing reversible , cholesterol-dependent assembly of p10 multimers in the plasma membrane . Results further indicate that: ( 1 ) ARV and NBV homomultimers segregate to distinct , cholesterol-dependent microdomains in the plasma membrane; ( 2 ) p10 homomultimerization and cholesterol-dependent microdomain localization are co-dependent; and ( 3 ) the four juxtamembrane MPER residues present in the multimerization motif dictate species-specific microdomain association and homomultimerization . The p10 ectodomain therefore constitutes a remarkably compact , multifunctional fusion module that directs syncytiogenic efficiency and species-specific assembly of p10 homomultimers into cholesterol-dependent fusion platforms in the plasma membrane .
More than two new species of human-infecting viruses are reported worldwide each year , and these novel viruses pose a substantial threat of becoming emerging human infections [1] . Bat populations may serve as a major reservoir for emerging human viral infections [2]–[5] . Genus Orthoreovirus is one of 15 recognized genera in Family Reoviridae , a large , diverse group of non-enveloped viruses with segmented , double-stranded RNA genomes [6] . Several novel orthoreoviruses of pteropine origin have recently been isolated , and six of these isolates were associated with acute respiratory infections in humans [7]–[11] . Moreover , a serological survey revealed 13% of 109 random serum samples of subjects in Malaysia were seropositive for pteropine reoviruses [12] , indicating zoonotic transmission of fusogenic bat reoviruses may be a frequent event . Monitoring agencies have recommended close monitoring of orthoreovirus evolution , as well as a necessity for further research to allow for prevention and control measures to be taken in the event of zoonotic transmission [13] . Genus Orthoreovirus is divided into two subgroups , fusogenic and non-fusogenic orthoreoviruses , based on the ability to induce cell-cell fusion and syncytium formation [14] . Human infections with non-fusogenic mammalian reoviruses ( MRVs ) are commonplace and generally asymptomatic [15] . In contrast , natural infections with fusogenic orthoreoviruses induce an array of pathologies in virus-infected animals; baboon reovirus ( BRV ) is associated with meningoencephalomyelitis in juvenile baboons [16] , reptilian reoviruses ( RRVs ) cause pneumonia and neurological dysfunction [17] , [18] , and avian reoviruses ( ARVs ) are the causative agents of a wide range of diseases that include arthritis and enteric disease syndromes [19] . Syncytium formation is mediated by non-structural , fusion-associated small transmembrane ( FAST ) proteins encoded by all fusogenic orthoreoviruses , and by fusogenic aquareoviruses in the closely related Genus Aquareovirus [20] . These are the only known membrane fusion proteins encoded by non-enveloped viruses . The FAST proteins are not involved in viral entry , but instead induce fusion between virus-infected and neighboring uninfected cells , thereby promoting virus dissemination [21] . The extent of syncytiogenesis correlates with viral pathogenicity [22] , and mice intranasally infected with a chimeric , FAST protein-expressing vesicular stomatitis virus ( VSV ) showed increased neuropathology [23] . The FAST proteins are therefore virulence determinants of the fusogenic reoviruses . There are six current members of the FAST protein family , each named according to their molecular mass: the orthoreovirus FAST proteins , p10 , p13 , p14 and p15 are encoded by ARV and Nelson Bay reovirus ( NBV ) , Broome reovirus ( BroV ) , RRV , and BRV , respectively [24]–[27] , while fusogenic aquareoviruses encode p16 and p22 FAST proteins [28] , [29] . Recent phylogenetic analysis suggests the fusogenic reoviruses arose from an ancestral , non-fusogenic virus by at least two separate gain-of-function events , possibly involving divergent evolution from one or more precursor non-structural proteins that were membrane-interactive , but non-fusogenic , [20] , [30] . The FAST proteins share no identifiable sequence similarity with each other , or with any other known membrane fusion protein , and each possesses a unique repertoire and arrangement of functional motifs . Defining attributes of family members include their small size ( 95–198 amino acids ) , presence of a single-pass transmembrane domain ( TMD ) with Nexoplasmic/Cendoplasmic membrane topology resulting in ectodomains of <40 residues , sites for fatty-acid modification ( N-terminal myristoylation or palmitoylation of membrane proximal Cys residues ) , a cluster of membrane-proximal basic residues on the cytoplasmic side of the TMD , a short hydrophobic or amphipathic motif that can be located on either side of the TMD , and an intrinsically disordered cytoplasmic tail [25] , [27] , [31]–[33] . Determining how these motifs function in a coordinated manner to mediate membrane fusion is critical to our understanding of how these unusual viral fusion proteins mediate syncytiogenesis , and how this process influences viral pathogenesis . Studies of enveloped viral fusion proteins suggest that all membrane fusion reactions share a common pathway progressing through membrane binding , close membrane apposition , outer-leaflet mixing , and pore formation/expansion [34] . Although enveloped virus fusion proteins possess considerable architectural diversity , remarkable similarities exist in the dynamic structural rearrangements undertaken throughout the fusion reaction [35] . Triggered conformational changes result in extensive structural remodeling of complex , multimeric ectodomains that expose and project hydrophobic fusion peptides ( FPs ) toward the target cell membrane , followed by folding back of the extended intermediate structure into a compact trimeric hairpin structure . FPs are critical for fusion activity , they are extremely sensitive to mutation , and can exist as either internal fusion loops or as N-terminal α-helices [36] . Membrane merger is believed to be mediated by bilayer deformation , using the energy released as the metastable , pre-fusion structure transitions to the lowest energy , post-fusion conformation and/or by membrane curvature effects induced by FP partitioning into bilayers [34] , [37] . Membrane-proximal external regions ( MPERs ) of several enveloped viral fusion proteins have also been shown to play active roles in the membrane fusion reaction [38]–[41] . These short , hydrophobic , frequently tryptophan-rich sequences partition into lipid bilayers where they may influence bilayer stabilization or curvature [42] . Despite extensive study of numerous enveloped virus fusogens , a clear picture on the specific role of FPs , MPERS , and quaternary structural transitions in the fusion reaction has not emerged . At 95–98 residues , the p10 FAST proteins of ARV and NBV are the smallest known protein fusogens and provide a simple system for investigating the mechanism of protein-mediated membrane fusion . As nonstructural proteins involved in cell-cell fusion , not virus-cell fusion , the relationship between syncytium formation and viral pathogenesis can also be analyzed in the absence of confounding virus entry effects . The p10 proteins are the only FAST proteins with numerous homologs , which segregate into two host-specific clades ( Figure 1 ) ; an avian clade and the newly emerged pteropine reoviruses associated with severe human respiratory infections . The rate of syncytiogenesis varies considerably between representatives of the avian and pteropine clades [21] . There is extensive sequence divergence between p10 clades ( <30% amino acid identity ) , but they share an identical arrangement of structural motifs ( Figure 1 ) . The ARV and NBV p10 FAST proteins comprise a central transmembrane domain flanked by approximately equal sized ( ∼40 residue ) ecto- and endodomains ( Figure 1 ) . The ectodomain contains a moderately hydrophobic region , termed the hydrophobic patch ( HP ) , flanked by two conserved Cys residues , which were shown to form an intramolecular disulfide bond , creating an 11-residue cystine loop FP [43] , [44] . The ectodomain also contains a 9-residue conserved motif ( CM ) of unknown function that is invariant in all p10 isolates . The endodomain is the most diverged region of p10 , but contains two conserved Cys residues immediately adjacent to the TMD , which in ARV p10 were shown to be palmitoylated [45] , followed by a short region of moderately conserved basic residues known as the polybasic ( PB ) motif ( Figure 1 ) ; both of these endodomain motifs are required for syncytium formation [45] . To date , relatively little attention has been given to bat reovirus p10 proteins , or to exploiting comparative approaches using the two p10 clades to better define structure-function relationships . Using ARV strain 176 ( hereinafter referred to as ARV ) and NBV p10 proteins as representatives of the two clades , we used chimeric p10 constructs to identify the ectodomain as the principal determinant of the differing syncytiogenic efficiencies of these two p10 proteins . Further fine structure mapping revealed the 40-residue ectodomain contains two distinct functional motifs , both of which are essential for syncytiogenesis: an N-terminal ∼25-residue motif governing formation of a cystine loop FP , and a C-terminal 13-residue membrane-proximal ectodomain region ( MPER ) directing p10 multimerization and clustering into plasma membrane microdomains . Cholesterol-dependent clustering and multimerization are co-dependent , reversible , and species-specific , with the four membrane-adjacent MPER residues dictating homomultimer specificity .
To confirm previous qualitative assessments of ARV and NBV p10 syncytiogenic activity [46] , these two representatives of the avian and pteropine p10 lineages were transfected into QM5 fibroblasts and Vero epithelial cells . A quantitative syncytiogenesis assay confirmed NBV p10-induced syncytium formation is markedly faster than ARV p10 in both cell types , with NBV p10 inducing ∼3–4-fold higher levels of cell-cell fusion than ARV p10 by 24 h post-transfection ( Figure S1A and B ) . Western blotting and FACS cell surface analysis using α-FLAG antibodies and FLAG-tagged p10 constructs indicated differing syncytiogenic kinetics were not due to differences in ARV and NBV p10 expression ( Figure S1C ) . We exploited the differing fusogenic activities of the ARV and NBV p10 proteins to define functional motifs that confer enhanced fusion capability . Using sequential PCR reactions , the homologous ectodomains , TMDs and endodomains of ARV and NBV p10 were exchanged to generate six chimeric p10 proteins ( Figure 2A ) . Chimeras were named to reflect the proportion of domains stemming from each parental p10 protein . For example , the chimeric ARV protein containing the NBV ectodomain is termed ARVectoNBV ( abbreviated AectoN ) . FACS analysis and western blotting indicated no significant differences in cell surface and total expression levels of FLAG-tagged versions of the chimeric proteins ( Figure 2B ) . Quantitative syncytiogenesis assays revealed the relative contribution of each domain to the overall syncytiogenic rate ( Figure 2C ) . Exchanging the TMDs or endodomains of ARV and NBV p10 had no effect on the rate or extent of syncytium formation; AtmN and AendoN had identical syncytiogenic kinetics as parental ARV p10 ( Figure 2C , bottom two panels ) . The same situation applied for NBV p10 containing an ARV p10 TMD or endodomain ( i . e . NtmA and NendoA chimeras ) , both of which shared the fusogenic activity of the parental NBV p10 ( Figure 2C , bottom two panels ) . In contrast , an NBV ectodomain substantially increased the syncytiogenic activity of ARV p10 while an ARV ectodomain impaired the fusogenic activity of NBV p10 ( Figure 2C , top panel ) . Interestingly , the two ectodomain chimeras had intermediate syncytiogenic activities relative to the parental p10 proteins , indicating the ectodomains directly contribute to the differing fusion activities of ARV and NBV p10 but require a homotypic TMD or endodomain to confer the full fusion phenotype of the parental proteins . ARV and NBV p10 ectodomain peptides ( 40 and 36 residues , respectively ) were chemically synthesized to include the intramolecular disulfide bond ( confirmed by mass spec analysis ) , which is required for ARV p10 FP activity [43] , and assessed for lipid mixing activity using a commonly employed fluorescent liposome-based assay [44] , [47] . Both ectodomain peptides displayed robust , dose-dependent lipid mixing activity , with the NBV peptide inducing approximately twice the extent of lipid mixing as the ARV peptide in a dose-dependent manner ( Figure 2D ) . The ectodomain is therefore the primary determinant of species-specific p10 syncytiogenic efficiency , a property that correlates with relative FP-induced lipid mixing activities . To further define motifs within the ectodomain that contribute to fusion potential , six additional constructs were created that exchanged portions of the ARV and NBV p10 ectodomains ( Figure 3A ) . These constructs exchanged sequences bracketing the cystine loop ( constructs A1 , A3 , N1 , and N3 ) , the sequences within the Cys-flanked HP ( constructs A2 and N2 ) , the non-conserved residue adjacent to the N-terminus of the CM ( constructs A4 and N4 ) , and the four membrane-proximal residues ( constructs A5 and N5 ) . Quantitative syncytiogenesis assays revealed chimeras composed of residues exchanged within or flanking the HP were all inactive for syncytium formation ( Figure 3B ) . A cell-surface biotinylation assay , previously employed to identify a cysteine loop in ARV p10 [43] , determined NBV p10 also forms a cystine loop; cells had to be treated with DTT before plasma-membrane localized NBV p10 could react with the thiol-specific biotinylation reagent ( Figure 3C ) . More notably , fusion-dead ARV and NBV p10 constructs ( A1–A3 and N1–N3 ) could all be biotinylated without DTT treatment , indicating the presence of free thiols and inability to form an intramolecular disulfide bond . Conversely , constructs with substitutions flanking the CM ( A4 , A5 , N4 and N5 ) remained fusion-active ( Figure 3B ) and retained the cystine loop ( Figure 3C ) . Thus , the N-terminal ∼25 ectodomain residues of ARV and NBV p10 define a single motif required for species-specific formation and fusion activity of an essential cystine loop-based FP . QM5 cells were co-transfected with N-terminally FLAG-tagged ARV p10 and N-terminally myc-tagged NBV p10 to allow simultaneous visualization of both proteins on the surface of non-permeabilized cells . Immunofluorescence microscopy revealed both p10 proteins displayed a punctate staining pattern randomly distributed over the cell surface , with little to no colocalization of the ARV and NBV puncta ( Figure 4A ) . In contrast , FLAG- and myc-tagged versions of the same parental protein perfectly colocalized in puncta ( Figure S2A ) , indicating each punctum comprises a homotypic clustering of more than one p10 protein . The previously created chimeric p10 constructs were N-terminally myc-tagged and co-expressed with N-terminally FLAG-tagged parental constructs to define which domain ( s ) govern segregated surface localization . ARV p10 constructs containing an NBV TMD or endodomain perfectly colocalized with parental ARV p10 , while the ARV construct containing the NBV ectodomain displayed almost no colocalization with ARV p10 ( Figure 4B , top row ) . The same situation occurred with the NBV constructs; an ARV ectodomain prevented colocalization with parental NBV p10 while constructs with an ARV TMD or endodomain colocalized with NBV p10 ( Figure 4B , bottom row ) . Colocalization was also observed in cells co-expressing FLAG-tagged ARV p10 and myc-tagged NectoA , and in cells co-expressing FLAG-tagged NBV p10 and myc-tagged AectoN ( Figure 4C ) , indicating p10 ectodomains are both necessary and sufficient to determine homotypic p10 clustering in the plasma membrane . A similar punctate plasma membrane staining pattern was previously noted for RRV p14 FAST protein , which localizes to cholesterol-rich microdomains that share features with lipid rafts [48] . Acute extraction of cholesterol from QM5 cells expressing N-terminally FLAG-tagged ARV p10 using methyl-β-cyclodextrin ( MβCD ) resulted in a dose-dependent alteration in p10 plasma membrane staining from punctate to diffuse ( Figure S2B ) . Live-cell fluorescence imaging with C-terminally GFP-tagged p10 indicated the punctate plasma membrane localization of p10 was rapidly converted to a diffuse staining pattern following addition of MβCD to cells ( Video S1 ) . Most interestingly , the diffuse , overlapping staining pattern of ARV and NBV p10 in the absence of cholesterol rapidly reverted to segregated , homotypic , punctate staining when membrane cholesterol was restored using cholesterol-loaded MβCD ( Figure 4D ) . Thus , the p10 ectodomain directs the reversible and preferential homotypic clustering of plasma membrane-localized ARV and NBV p10 into species-specific , cholesterol-dependent microdomains . A series of ARV constructs containing point substitutions in the ectodomain were used for fine structure mapping of sequences governing p10 clustering in membrane microdomains . A C9S substitution in ARV p10 , which abrogates cystine loop formation and syncytiogenesis [43] , had no effect on p10 clustering ( Figure S2C ) , indicating the cystine loop FP motif does not influence p10 cell surface distribution . To determine whether the CM plays any role in p10 microdomain association , each non-Ala/Gly residue of the ARV p10 CM was individually point-substituted with either a conserved residue or with Ala , while each Ala/Gly residue was individually substituted to the obverse ( Figure 5A ) . Surface immunofluorescence and syncytiogenesis assays with these 14 CM point substitutions revealed changes to the CM eliminated p10-induced syncytium formation ( Figure 5A ) and generated a diffuse surface-staining pattern ( Figure 5B , top and middle rows ) . The four diverged , juxtamembrane residues of the MPER , hereinafter referred to as the neck region , connect the TMD to the CM and can be exchanged in the ARV and NBV p10 proteins with no effects on syncytiogenesis ( Figure 3B ) . However , substitution of these neck residues with Ala had the same effect as CM substitutions , abrogating both p10-induced syncytium formation ( Figure 5A ) and clustering in membrane microdomains ( Figure 5B , bottom row ) . The p10 MPER , encompassing CM and neck residues , therefore governs the essential clustering of p10 in cholesterol-rich membrane microdomains . To identify regions responsible for homotypic clustering of ARV and NBV p10 , the previously constructed chimeric ectodomain constructs ( i . e . , those constructs where diverged ectodomain sequences were exchanged ) were co-expressed with each of the parental p10 constructs . Surface immunofluorescence microscopy revealed ARV and NBV p10 each colocalized homotypically with chimeric constructs containing heterotypic sequences in or flanking the HP , or on the N-proximal side of the CM ( the A/N 1 , 2 , 3 and 4 constructs ) ( Figure 5D ) . In contrast , exchange of the four-residue neck region converted homotypic clustering to heterotypic clustering; NBV p10 perfectly colocalized with ARV p10 containing the tetra-peptide neck residues from NBV , and visa versa ( Figure 5D and E ) . Thus , the juxtamembrane tetra-peptide is solely responsible for species-specific homotypic clustering of p10 in cholesterol-dependent plasma membrane microdomains . Neck region exchanges redirected p10 to a heterotypic clustering pattern ( Figure 5D and E ) but had no effect on disulfide loop formation ( Figure 3C ) or syncytiogenic activity ( Figure 3B ) , indicating localization of ARV and NBV p10 to discrete membrane microdomains does not contribute to differing fusion efficiencies of these FAST proteins . To further explore the relationship between microdomain association and fusion activity , syncytium formation was assessed in QM5 cells co-transfected with a parental p10 protein and equivalent amounts of plasmid DNA expressing one of several chimeric or substituted constructs that had various effects on p10 clustering and syncytiogenesis . Co-transfecting cells with the same parental p10 ( i . e . twice the dose of p10-expressing plasmid DNA ) increased syncytiogenesis ∼50–100% relative to cells co-transfected with empty vector ( e . g . , ARV+ARV relative to ARV+vector , and NBV+NBV relative to NBV+vector; Figure 6A ) . Co-expressing parental p10 with constructs defective in both syncytium formation and clustering ( e . g . CM mutants or Ala substitutions of the neck regions ) had the same effect as co-transfecting empty vector; these constructs had no effect on parental p10-induced syncytiogenesis ( Figure 6A ) . Co-expressing the ARV p10 C9S mutant , which is fusion-incompetent but retains the MPER homotypic microdomain association motif , eliminated ARV p10-induced syncytium formation but had no effect on NBV p10 syncytiogenesis ( Figure 6B ) . Conversely , a C9S substitution in an ARV p10 A5 background ( i . e . , ARV p10 containing the NBV p10 neck region ) suppressed NBV p10-induced syncytium formation by ∼99% ( Figure 6B ) . Varying the proportion of functional NBV p10 and non-functional ARV A5 C9S yielded a dose-response curve , with 50% maximal syncytiogenesis occurring at a functional∶non-functional p10 ratio of ∼4∶1 ( Figure 6C ) . Fusion-deficient constructs therefore exert a potent dominant-negative effect , but only when co-localized to the same membrane microdomain as a parental p10 protein . Lastly , co-expressing ARV or NBV p10 with heterotypic neck region substitution constructs that co-localize with each other ( i . e . , ARV p10 with NBV N5 , or NBV p10 with ARV A5 in Figure 5D ) resulted in approximately equivalent , intermediate fusion phenotypes; syncytium formation was more extensive than that induced by ARV p10 alone but reduced relative to NBV p10 alone ( Figure 6A ) . Taken together , these results indicate each punctum represents a “fusion unit” , with overall fusion efficiency being dictated by the population of p10 proteins present in individual puncta . To determine if puncta contain p10 multimers , we employed fluorescence resonance energy transfer ( FRET ) in cellulo . Since the efficiency of energy transfer between fluorophores varies inversely to the 6th power of the distance between donor and acceptor fluorophores , FRET only occurs if the average spatial separation of the fluorophores is <5–10 nm , a distance only stably observed during direct protein-protein interactions [49] . ARV and NBV p10 proteins were C-terminally tagged with either enhanced green fluorescent protein ( EGFP ) or a monomeric derivative of red fluorescent protein ( mCherry ) , a FRET pair with good spectral overlap but low donor-acceptor cross-talk levels [50] . The PixFRET Image-J plug-in [51] was used to calculate donor and acceptor spectral bleed-through ( SBT ) values and normalized FRET ( NFRET ) intensities using pixel-by-pixel analysis of sensitized emission FRET , and mean NFRET ( mNFRET ) values were determined for 10 cells in each of two separate experiments . Positive controls included a uni-molecular FRET pair ( i . e . , EGFP directly attached to mCherry via a flexible linker ) and a bi-molecular FRET pair , the multimeric p14 FAST protein , both of which gave positive FRET signals ( Figure 7A ) . The negative controls , ARV-GFP co-expressed with free mCherry ( Figure 7B ) and ARV-mCherry co-expressed with free GFP ( data not shown ) , yielded no detectible FRET signals . As shown in the fluorescence images ( Figure 7A ) and NFRET quantification ( Figure 7B ) , ARV and NBV p10 both formed homomultimers in cells but failed to heteromultimerize . Two representative CM substitutions that abrogate ARV p10 plasma clustering ( D31E and T35A ) also eliminated multimerization as shown by loss of FRET signal ( Figure 8A ) . Similarly , co-expressing ARV p10-GFP with mCherry-tagged ARV p10 containing either Ala substitutions of the four neck residues ( A-neck construct ) or the heterologous neck residues from NBV p10 ( ARV A5 construct ) eliminated the FRET signal , as did similar co-transfections with NBV p10 constructs ( Figure 8A ) ; all of these neck constructs also abrogated fusion activity and p10 clustering in the plasma membrane ( Figure 5 ) . Conversely , co-expression of ARV p10 with the NBV N5 construct ( NBV p10 with the four neck residues of ARV p10 ) , or co-expression of NBV p10 with the corresponding ARV A5 construct , both generated positive FRET signals ( Figure 8A ) and co-clustered in the plasma membrane ( Figure 5 ) . Additionally , cholesterol extraction with MβCD eliminated the detectible FRET signal ( Figure 8B ) , indicating p10 multimerization and p10 clustering in the plasma membrane are both cholesterol-dependent . The MPER therefore controls cholesterol-dependent , reversible homomultimerization and clustering of p10 in plasma membrane fusion platforms , while the tetra-peptide neck residues are solely responsible for determining multimer specificity .
Reovirus-induced syncytium formation is a correlate of virulence [22] , [23] , the fusogenic reoviruses evolved from at least two ( and possibly three or more ) gain-of-function events suggesting additional fusogenic reoviruses might emerge [20] , and the fusogenic pteropine orthoreoviruses represent a potential threat as emerging human pathogens [13] . Understanding what dictates the fusogenic capacity of the FAST proteins is therefore of some interest . Our comparative analysis of the homologous avian and pteropine orthoreovirus p10 FAST proteins revealed the small ectodomains of these rudimentary fusogens operate as remarkably compact , multifunctional fusion modules that function as the predominant determinant of species-specific p10 fusion efficiency . Mechanistically , the ectodomain comprises two separate components , one of which directs formation of a cystine loop FP and the other co-dependent clustering and multimerization of p10 in plasma membrane microdomains . Plasma membrane clustering and multimerization are cholesterol-dependent , required for syncytiogenesis , reversible , and governed by the MPER , which includes a juxtamembrane tetra-peptide solely responsible for species-specific p10 clustering in microdomains and homomultimerization . Most notably , co-transfections revealed that overall syncytiogenic efficiency is dictated by the p10 composition of individual microdomains , identifying these p10 plasma membrane clusters as multimeric FAST protein fusion platforms . Differences in p10 fusion efficiency map primarily to the more conserved ectodomain rather than to the highly diverged endodomains ( 44% versus 19% sequence identity , respectively ) . While fusion efficiency correlated with the parental source of the ectodomain , chimeras required a homotypic TMD or endodomain to display the full fusion capacity of the parental p10 protein contributing the ectodomain . A similar situation occurs with influenza virus hemagglutinin ( HA ) , although in this instance it reflects the preference for a homotypic combination of TMD and endodomain [52] . How these individual domains function in concert during the fusion process has not been determined , either for the FAST proteins or for enveloped virus fusogens . However , prior studies determined FAST protein TMDs and endodomains are required for fusion pore formation and expansion [32] , [53] , [54] , while their ectodomains mediate the early lipid mixing stage of membrane fusion [43] , [44] , [47] , [55] . Ectodomain-mediated lipid mixing leading to pore formation may therefore be more efficient when pore stabilization and expansion are mediated by a homotypic TMD and/or endodomain . The present results also implicate relative lipid mixing activities of the ARV and NBV p10 ectodomain cystine loop FPs in the differing syncytiogenic efficiencies of these FAST proteins . Changing the clustering pattern of p10 by exchanging MPER neck motifs did not alter syncytiogenic activity of the parental p10 constructs . For example , the A5 construct ( ARV p10 with an NBV neck ) had an NBV plasma membrane segregation phenotype ( Figure 5 ) but an ARV p10 fusion phenotype ( Figure 3 ) , indicating fusion efficiency is not affected by the species-specific microdomain with which p10 associates . Lipid mixing activities of the p10 ectodomain peptides did , however , correlate with p10 fusion efficiency ( Figure 2D ) . The ARV p10 ectodomain was recently shown to form an essential cystine loop FP [43] , results we have now extended to NBV p10 ( Figure 3 ) . Furthermore , the ARV and NBV cystine loops are not interchangeable , a reflection of the remarkable sensitivity of disulfide bond formation to species-specific sequences within , and flanking , the p10 loop ( Figure 3 ) . Remarkably , despite the extreme sensitivity of disulfide bond formation to minor sequence changes , substitutions in the CM ( D31E and T35A ) that disrupt clustering and multimerization did not affect disulfide bond formation ( Figures 5 and 8 ) , indicating monomeric p10 contains the essential cystine loop FP and is theoretically functional ( see below ) . The N-terminal 21–25 residues of p10 have therefore evolved as an independent structural and functional motif whose sequence-specific geometry is required for formation or stability of an essential cystine loop FP . We also identified a second independent ectodomain motif that defines a new role for a viral fusogen MPER , namely cholesterol-dependent homomultimerization . MPERs in viral fusogens tend to be short ( ∼10–20 residues ) , hydrophobic or amphiphilic sequences frequently enriched in aromatic residues [42] . Their ability to adopt amphipathic helical structures , self-aggregate , and partition into membrane interfacial regions suggests the role of MPERs in membrane fusion reflects their ability to perturb bilayer structure [38] , [39] , [41] , [56] , [57] . In contrast , the 13-residue MPERs of ARV and NBV p10 are mostly polar and almost devoid of aromatic residues , and their essential role in membrane fusion is to direct reversible , cholesterol-dependent p10 multimerization . Although noted as an invariant sequence in all p10 isolates from avian and pteropine hosts , no functions have previously been ascribed to the CM . FRET analysis and a panel of CM point substitutions suggested a striking requirement for sequence-specific , spatial self-complementarity in the CM to generate a binding interface suitable for p10 multimerization . While homotypic CM interactions are necessary for multimerization , they are not sufficient; stable multimerization only occurred in conjunction with cholesterol-dependent clustering of p10 in plasma membrane microdomains . Cholesterol-dependent lateral segregation of proteins and lipids creates submicroscopic membrane domains with unique biophysical properties , such as increased membrane thickness and altered lipid order [58]; these or other features of lipid microdomains appear to be needed to stabilize low-affinity p10 multimer interactions . Similar co-dependent raft assembly and multimerization have been reported in other systems [59]–[61] . While cholesterol microdomains have been implicated in the function of numerous enveloped virus fusogens [62] , we are unaware of any instance where they are needed for enveloped virus fusogen multimerization . The one exception is Semliki Forest virus E1 , where FP binding to sterol- and sphingolipid-rich microdomains in the target membrane is required for low-pH-dependent conversion of E1 pre-fusion heterodimers to fusion active homotrimers [63] . In contrast , the cholesterol-dependence of p10 multimerization reflects an undefined functional interaction between the MPER and donor membrane cholesterol needed to convert p10 monomers to stable multimers . Indirect evidence also suggests p10 microdomains contain higher-order p10 complexes and function as fusion platforms . The ARV p10 construct containing the neck region of NBV ( A5 construct ) co-clusters with NBV p10 ( Figure 5 ) , while an ARV p10 C9S substitution prevents formation of the cystine loop and eliminates cell-cell fusion ( Figure 5 ) but has no effect on p10 homotypic clustering ( Figure S2C ) . Co-expression of NBV p10 with an equimolar amount of an A5 construct containing the C9S substitution resulted in a 99 . 15% decrease in NBV p10 syncytiogenesis ( Figure 6B ) . Assuming incorporation of one non-functional p10 polypeptide into a punctum is sufficient to eliminate fusion activity of the complex , then the 0 . 85±0 . 53% ( SEM , n = 3 ) of retained syncytiogenic activity would be attributable to clusters containing only parental NBV p10 . Based on a binomial distribution , the probability of this percent of clusters containing just parental NBV p10 polypeptides is best approximated by heptameric or octameric clustering ( probability of 0 . 78% and 0 . 39% , respectively ) . This interpretation assumes there is a single fusion site formed per cell . It is likely that several functional “fusion platforms” are formed on a single cell . By varying the proportion of functional NBV p10 and non-functional ARV A5 C9S ( Figure 6C ) , we were able to consider the influence of multiple fusion sites on our previous analysis , as previously reported for rabies virus G protein [64] . Residual fusion activity should equal ( equation 1 ) , where f is the proportion of functional p10 proteins , n is the number of p10 proteins in a cluster , and s is the number of fusion sites . Fitting our titration data ( Figure 6C ) to this equation returned best-fit values of n = 8 . 74±0 . 70 and s = 5 . 17±0 . 94 . We interpret these estimates as an indicator that clusters comprise higher-order assemblies of p10 monomers in the range of octamers . Cholesterol-dependent , higher-order multimerization is evident with other proteins , such as HIV-1 Gag , where initial membrane binding by lower-order ( dimers ) Gag multimers progresses to higher-order Gag-Gag interactions in lipid rafts [65]–[67] . The clear correlation between p10 clustering and multimerization in membrane microdomains with p10 fusion competency also suggests microdomains increase local plasma membrane concentrations of p10 to generate fusion platforms , similar to cholesterol-dependent assembly of exocytic SNARE complexes into fusion sites [68] . This conclusion is strengthened by results indicating only non-functional constructs that co-clustered and co-multimerized with parental p10 proteins exerted dominant-negative effects . For example , CM substitutions ( D31E and T35A ) display a diffuse plasma membrane staining pattern ( Figure 5B ) , do not multimerize with ARV p10 ( Figure 8 ) and exert no dominant-negative effects ( Figure 6A ) . Conversely , fusion dead constructs that colocalized with parental p10 ( e . g . ARV p10 C9S ) inhibited parental fusion activity ( Figure 6B ) , while functional neck region chimeras generated heterotypic puncta and a syncytiogenic phenotype intermediate between ARV and NBV p10 proteins ( Figure 6A ) . Overall syncytiogenic efficiency is therefore determined not by the number of p10 microdomains but by their compositional p10 bias , indicating individual puncta define a fusion unit . While the CM and neck region of the MPER both control microdomain association and multimerization , the tetra-peptide neck is solely responsible for segregation of p10 into species-specific microdomains . How proteins partition into lipid rafts remains an unresolved issue in membrane biology [58] . Recent developments in the field suggest the presence of compositionally distinct membrane microdomains differing in their lipid composition , length and asymmetry of N-acyl fatty acids , and hydration status of polar headgroups . These biophysical properties provide lipid environments capable of discriminating between different membrane-associated proteins [69] . For example , HIV env and Ebola GP segregate to different lipid microdomains , both of which can be recruited by Gag to generate HIV particles with a single type of glycoprotein [70] . Proteins may also directly contribute to raft heterogeneity by specifically recruiting and stabilizing a localized lipid environment . Palmitoylation of membrane-proximal cysteines and hydrophobic matching between the lengths of acyl chains and TMDs are the two most common protein features associated with raft recruitment and protein localization to membrane microdomains [71] , [72] . As we now show , MPERs can also function as microdomain sorting signals . The tetra-peptide neck of p10 provides a remarkably specific sorting signal capable of segregating two homologous membrane proteins into distinct cholesterol-dependent microdomains . This sorting event occurs in the plasma membrane , and is fully reversible following cholesterol depletion and repletion . There are several feasible explanations for how the p10 neck might provide both microdomain and multimerization specificity . In a protein-centric-model , the neck would function as the specificity determinant of the MPER multimerization motif , directing homotypic lower-order multimerization . Lipid recruitment by these p10 nanoclusters to generate a sterically favored lipid environment would then recruit additional monomers or lower-order multimers to create a fusion platform . In a lipid-centric model , the tetra-peptide p10 sorting signal functions primarily as a lipid recognition or adaptation motif . Neck residues could interact with specific lipid headgroups in the exoplasmic leaflet , similar to the reported cholesterol-binding activity of the conserved LWYIK juxtamembrane peptide of HIV gp41 [73] . Alternately , the neck may affect hydrophobic matching of the p10 TMD to acyl chain lengths of adjacent lipids . Recent molecular dynamic simulations and studies in artificial bilayers indicate cholesterol decreases the effects of hydrophobic mismatch by altering bilayer thickness and lipid packing , and by inducing changes in TMD helix length [71] . We note that the ARV and NBV p10 necks display very different arrangements of polar and non-polar residues ( Figure 1 ) ; ARV p10 has a polar-apolar-TMD arrangement while NBV p10 is apolar-polar-TMD . These different arrangements may influence cholesterol-mediated hydrophobic matching and lateral segregation of ARV and NBV p10 into distinct lipid microenvironments . The co-dependent and inseparable relationship between p10 multimerization and microdomain clustering makes it unclear , in either protein- or lipid-centric models , whether p10 associates with , or assembles , lipid rafts to promote multimerization . Lastly , the readily reversible nature of p10 multimerization in the plasma membrane was an intriguing , and potentially functionally significant , observation . In contrast to the functionally reversible nature of p10 multimerization , enveloped fusogens are locked in metastable dimeric or trimeric pre-fusion conformations that generally undergo an irreversible conformational conversion to a post-fusion , stable trimeric conformation , releasing energy to drive the fusion reaction [34] , [36] . Thermodynamically favorable , reversible p10 multimerization implies fusion is not dependent on such energy releasing conformational changes . Furthermore , the co-dependence of multimerization and microdomain association suggests p10 may be capable of shuttling in and out of lipid rafts during the fusion reaction , similar to what occurs with the fusion machinery during peroxisome fusion [74] . Taken together with results indicating monomeric p10 still contains the essential cystine noose FP , this suggests there may be a role for reversible p10 multimerization during the fusion process . Interestingly , conformational changes of enveloped virus fusogens during transition from the pre-fusion structure to the post-fusion , trimeric hairpin conformation are impossible unless the ternary structure is dissociated [75] . Recent studies suggest that monomeric intermediates exist for both tick-borne encephalitis ( TBE ) virus E [76] and VSV G fusion proteins [77] . Monomeric intermediates and dynamic clustering and dispersion of protein fusogens at sites of fusion , either raft-mediated or otherwise , may therefore be a functionally relevant feature of viral fusogens .
QM5 and Vero cells were grown and maintained as previously described ( Corcoran and Duncan , 2004 ) . Rabbit antisera generated against full-length ARV p10 or ARV p10 endodomain were previously described [43] , [78] . Monoclonal mouse anti-FLAG ( Sigma-Aldrich ) and rabbit anti-c-myc ( Sigma-Aldrich ) antibodies , horseradish peroxidase ( HRP ) -conjugated goat anti-rabbit ( Santa Cruz ) and goat-anti mouse ( Santa Cruz ) antibodies , Alexa Fluor 488-conjugated goat anti-mouse ( Invitrogen ) and goat anti-rabbit ( Invitrogen ) , antibodies , Alexa Fluor 647-conjugated goat anti-rabbit IgG ( Invitrogen ) , maleimide-PEG2-biotin ( Thermo Scientific ) , neutravidin agarose resin ( Thermo Scientific ) , and methyl-β-cyclodextrin ( Sigma-Aldrich ) were purchased from the indicated commercial sources . ARV p10 and NBV p10 subcloned into pcDNA3 mammalian expression vectors were previously described [79] . Full domain exchanges and ectodomain segment exchanges between ARV and NBV p10 were created using sequential PCR reactions with custom oligonucleotide primers and clones into pcDNA3 . The QuickChange site-directed mutagenesis kit ( Agilent Technologies ) was used according to the manufacturer's instructions to generate point substitutions for all CM and neck region constructs . A triple FLAG tag was added to the N-terminus of indicated p10 constructs by PCR amplification and cloning . Custom oligonucleotide primers were purchased from IDT , and all constructs were confirmed by sequencing . QM5 or Vero cells were transfected using Polyethylimine ( PEI , Polysciences Inc . ) as per manufacturer's instructions . Syncytial indexing was performed on 50% confluent monolayers in 12-well plates transfected with 0 . 5 µg of plasmid DNA ( unless otherwise stated ) . At 4 h post- transfection , the transfection mix was replaced with Earle's 199 growth media ( Gibco ) supplemented with 10% fetal bovine serum ( Sigma-Aldrich ) . At indicated times post-transfection , cells were fixed with methanol and Wright-Giemsa stained ( Siemens Healthcare Diagnostics ) . Stained monolayers were imaged using a Nikon DIAPHOT-TMD under 200× magnification . The numbers of syncytial nuclei were counted from five random fields per well , in three separate experiments , and the syncytial index reported as the mean ± SEM propagating errors within and across experiments . Data in Figure 6C was fit to the following equation , described in Roche and Godin [64] , using the curve fitting toolbox 3 . 3 application in MATLAB R2012b: ( 1 ) where f is the proportion of functional p10 proteins , n is the number of p10 proteins in a cluster , and s is the number of fusion sites . SDS-PAGE and Western blotting were carried out as previously described [43] , [53] . A 1∶10 , 000 dilution of anti-FLAG mouse antiserum followed by a 1∶10 , 000 dilution of HRP-conjugated goat anti-mouse secondary antibody was used to measure overall protein expression levels . Western blots were developed using ECLplus western blotting reagent ( GE Healthcare ) and imaged on a Kodak 4000 mm Pro CCD imager . Transfected QM5 monolayers were incubated for 24 h in Earle's 199 growth media supplemented with 10% fetal bovine serum . Live cells were immunolabeled at 4°C with 1∶200 dilution of mouse anti-FLAG primary antibody , followed by 1∶2000 dilution of Alexa 647-conjugated goat anti-mouse antibody , as previously described [43] , [53] . Cells were resuspended with 50 mM EDTA in PBS , fixed in 3 . 7% formaldehyde , and surface expression was quantified using a FACSCalibur flow cytometer ( Becton Dickinson ) by counting 20 , 000 cells . Cell surface fluorescence was analyzed using FCS Express 2 . 0 ( De Novo Software ) . The presence on an intramolecular disulfide bond in the ectodomain of FLAG-tagged p10 constructs was detected as previously described [43] . QM5 cells seeded in 10 cm dishes ( Corning ) at ∼50% confluency were transfected with the indicated constructs . At 24 h post-transfection cells were washed twice with Hanks buffered saline solution ( HBSS ) before being incubated in HBSS with or without 0 . 1 mM dithiothreitol ( DTT ) for five min . Cells were washed three times with HBSS , then incubated with shaking in 1 µg/ml maliemide-PEG2-biotin ( a membrane impermeable biotinylation reagent ) for 25 min at 4°C to biotinylate free thiol groups on the cell surface . Cells were washed four times with HBSS to remove excess biotin reagent , then once with HBSS containing 1% BSA to quench residual biotinylation reagent . Cells were washed twice with PBS , then resuspended with 50 mM EDTA in PBS , lysed in RIPA buffer ( Tris , pH 8 . 0 , 150 mM NaCl , 1 mM EDTA , 1% NP40 , 0 . 5% NaDOC ) with protease inhibitors ( Pierce ) , the lysate was incubated overnight with neutravidin agarose resin to pull down biotinylated proteins , and pellets were boiled in protein sample buffer with 100 mM DTT to release biotinylated proteins . Samples were then analyzed via SDS-PAGE and Western blotting . QM5 cells grown on coverslips at approximately 50% confluence were transfected with FLAG- or c-myc-tagged versions of ARV or NBV p10 constructs as described above . At 24 h post-transfection , cells were fixed with paraformaldehyde , blocked with 1% BSA in PBS for 30 min , and then incubated with 1∶1000 dilutions of mouse anti-FLAG and rabbit anti-c-myc monoclonal antibodies for 1 h at room temperature . After thorough washing with PBS , the cells were incubated with 1∶1000 dilutions of Alexa 488-conjugated goat anti-mouse antibody and Alexa 647-conjugated goat anti-rabbit antibody for 1 h at room temperature . Coverslips were then mounted and sealed on slides using prolong gold anti-fade reagent ( Invitrogen ) , then imaged using an Axiovert 200M inverted microscope ( Zeiss ) . A fluorescence resonance energy transfer ( FRET ) -based mixing assay was employed to monitor the lipid-mixing potential of synthetic p10 ectodomain peptides , as previously described [47] . Large unilamellar vesicles composed of a 40∶20∶20∶20 ratio of 1 , 2-dioleoyl-sn-glycero-3- phosphocholine ( DOPC ) , 1 , 2-dioleoyl-sn-glycero- 3-phosphoethanolamine ( DOPE ) , cholesterol and sphingomylein ( Avanti Polar Lipids ) in 10 mM phosphate buffer with 100 mM NaF , pH 7 . 4 , were prepared by extrusion through 100 nm polycarbonate filter . A fluorescently-labeled liposome population was similarly prepared to contain 2 mol% each of 1 , 2-dioleoyl-sn-glycero-3-phosphoethanolamine-N- ( 7-nitro-2-1 , 3- benzoxadiazol-4-yl ) ( NBD-DOPE ) and 1 , 2- dioleoyl-sn-glycero-3-phosphoethanolamine-N-4 ( lissamine rhodamine B sulfonyl ) ( Rho-DOPE ) . Non-labelled and labeled liposomes were incubated at 37°C in a 9∶1 ratio at a concentration of 100 mM prior to addition of ARV or NBV p10 synthetic peptides dissolved in the same buffer as liposomes and added to the indicated final concentrations , or a buffer-only control . Fluorescence was recorded for up to ten minutes . The percent lipid mixing was calculated using equation ( 2 ) : ( 2 ) A third liposome population of 0 . 2 mol% each of NBD-DOPE and Rho-DOPE represented the theoretical maximum ( FMAX ) level of lipid mixing . All experiments were performed in triplicate . QM5 cell monolayers grown on glass coverslips were transfected with EGFP-tagged NBV p10 . At 24 h post-transfection , live cells were imaged using a spinning-disc confocal microscope ( 3i Intelligent Imaging Innovations , Denver , CO ) consisting of a Cell Observer Z1 microscope ( Zeiss ) , an Evolve EMCCD camera ( Photometric ) , and a CSU-X1 spinning disk head ( Yokagawa ) . Images were captured at 1 min intervals to monitor the movement of fluorescently-tagged proteins . At t = 100 sec , 20 mM MβCD in HBSS was circulated into the cell chamber to chelate cholesterol . The diffusion of p10 proteins was followed for an additional 800 sec . Images were acquired using the Slidebook imaging software ( Version 5 . 0 ) . Additionally , QM5 cell monolayers transfected with FLAG-tagged ARV p10 were treated with 2 , 5 , 10 or 20 mM MβCD for 20 min , then fixed with 3 . 7% paraformaldehyde . Immunofluorescence staining for surface-localized proteins was performed as described above . Cholesterol repletion experiments were performed as previously described [48] on QM5 cell monolayers transfected with FLAG-tagged ARV p10 and c-myc-tagged NBV p10 . Cholesterol-loaded MβCD was made by dissolving cholesterol ( 6 mg/ml ) in serum-free medium containing 20 mM MβCD by vigorous vortexing and heating at 37°C for 30 min , then filtering to remove insoluble cholesterol . Immunofluorescence staining for surface-localized proteins was performed as described above . QM5 cells grown on glass coverslips at approximately 50% confluence were transfected with EGFP- and/or mCherry-tagged versions of indicated ARV and NBV p10 mutant constructs . A Zeiss Laser Scanning Microscope ( LSM ) 510 equipped with a META detector was used in wide field mode to detect sensitized emission FRET . Cells were imaged under a 100× oil immersion , 1 . 4 NA Plan Apochromat objective . EGFP was excited using a 40 mW Argon laser at 488 nm , and mCherry was excited using a HeNe 548 nm laser . For microscope set-up and control experiments , additional transfections of free EGFP alone , free mCherry alone , free EGFP and free mCherry together , and EGFP directly linked to mCherry ( EGFP-mCherry ) were done concurrently . The donor and acceptor spectral bleed-through ( SBT ) values were visually minimized using the free EGFP and free mCherry samples , respectively , prior to detecting FRET in sample conditions . Donor and acceptor SBT ratios were modeled using exponential relationships with fluorophore intensity after exclusion of aberrant background values at low intensities and the application of a Gaussian blur . With the SBT values determined , the normalized FRET ( NFRET ) for the p10-p10 interaction was calculated . A series of three images was acquired for each cell imaged: ( 1 ) sensitized emission FRET image ( donor excitation , acceptor emission; ( 2 ) donor image ( donor excitation , donor emission ) ; and ( 3 ) acceptor image ( acceptor excitation , acceptor emission ) . Background subtraction and Gaussian blur of the donor , acceptor and FRET channels were performed on each image stack prior to analysis . Ten images were acquired for each sample condition in duplicate ( total of twenty images ) . Using the PixFRET Image J plugin [51] , the FRET intensity of each pixel was normalized to the donor and acceptor expression levels by dividing the FRET-channel pixel intensity by the square-root of the product of the corresponding donor- and acceptor-channel pixels using equation ( 3 ) : ( 3 ) This normalization provided a measure of sensitized emission FRET that is comparable between different samples [80] . Pixel amplitude distributions of the 8-bit NFRET images generated by the PixFRET software were summarized as histograms with a bin width of 0 . 03906 NFRET units . Each histogram was fit with four Gaussian distributions , and that with the highest calculated R2 value was used for further analysis . The mean NFRET ( mNFRET ) was determined for each image from the best-fit Gaussian distribution . The Gaussian-fitted NFRET histograms were also used to calculate the average pixel amplitude from each condition . GenBank accession numbers for avian and pteropine FAST proteins analyzed in this study . ARV isolates: 176 , AAF45151; S1133 , AAK18186; Muscovy duck , ABA33820; 138 , AAF45154; RAM-1 , AAA57266; NC/SEP-R108/03 , ABN46970; NC/98 , ABL96273; NC/SEP-R61/03 , ABN46972; TX99 , ABN46974; NC/PEMS/85 , ABN46971; Psittacine , ABY78878 . Pteropine isolates: Melaka , YP_007507326; Sikamat , AES12474; Pulau , AAR13231; Kampar , ACC77635; NBV , AAF45157 .
|
Natural infections by fusogenic orthoreoviruses can result in severe afflictions ranging from neuropathogenicity to pneumonia and death . The fusogenic capacity of these viruses , attributable to a unique family of fusion-associated small transmembrane ( FAST ) proteins , is a correlate of virulence . The FAST proteins are the only known examples of nonenveloped virus membrane fusion proteins , and they are the smallest known viral fusogens whose structural and functional attributes are incompatible with current models of protein-mediated membrane fusion . Exploiting the sequence divergence and distinct syncytiogenic rates of representative p10 FAST proteins from avian and bat reovirus isolates , we determined the p10 ectodomain is a compact , complex fusion module comprising two independent functional motifs . One motif determines species-specific p10 fusion efficiency by governing formation of a cystine loop fusion peptide , while the other directs reversible clustering and multimerization of p10 in cholesterol-dependent membrane microdomains . Remarkably , a juxtamembrane tetra-peptide is solely responsible for co-dependent clustering and multimerization of p10 in distinct , species-specific fusion platforms . This is the first example of a viral fusogen utilizing a membrane-proximal ectodomain region ( MPER ) to direct cholesterol-dependent multimerization and assembly into fusion platforms .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"lipids",
"transmembrane",
"proteins",
"protein",
"interactions",
"host",
"cells",
"proteins",
"viral",
"transmission",
"and",
"infection",
"virology",
"emerging",
"viral",
"diseases",
"protein",
"structure",
"membranes",
"and",
"sorting",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"molecular",
"cell",
"biology",
"sterols"
] |
2014
|
A Compact, Multifunctional Fusion Module Directs Cholesterol-Dependent Homomultimerization and Syncytiogenic Efficiency of Reovirus p10 FAST Proteins
|
Israeli acute paralysis virus ( IAPV ) is a widespread RNA virus of honey bees that has been linked with colony losses . Here we describe the transmission , prevalence , and genetic traits of this virus , along with host transcriptional responses to infections . Further , we present RNAi-based strategies for limiting an important mechanism used by IAPV to subvert host defenses . Our study shows that IAPV is established as a persistent infection in honey bee populations , likely enabled by both horizontal and vertical transmission pathways . The phenotypic differences in pathology among different strains of IAPV found globally may be due to high levels of standing genetic variation . Microarray profiles of host responses to IAPV infection revealed that mitochondrial function is the most significantly affected biological process , suggesting that viral infection causes significant disturbance in energy-related host processes . The expression of genes involved in immune pathways in adult bees indicates that IAPV infection triggers active immune responses . The evidence that silencing an IAPV-encoded putative suppressor of RNAi reduces IAPV replication suggests a functional assignment for a particular genomic region of IAPV and closely related viruses from the Family Dicistroviridae , and indicates a novel therapeutic strategy for limiting multiple honey bee viruses simultaneously and reducing colony losses due to viral diseases . We believe that the knowledge and insights gained from this study will provide a new platform for continuing studies of the IAPV–host interactions and have positive implications for disease management that will lead to mitigation of escalating honey bee colony losses worldwide .
Honey bees are the most economically valuable pollinators of agricultural crops worldwide . In the U . S . alone , the value of agricultural crops pollinated by bees each year is more than $17 billion dollars [1] . In 2006 , an enigmatic phenomenon labeled Colony Collapse Disorder ( CCD ) was observed in U . S . beekeeping operations . CCD is defined as an unusually sudden decrease in the numbers of worker honey bees , without expected signs of disease , starvation , or reproductive failure [2] . Such rapid declines have been observed throughout the history of beekeeping , and their causes often remain enigmatic . Since 2006 , colony losses have been noted in beekeeping operations in much of the world [3] , posing a significant threat to the pollination of many agricultural crops [4] . There is no single agent yet identified that causes CCD . Instead , it appears that CCD results from a combination of factors that include pathogens/parasites , pesticides , malnutrition , environmental stress , low genetic diversity , and migratory beekeeping practices . It is also conceivable that synergistic effects of two or more insults are behind recent declines . To that end , there is some evidence that interactions between pathogens and neuro-active pesticides can synergistically affect honey bee mortality , contributing to colony depopulation [5] , [6] . An early survey [7] of healthy and CCD-affected colonies in the U . S . found a significant correlation between CCD-affected colonies and Israeli acute paralysis virus ( IAPV ) , an RNA virus first identified in 2004 [8] . The result drew immediately international attention to the risks of virus infection in honey bees . The role of IAPV in triggering colony declines , alone or in concert with other factors , remains a research priority . The parasitic mite Varroa destructor has long been considered the primary threat to honey bees [9] , in part because this mites serves as a vector of honey bee viruses [10] . For example , levels of Deformed wing virus ( DWV ) , a common virus that has killed billions of honey bees across the globe , are greatly increased following Varroa transmission [11] . A recent study showed that Varroa mites can also serve as vectors of IAPV; furthermore , the mite/virus association was shown to reduce host immunity and promote elevated levels of IAPV replication [12] , providing more evidence for the damaging effects of viruses associated with Varroa mite infestations . In this study , we investigated the molecular basis of pathogenesis , transmission and genetic diversity of IAPV in honey bees and evaluated the impacts of IAPV infection on colony losses . We also determined the global transcriptional profiles of honey bee responses to viral infection . Finally , we examined the inhibitory effect of small interfering RNA ( siRNA ) that targets putative virus-encoded proteins ( VSR ) on IAPV replication . The replication of single-stranded positive-sense RNA viruses results in the synthesis of complementary negative-stranded RNA , thereby producing dsRNA replicative intermediates that are attractive targets for defenses based on RNA interference . To counteract host RNAi antiviral defense , viruses have evolved strategies to suppress the antiviral effects of RNAi . A recent study with Cricket paralysis virus ( CrPV ) showed that the sequences upstream of a highly conserved sequence ( DVEXNPGP ) within the N-terminal region of CrPV ORF-1 encode a potent suppressor that mutes the RNAi antiviral defense in Drosophila [13] . As a result , we speculated that IAPV may possess a similar mechanism to counteract the antiviral response of hosts . We believe that knowledge gained from this study will lead to better understanding of the dynamics of virus disease pathogenesis in honey bees and help mitigate escalating colony losses worldwide .
Although the bee colonies in this study showed no clinical signs of infection , IAPV was found widely in surveyed honey bees colonies . IAPV-positive PCR signal was detected in eggs , larvae , pupae , adult workers , drones , and queens as well as V . destructor that fed on the bees ( Figure 1A ) . In addition , IAPV-specific PCR signal was also detected in royal jelly , honey , pollen , queen feces and drone semen collected from IAPV positive colonies ( Figure 1B ) . Strand specific RT-qPCR assays revealed that IAPV causes systemic infection in honey bees . IAPV replication was detected in hemolymph , brain , fat body , salivary gland , hypopharyngeal gland , gut , nerve , trachea , and muscle . However , the relative abundance of negative stranded RNA copies of IAPV in the different tissues varied significantly . The hemolymph ( i . e . , hemocytes ) harbored the lowest level of IAPV among the examined tissues and therefore was chosen as the calibrator . The difference in IAPV abundance in other tissues relative to hemolymph ranged from 2 . 23- to 167-fold in the following order from lowest to highest concentration: muscle<fat body<brain<trachea<salivary gland<hypopharyngeal gland<nerve<gut ( Figure 2A ) . In situ hybridization showed IAPV specific signals localized in egg , gut , ovaries , and spermatheca of infected queens . IAPV was found to be the third most common virus infection in bee colonies after DWV and Black Queen Cell Virus ( BQCV ) . Over the 4-year study period , the infection IAPV detected in the brood was significantly higher than in adult bees ( p<0 . 001 ) . When we divided our experimental bee colonies into those with more than ten frames covered with adult workers and more than six frames filled with brood and food stores ( ‘strong’ ) versus those with fewer than ten frames of adult bees , less than six combs with brood and small patches of food stores ( ‘weak’ ) , we found a measurable difference in IAPV infection levels . The average rate of IAPV infection per month was 49% for brood and 19 . 5% for adults in weak colonies and 26% for brood and 3 . 25% for adults in strong colonies . The overall rate of IAPV infection in weak colonies was significantly higher than in the strong colonies ( p<0 . 01 for brood and p<0 . 001 for adults ) . While no statistically significant seasonal variation in IAPV infection was observed in the strong colonies , the infection rate of IAPV in adult bees in weak colonies increased from spring to summer and fall and peaked in winter . While strong colonies in our survey survived through the cold winter months , almost all weak colonies collapsed before February ( Figure 3 ) . While strong colonies in our survey survived through the cold winter months , almost all weak colonies collapsed before February ( Figure 3 ) . The complete genomes of IAPV strains collected in the US states of Maryland , California , and Pennsylvania were obtained by direct sequencing of overlapping RT-PCR fragments and partial sequences from both 5′UTR and 3′UTR and deposited in GenBank with accession numbers , EU224279 , EU218534 , and EU224280 , respectively . Comparison of US , Chinese and Australian IAPV strains with the first reported Israeli IAPV strain at the genome level showed a significant genetic divergence among different strains , providing evidence of quasi-species dynamics in IAPV populations . The polymorphisms in IAPV were found more frequently in 5′ UTR and functional protein coding regions compared to the capsid protein coding region and 3′ UTR ( Figure 4A ) . Phylogenetic analysis using full-length viral genomes showed that the Australian IAPV strain constitutes the earliest lineage of the phylogenetic tree . The US strains branch to form a distinct lineage distantly related to the Israeli and Chinese strains of IAPV ( Figure 4B ) . The results of microarray analyses yielded a large group of differentially expressed genes . The principal component analysis ( PCA ) mapping showed that the total accumulative variance of the first three PCs was 78% for adult and 67 . 4% for brood , respectively , and suggested that two kinds of experimental populations ( IAPV positive vs IAPV negative ) were well separated for both adults and brood . The cluster analysis showed overall similar data patterns ( Figure S3A and B ) , indicating that inter-individual differences had a minimum effect on gene expression data . The treatment variance ( IAPV-infected versus uninfected ) was significantly higher than error variance for both adult bees and brood ( both p<0 . 01 ) ( Figure S3C ) . This confirmed that variation among samples was largely due to IAPV and suggested the good data quality for two ANOVA data analysis in both adults and brood . The distribution of differentially expressed genes in both adults and brood are presented by volcano plots ( Figure S3D ) . All microarray data were deposited in the NCBI public database with accession number GSE46278 . Overall , the transcriptional response to IAPV infection was substantially different between adults and brood . There were 2 , 522 up-regulated and 2 , 093 down-regulated genes identified in IAPV-positive adults , but only 825 up-regulated and 525 down-regulated genes identified in IAPV-positive brood with a very small fraction of overlapping genes between the two groups ( Figure 5A ) . Of the up-regulated and down-regulated genes , overlapping genes between adult and brood were 268 and 68 , respectively . A heat map illustrates the differential expression of enriched functional genes between adults and brood ( Figure 5B ) . Of the genes transcriptionally altered by IAPV infection , 2 , 150 genes identified in adults and 716 genes identified in brood could be assigned a putative function based on orthology to D . melanogaster genes . The GO-enriched analysis of the genes that displayed fold-changes of more than 1 . 5 ( False Discovery Rate adjusted ρ value≤0 . 05 ) and had putative D . melanogaster orthologs by the Database for Annotation , Visualization and Integrated Discovery ( DAVID ) revealed major functional clusters including metabolism , host cell transcription , signal transduction , cell cycle , hormone synthesis , endocytosis , phagocytosis , autophagy , and innate immune response ( Table S1 and S2 ) . The majority of genes with up-regulated expression were related to the regulation of signaling transduction and immune response , while the majority of those with down-regulated expression were involved with metabolic energy generation . Of the top functional clusters , genes that were related to immune response functions were of particular interest in this study . We examined the integrated networks and pathways of genes that were up- and down- regulated in response to IAPV infection . The global canonical pathway analysis of 2 , 150 genes identified in adults using the Ingenuity Pathway Knowledge Base led to identification of five top canonical pathways , including mitochondrial dysfunction , TCA cycle II , protein ubiquitination pathway , eIF2 signaling and γ-glutamyl cycle , with mitochondrial dysfunction ( 37 molecules , ρ-value 3 . 93E-17 ) as the most significantly affected pathway . Among five significantly disturbed canonical pathways , four showed significant up-regulation and only the γ-glutamyl cycle pathway showed significant down-regulation . The analysis of 716 IAPV regulated genes in brood identified five top canonical pathways , eIF2 signaling , mitochondrial dysfunction , mTOR signaling , TCA cycle II , regulation of eIF4 and P70S6K signaling , with eIF2 signaling ( 25 molecules , ρ-value 6 . 15E-16 ) as the most significantly affected canonical pathway . All pathways showed significant up-regulation . Among 25 networks identified , one was centered by viral infection in adults ( Figure 6 ) and contains both up- and down- regulated genes that are involved in pathways related to host defense responses such as oxidative phosphorylation , ABC transporter , endocytosis , phagocytosis , TGF-beta signaling pathway , mTOR signaling pathway , MAPK signaling pathway , JAK-STAT pathway , and lysosome . IAPV infection triggers multiple immune signaling in adult bees . qRT-PCR confirmation of immune related genes showed the components of the Janus Kinase/Signal Transducers and Activators of Transcription ( JAK-STAT ) pathway including Cbl , STAT , PIAS , and Hopscotch had ≤2 fold elevated expression in response to IAPV infection . The components of Mammalian Target of Rapamycin ( mTOR ) signaling pathway including GβL , MO25 , Dmel , and eIF4B had ≤2 fold elevated expression in response to IAPV infection . The expression of genes including Pointed , Phi , and Corkscrew that had functional association with Mitogen-activated Protein Kinases ( MAPK ) pathway was upregulated to 2 . 3- , 2 . 91- and 1 . 92-fold respectively , in response to IAPV infection . The expression of genes EGFR , PastI , Rabenosysn , and CG1115 , involved in endocytosis was also upregulated by 2 . 1- , 3 . 18- , 1 . 88- , and 3 . 1- fold , respectively . IAPV infection also caused the down-regulation of mTOR pathway gene such as Raptor , MAPK pathway genes , TII and Ras , and endocytosis gene CG6259 ranging from −2 . 14 to −3 . 9 fold . qRT-PCR analysis of immune related genes in IAPV-infected adults showed considerable concordance with the normalized microarray data ( Figure 7 ) . The sequence motif of DIEENPGP was identified in the N-terminal region of ORF-1of IAPV and other members of the Dicistroviridae family infecting honey bees such as KBV , and ABPV , where the uppercase letters of the sequence motif indicates residues with absolute sequence conservation ( Figure 8A ) . An RNAi-mediated knockdown experiment showed that silencing putative VSR in IAPV genome could effectively inhibit replication of IAPV and confer significant antiviral activity in honey bees . Quantification of the titer of negative strand RNA of IAPV showed that feeding siRNA resulted in a remarkable reduction in IAPV replication . The bees in Group I ( IAPV+siRNA ) had the lowest IAPV titer among four experimental groups at all time points ( days 1 , 3 , 5 and 7 ) and this group was therefore chosen as a calibrator . Compared to the calibrator , Group-II ( IAPV ) , Group III ( IAPV+Varroa+siRNA ) , and Group IV ( IAPV+Varroa ) averaged 4 . 78±0 . 25 , 17 . 5±0 . 56 , and 451 . 5±2 . 72 ( Mean±SE , N = 3 ) folder higher titers of negative strand RNA of IAPV , respectively . The significant reduction in virus replication observed in Groups-I and III at day 1 post treatment indicated that the impact of siRNA on the virus life cycle takes place within 24 hours . There was no significant difference in virus titer among different time points for each group ( ρ<0 . 05 , ANOVA ) . The highest titer of virus replication seen in Group IV challenged by V . destructor with no siRNA treatment provides additional evidence for the role of V . destructor in virus transmission and activation in honey bees ( Figure 8B ) . The antiviral effects of siRNA from this study ( siRNA-suppressor ) were compared to those of siRNA targeting the 5′ Internal Ribosomal Entry Site ( IRES ) of IAPV ( siRNA-5′IRES ) that was shown to confer antiviral activity in bees in our previous study [14] . The virus titer in bees fed siRNA-5′IRES was 3 . 3±0 . 54 , 4 . 5±0 . 33 , 3 . 9±0 . 21 and 5 . 2±0 . 67 ( Mean±SE , N = 3 ) fold higher than the group fed with siRNA-suppressor at Day 3 , Day 5 , and Day 7 post treatment , respectively . However , no significant difference ( p valve>0 . 05 ) was observed between groups received dsiRNAs targeting different genomic regions when Varroa mites were introduced .
The association of IAPV with honey bee declines has led to an increased awareness of the risks of viral infections on bee health . In this paper , we present a long-term study of the biological and molecular features of IAPV infection in honey bees . Our results showed that IAPV is established as a persistent infection in honey bee population and infects all developmental stages and different sexes of honey bees . The tissue tropism study showed that IAPv replicates within all bee tissues but tends to concentrate in gut and nerve tissues and in the hypopharyngeal glands . The highest titer of IAPV was observed in gut tissues and , in conjunction with detection of IAPV in colony food , suggests that food serves as a vehicle for within-colony horizontal transmission . The next highest titer of IAPV replication was observed in nerve tissue and indicates tropism of IAPV to the bee nervous system , consistent with observed pathologies . Specifically , while IAPV-infected bees in our study remained asymptomatic , infected bees can exhibit shivering wings and progressive paralysis , typical symptoms of nerve-function impairment [8] . The third highest titer of IAPV was identified in hypopharyngeal glands and may explain the presence of the virus in royal jelly , a product synthesized in these glands and fed to queens and larvae . Royal jelly , along with nectars shared among adult workers , thus provide an important route for viral movement within the colony . A previous study showed that honey bees became infected with IAPV after exposure to V . destructor that carried the virus [12] , illustrating vector-mediated horizontal transmission . In addition , the detection of IAPV in the digestive tracts and feces of queen bees along with detection of the virus in colony food supplies suggest a food-borne transmission pathway , arguably driven by frequent trophallaxis ( mouth-to-mouth sharing of food ) between colony members . The detection of IAPV in eggs and larvae not exposed to V . destructor that serves as a vector to facilitate the horizontal transmission of the virus to their honey bee hosts , together with detection of IAPV in queen ovaries suggests a vertical transmission pathway from queens to their progeny . Further , the detection of IAPV in drone semen , and in the spermatheca used to store sperm in queens for fertilizing eggs , suggests that venereal ( sexual ) infection is another plausible mechanism by which this virus is transmitted . We suspect that IAPV manifests itself in a way similar to the iflavirus Deformed wing virus . Namely , when colonies are healthy , the virus persists via vertical transmission and exists in a latent state without perturbing host immunity . When honey bees live under stressful conditions such as Varroa mite infestation and overwintering stress , the virus replicates quickly and becomes more infectious , leading to the death of hosts and possible collapse of the colony . RNA viruses are characterized by their quasi-species population structure , that is , clouds of genetically related variants that collectively determine pathological characteristics of the population [15] . Genome analyses of IAPV strains shows several lineages . Previous genetic analysis of IAPV suggested the existence of at least three distinct IAPV lineages , two of them present in the US [16] . Our phylogenetic analysis confirmed this finding but showed a long period of independent evolution of IAPV strains in different collections . Genetic variation may account for the difference in virulence properties and severity of disease manifestations among IAPV strains and in fact , Cornman et al . [17] noted an especially high rate of nucleotide divergence among IAPV isolates sequenced from heavily impacted populations . Future studies using a combination of genome sequencing and single-nucleotide polymorphism analyses based on sequencing RNA pools ( deMiranda et al . 2010 , Cornman et al . , 2013 ) , should provide more insights into the evolutionary history , functional variation , and pathogenicity of this virus . The rate of IAPV infection in brood was higher than in adult bees for both strong and weak colonies . IAPV infection triggers a more profound alteration of gene expression in adult bees than in brood , shown by the fact that the number of genes with altered expression was four times higher in adults than in brood . The gene expression data did not provide obvious clues to the molecular mechanism ( s ) underlying the maintenance of the viral latency in brood . Genes involved in immune response showed no clear trend in expression in IAPV-positive brood . Genes involved in host immunity were significantly invoked in IAPV-infected adults , indicating that IAPV infection triggers active immune responses in adult bees . The transition of the virus from latency to activation of host immune response was likely triggered by exogenous stressors that affect bees at the adult stage . The evidence that mitochondrial dysfunction was the most significantly affected canonical pathway in IAPA-infected adults suggests that IAPV likely caused pathogenesis of energy-related host processes and functions , a condition that tends to worsen host nutritional status and impair host defenses mechanisms [18] . JAK-STAT was reported to be involved in the control of the viral load in DCV-infected Drosophila [19] . Components of the JAK-STAT pathway were up-regulated in response to IAPV infection . Other signaling cascades such as mTOR and MAPK pathways reported to be involved in antiviral immune responses [20] , [21] , also showed expression changes in response to the IAPV infection . However , components of the Toll and Imd signaling pathways , implicated in antiviral immunity in insects [22] , [23] were not up-regulated by IAPV infection . Toll and Imd are not always linked with antiviral immunity and , in particular , these pathways were not a factor during infection of D . melanogaster by Drosophila C virus , a relatively close relative of IAPV [19] , suggesting that different viruses trigger distinct antiviral responses . Knowing which pathways respond specifically to viral infections will enable more targeted pharmacological or genetic control strategies . Our results show that silencing a putative immune-suppressive protein encoded by IAPV led to significant reduction in IAPV replication without detrimental effect on bee hosts . This suggests that IAPV may also encode an RNAi suppressor . RNAi technology has been employed in previous work to combat virus infection in honey bees . The injection and feeding of Remebee , a dsRNA homologous to IAPV has proven effective in not only reducing the intensity of IAPV infection in honey bees [24] , but also strengthening honey bee colonies [25] . A recent study showed that the feeding of siRNA targeting an Internal Ribosomal Entry Site ( IRES ) of IAPV required for protein translation could confer antiviral activity in bees [14] . That feeding siRNAs targeting VSR in this study led to suppressed IAPV replication reinforces the therapeutic potential of RNAi for treatment of viral diseases in honey bees , by showing that carefully designed constructs can temper a potent counter-response to the host immune system . Further exploration of antiviral effects of putative suppressors of RNAi of other bee viruses such as KBV and ABPV , which share the same sequence motif of DIEENPGP with IAPV , is warranted . IAPV has a longstanding presence in managed honey bees [26] . While IAPV is not consistently tied to CCD , its ability to cause increased mortality in honey bees has been firmly established . Our results showed that host health status and environmental conditions indeed play a critical role in IAPV infection dynamics . While the simultaneous presence of multiple viruses in honey bees makes Koch's postulates of disease causality difficult to fulfill [27] , the presence and diversity of viruses in bee colonies has high predictive value for colony mortality [28] . The negative correlation between the level of IAPV infections and the size of host populations , in combination with other stress factors , has significant negative impact on colony survival and is likely a contributing factor to poor winter survivorship of honey bee colonies . The present study provides an improved starting point for continuing studies of the virus-host interactions and for efforts to formulate strategies to reduce colony losses due to viral diseases .
A brood frame containing bee samples of various ages and food stores was removed from each of three declining colonies colony selected from each colony maintained in a northern California queen-breeding operation in Spring . Honey bees ( Apis mellifera ligustica ) of different developmental stages and sexes ( eggs , larvae , pupae , workers , adult drones , and queens ) and colony foods ( honey , pollen , and royal jelly ) , as well as parasitic mites , V . destructor , were sampled for the detection of IAPV infection using RT-PCR method . Clear fecal material , 20–25 µl per queen , was collected by isolating queens individually in a 100×15 mm petri dish for approximately 30 minutes to allow them to defecate . Approximately 20–25 ul of semen was also collected from 25 drones of each colony . To determine the ability of IAVP to spread and replicate within honey bee hosts , fifteen adult worker bees were collected from each of the three colonies maintained placed in two USDA Bee Research Laboratory apiaries in Beltsville , MD and identified to be IAPV positive and subjected to tissue dissection . Under a dissecting microscope , each worker was fixed on the wax top of a dissecting dish with steel insect pins and 10–15 µl of hemolymph was micropipetted from a small hole made with a sterile needle on the dorsal thorax . Following hemolymph collection , a dorsal mid-line cut was made from the tip of the abdomen to the head with scissors , and tissues including hemolymph , fat body , brain , salivary gland , hypopharyngeal gland , gut , nerve , trachea , and muscle were individually removed from each worker . The scissors and forceps were cleaned between dissections with a cotton pad soaked with 10% bleach ( 0 . 003 sodium hypochlorite ) and another soaked with 70% alcohol , followed by a final rinse in sterile water . To prevent contamination with hemolymph , all tissues were rinsed once in 1× phosphate-buffered saline ( PBS ) and twice in nuclease-free water . The washing solution was changed after each tissue collection to prevent cross-contamination . The same tissues of different bees of the same colony were pooled together for subsequent RNA extraction . All freshly dissected tissues were immediately subjected to RNA extraction and then stored in −80°C freezer in the presence Invitrogen RNaseOUT Recombinant Ribonuclease Inhibitor until quantitative examination of the tissue tropism by strand-specific TaqMan quantitative RT-PCR ( RT-qPCR ) . Additionally , twenty eggs were also collected from the colonies identified to be IAPV positive using a fine brush . The eggs were washed in 5% bleach solution for five minutes then rinsed in sterile water to eliminate surface contamination of the virus [29] . Queens from the same IAPV-positive colonies were collected and tissues of gut , ovaries and spermatheca were excised following the methods described above . Both eggs and queen tissues were fixed in 4% paraformaldehyde in 100 mM PBS ( pH 7 . 0 ) , then stored in 70% ethanol ( 200 Proof ) at 4°C until in situ hybridization ( ISH ) assays for localization of the virus . To determine seasonal activities and impacts of IAPV on honey bee health , samples of adult workers and brood ( 4th and 5th instar mature larvae , prepupae , and white-eyed pupae ) of A . mellifera ligustica were collected from ten bee colonies maintained in apiaries of the USDA Beltsville Bee Research Laboratory from March 2008 to February 2012 and were subject to RT-PCR assay for presence of IAPV . The experimental colonies were divided into healthy and weak colonies based on the size of adult populations , amount of sealed brood , and presence of food stores . 20 adult workers and 20 unsealed brood were collected individually from each of five strong and five weak colonies every month and examined for the virus infection individually . For each colony , the rate of the virus infection and strength of individual colonies were recorded every month and the infection rate was calculated based on percentage of tested bees ( adult or brood ) that were infected ( N = 20 ) . The average infection rate each month for both strong and weak colonies was calculated by combining the date from five colonies each month and four years of the same month ( N = 5×4 ) . The infection rates of IAPV were compared for colony strength ( healthy vs . weak ) , developmental stages ( adult vs . brood ) and months of the year . Because the data are binomial in nature ( for each sample , the number of uninfected of 40 total ) , analysis was based on a generalized linear mixed model ( because random effects were included ) , using the logit link and the R software ( R Core Team 2012 ) with the lme4 package [30] . The combination of lowest AIC and main effects retention ( i . e . preserve main effects in the model even if not significant as long as higher order terms involving these main effects were significant ) was used to select a model that captured the important features of the data . Invitrogen Trizol reagent was used for isolation of total RNA from whole bees and bee tissues , as well as from colony foods , queen feces , drone semen , and Varroa mites , in accordance with the manufacturer's instructions . After confirmation of IAPA positive status by RT-PCR , total RNAs intended for microarray analysis were further purified with Qiagen RNeasy Microarray Tissue Mini Kit . RNA integrity was assessed with a 2100 Bioanalyzer system ( Agilent Technologies , Palo Alto , CA ) and RNA Lab Chip . Only samples with an RNA integrity number ( RIN ) of 6 or more were used [31] . The Promega one-step access RT-PCR system ( Madison , WI ) was used for IAPV detection as previously described [32] . Negative and positive controls ( previously identified positive sample ) were included in each run of RT-PCR reaction . The specificity of the amplified products was confirmed by sequence analysis of PCR products . RNA samples extracted from different tissues of adult workers were analyzed for the abundance of negative-stranded RNA , a replicative intermediate form of positive strand RNA viruses , using strand-specific reverse transcription coupled with TaqMan quantitative PCR ( RT-qPCR ) [33] , [34] . For each tissue sample , the first strand of cDNA was synthesized from total RNA using Superscript III reverse transcriptase ( Invitrogen ) with Tag-sense primer , Tag-IAPV-F1 ( 5′-AGCCTGCGCACGTGG gcggagaatataaggctcag -3 ) , where the capitalized sequences of Tag were published by Yue and Genersch [35] . The resulting synthesized cDNAs were then purified using MinElute PCR purification kit ( Qiagen ) followed by MinElute Reaction Clean kit ( Qiagen ) to remove short fragments of oligonucleotides and residue of enzymatic reagents to prevent amplification of non-strand specific products [34] . The resulting cDNA derived from negative stranded RNA was amplified using the Platinum Taq High Fidelity Polymerase ( Invitrogen ) in a 25-ul reaction containing 2 µl cDNA , 0 . 25 µM of Tag primer ( 3′-AGCCTGCGCACCGTGG- 5′ ) , 0 . 25 µM of antisense primer , IAPV-R1 ( 5′-cttgcaagataagaaaggggg-3′ ) , a 0 . 2 µM TaqMan probe ( 5′ FAM - CGCCTGCACTGTCGTCATTAGTTA - TAMRA 3′ ) , 0 . 2 mM each dNTP , 1 units of DNA polymerase , 1× PCR buffer , and 2 mM MgCl2 . qPCR was carried out using a cycling sequence of 95°C for 2 min followed by 35 cycles of 95°C for 30 sec , 55°C for 30 sec and 68°C for 1 min , which was then followed by a final extension of 68°C for 7 min . To normalize the qPCR result , amplification of a housekeeping gene β-actin was performed for each sample with a previously reported primer set and dual-labeled probe [32] . After confirmation of the approximately equal amplification efficiencies of the RT-qPCR assay for both IAPV and β-actin ( Figure S2 ) , the concentration of IAPV in different tissues was interpreted using the comparative Ct method ( ΔΔCt Method ) . The mean value and standard deviations of triplicate measurements of IAPV in each tissue was normalized using the Ct value corresponding to the triplicate measurements of endogenous control , β-actin following the formula: ΔCt = ( Average Ct DWV ) − ( Average Ct β-actin ) . The hemolymph , with the lowest virus level of IAPV , was chosen as a calibrator . Each of the normalized target values was subtracted by the normalized value of the calibrator to yield ΔΔCt . The concentration of IAPV in each tissue was calculated using the formula 2−ΔΔCt and expressed as the fold-change . Purified IAPV amplicons from primer pair IAPVF1/R1were incorporated into a pCR2 . 1 TA cloning vector ( Invitrogen ) which has a T7 site downstream of the insert and the orientation of the inserts was determined by sequence analysis . Probe complementary of genomic RNA of IAPV was generated from linearized plasmid using DIG-RNA Labeling Kit ( T7 ) ( Roche Applied Science , Indianapolis , IN ) . Eggs and queen tissues , including spermathecae , ovaries and gut , were subjected to dehydration by successive incubation in ethanol ( 70% , 95% and 100% ) and xylol ( 2×5 min each ) and then embedded in paraffin . Paraffin sections were cut ∼3–5 micron thick and mounted on poly-L-lysinated slides and stored at 4°C overnight . The sections were then rehydrated through a descending concentration of ethanol ( 100% , 95% and 70% ) , dewaxed in xylol , treated with proteinase K ( 10 ug/ml ) for 30 minutes , and acetylated with 0 . 33% ( v/v ) acetic anhydride in 0 . 1 M triethanolamine-HCl ( pH 8 . 0 ) for ten minutes prior to hybridization . The sections were prehybridized in prehybridization solution ( 50% formamide , 5× SSC , 40 ug/ml salmon sperm ) at 58°C for 2 hours and incubated in hybridization buffer with Dig-labeled IAPV probe solution to a concentration of 100–200 ng/ml probe in pre-hybridization solution at 58°C overnight . Negative control reactions included regular dUTP instead of DIG-labeled viral probe . After hybridization , the sections were washed in low stringency wash solution ( 2× SSC , 0 . 1% SDS ) at room temperature for 5 minutes and washed twice in high stringency wash solution ( 0 . 1× SSC , 0 . 1% SDS ) at 52°C for 15 minutes , and finally incubated with alkaline phosphatase ( AP ) -labeled sheep anti-DIG antibody conjugate . The hybridization signals were detected with alkaline phosphatase ( AP ) -labeled sheep anti-DIG antibody conjugate ( Roche Applied Science ) . The conjugate solution was added to the dry sections and incubated at 4°C for 2 hours in a humid chamber . Color development was performed by adding the buffer solution containing nitroblue tetrazolium ( NBT ) and 5-bromo-4-chloro-3-indoyl phosphate ( BCIP ) to the tissue sections and incubating for 3–6 hours at room temperature with protection from light . Dark purple to blue coloring suggested the presence of the virus where the DIG-labeled probe bound directly to the viral RNA , while pink staining was shown in negative controls where no IAPV probe was included . To determine the levels of genetic diversity of IAPV , the complete genome sequences of IAPV strains from infected bees collected in MD , CA , and PA were determined by combining primer walking and long-range RT-PCR amplification using Invitrogen SuperScript One-Step RT-PCR System for Long Templates . The seven overlapping fragments of IAPV were amplified simultaneously . The sequences of the genome termini were determined by Invitrogen 3′ and 5′ RACE systems . The primers used to amplify overlapping long RT-PCR fragments and 3′ and 5′-RACE nested PCR were shown in Figure S1 . The information regarding sequences and genomic positions of primers used in this study is shown in Table 1 . Overlapping sequences were assembled into complete virus genomes using SeqMan ( DNASTAR , Madison , WI , USA ) . The entire genome sequences of IAPV isolates from this study , as well as IAPV strains identified in Australia , China , and Israeli and previously deposited in GenBank were compared with the first reported strain of IAPV ( GenBank Accession# NC_009025 ) in order to get a clear global picture of genetic diversity of IAPV strains . A phylogenetic tree was generated using all available complete genome sequences of IAPV . The sequence of KBV ( GenBank Accession# NC_004807 ) was used as an outgroup to root the tree . Phylogenetic analysis was conducted in MEGA4 [36] . A tree was built using the Neighbor-Joining method and the reliability of the phylogenies was assessed by bootstrap replication ( N = 1000 replicates ) . Node labels correspond to bootstrap support and those values >50% were regarded as providing evidence of phylogenetic grouping . The global host responses of honey bees to IAPV infection in both adult and brood stages were investigated using microarray analysis . Adult worker bees ( nurse bees inside the hive ) and brood ( 4th and 5th instar larvae prior to capping ) were collected from three colonies that were confirmed by RT-PCR to be infected with IAPV . The ubiquitous presence of DWV in both IAPV-positive and IAPV-negative bees was considered to be a background infection . Total RNAs from 10 IAPV-infected and 10 uninfected workers as well as 10 IAPV-infected and 10 uninfected brood were individually reverse transcribed into cDNA using Superscript III reverse transcriptase ( Invitrogen ) with random hexamers . The cDNA was labeled with Cytidine 3 ( Cy3 ) and Cytidine 5 ( Cy5 ) , respectively , and reversed for the dye-swap analysis . The Cy5- or Cy3-labeled cRNA were mixed in the same amount and hybridized to honey bee oligonucleotide microarray slides fabricated at the University of Illinois . Slides were hybridized , washed , dried , and scanned using methods previously described [37] . The signal intensities were normalized based on the mean signal intensity across all genes on the arrays . The signal-to-noise ratio ( SNR = <signal mean – background mean</<background standard deviation>] was then calculated for each spot to discriminate true signals from noise . Only spots with an SNR equal to or greater than 2 . 0 were considered positive . All negative , poor and empty spots were flagged and discarded . The normalized data were analyzed using Partek Genomics Suite Version 6 . 4 ( Partek Inc . , St . Louis , MO ) . Principal component analysis ( PCA ) and hierarchical clustering analysis were conducted with Partek with default settings . The fold changes of each gene expression in response to IAPV infection were calculated against the uninfected samples ( negative control ) . Statistically significant genes were identified using mixed model analysis of variance ( one-way ANOVA ) with the Benjamini & Hochberg false discovery rate set to ≤0 . 05 . The genes that displayed fold-changes of more than 1 . 5 ( False Discovery Rate adjusted ρ value≤0 . 05 ) and had putative D . melanogaster orthologs were analyzed by DAVID Bioinformatics Resources 6 . 7 ( http://david . abcc . ncifcrf . gov ) , and GO browser and search engine AmiGO ( http://www . geneontology . org ) to define identify enriched biological themes in gene lists of both adult and brood . Additionally , the genes homologous to the their Drosophila gene counterparts were further analyzed for canonical pathways , biological functions/diseases , and functional molecular networks by Ingenuity Pathway Analysis ( IGA ) ( Ingenuity Systems , Redwood City , CA ) . The Fisher's exact test was used to calculate a ρ- value to determine the probability that the association between the gene in the dataset and the predefined pathways and functional categories in the Ingenuity Pathway Knowledge Base is due to random chance alone . A list of 20 genes involved in host immune responses were validated by SYBR Green real-time qRT-PCR in IAPV infected adult bees . The primers used in qRT-PCR are included in Table 2 . The ΔΔCt method was chosen for interpretation of gene expression in response to IAPV infection following the same procedures described above . The approximately equal amplification efficiencies of the RT-qPCR assay for housekeeping gene β-actin and target immune genes were confirmed individually ( the slope of normalized Ct vs . log input RNA≤0 . 1 ) . The data output of each gene was expressed as a fold-change indicating whether the expression of the target gene in IAPV infected bees was up-regulated or down-regulated compared to the expression of the same gene in uninfected bees . Complete predicted protein sequences of IAPV ( NC_009025 . 1 ) , along with other honey bee viruses , including KBV ( NC_004807 . 1 ) , ABPV ( NC_002548 . 1 ) , CrPV ( NC_003924 ) , and DCV ( NC_001834 ) were retrieved from GenBank and scanned for the DvExNPGP sequence motif where the upstream sequences of the DvExNPGP motif was reported to encode a RNAi suppressor [13] . A DvExNPGP sequence motif was identified in IAPV and the upstream sequences of the DvExNPGP motif at the 5′ terminus of the IAPV genome was therefore assumed to be a putative IAPV-encoded suppressor ( Figure S1 ) . siRNA corresponding to upstream sequences of DvExNPGP was designed using online siRNA design tool siDirect version 2 . 0 ( http://sidirect2 . rnai . jp/ ) . The sequences of the siRNAs used in this study are as follows: 5′-UACAACUUAUUCAAGAAUCCA-3′ and 5′- GAUUCUUGAAUAAGUUGUACC-3′ . The chemically modified , 21-mer , double-stranded and in vivo ready siRNAs were synthesized in a 250 nmol scale by Ambion Life Technologies ( CA , USA ) . The impact of siRNA corresponding to a putative IAPV-encoded VSR on IAPV replication was investigated by a laboratory cage study as described previously [38] . Briefly , the frames with emerging brood were removed from the colonies left untreated for V . destructor for 2–3 moths and identified with IAPV infection by RT-PCR assay , and newly emerged bees were collected the following day . Forty bees were placed in each rearing cage for the assay . A scintillation vial filled with a 1∶1 ratio sucrose-water solution was inverted over the top of the rearing cup as provision for the caged bees . The caged bees were divided into four groups: Group-I consisting of siRNA-treated IAPV-infected bees not exposed to parasitic mites V . destructor , Group-II consisting of untreated IAPV-infected bees not exposed to V . destructor , Group-III consisting of treated IAPV-infected bees challenged by V . destructor , and Group-IV consisting of untreated IAPV-infected bees challenged by V . destructor . The Varroa mites used in the study were collected from a colony that was heavily infested with mites; both honey bees and mites were shown to be infected with IAPV using RT-PCR assay . Twelve Varroa mites were introduced to each cage to create 30% Varroa mite infection . For groups receiving siRNA feeding , siRNA was mixed with sugar water in the scintillation vials , resulting in a 50 nM/ul working concentration of siRNA . Ten experimental bees along with 3 mites were collected at day 1 , day 3 , day 5 , and day 7 post-treatment . The assay was repeated three times . The effect of silencing putative VSR on IAPV replication was analyzed by quantifying the titer of negative-stranded RNA of IAPV in bees from each group by real time RT-qPCR following the method described above . No specific permits were required for the described studies . Observations were made in the USDA-ARS Bee Research Laboratory apiaries , Beltsville , Maryland , USA; therefore , no specific permissions were required to be obtained for these locations . The apiaries are the property of the USDA-ARS and are not privately-owned or protected in any way . Studies involved the European honey bee ( Apis mellifera ) , which is neither an endangered nor protected species .
|
The mysterious outbreak of honey bee Colony Collapse Disorder ( CCD ) in the US in 2006–2007 has attracted massive media attention and created great concerns over the effects of various risk factors on bee health . Understanding the factors that are linked to the honey bee colony declines may provide insights for managing similar incidents in the future . We conducted this study to elucidate traits of a key honey bee virus , Israeli acute paralysis virus . We then developed an innovative strategy to control virus levels . The knowledge and insights gained from this study will have positive implications for bee disease management , helping to mitigate worldwide colony losses .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology",
"and",
"life",
"sciences",
"agriculture"
] |
2014
|
Israeli Acute Paralysis Virus: Epidemiology, Pathogenesis and Implications for Honey Bee Health
|
Formalin-inactivated Japanese encephalitis virus ( JEV ) vaccines are widely available , but the effects of formalin inactivation on the antigenic structure of JEV and the profile of antibodies elicited after vaccination are not well understood . We used a panel of monoclonal antibodies ( MAbs ) to map the antigenic structure of live JEV virus , untreated control virus ( UCV ) , formalin-inactivated commercial vaccine ( FICV ) , and formalin-inactivated virus ( FIV ) . The binding activity of T16 MAb against Nakayama-derived FICV and several strains of FIV was significantly lower compared to live virus and UCV . T16 MAb , a weakly neutralizing JEV serocomplex antibody , was found to inhibit JEV infection at the post-attachment step . The T16 epitope was mapped to amino acids 329 , 331 , and 389 within domain III ( EDIII ) of the envelope ( E ) glycoprotein . When we explored the effect of formalin inactivation on the immunogenicity of JEV , we found that Nakayama-derived FICV , FIV , and UCV all exhibited similar immunogenicity in a mouse model , inducing anti-JEV and anti-EDII 101/106/107 epitope-specific antibodies . However , the EDIII 329/331/389 epitope-specific IgG antibody and neutralizing antibody titers were significantly lower for FICV-immunized and FIV-immunized mouse serum than for UCV-immunized . Formalin inactivation seems to alter the antigenic structure of the E protein , which may reduce the potency of commercially available JEV vaccines . Virus inactivation by H2O2 , but not by UV or by short-duration and higher temperature formalin treatment , is able to maintain the antigenic structure of the JEV E protein . Thus , an alternative inactivation method , such as H2O2 , which is able to maintain the integrity of the E protein may be essential to improving the potency of inactivated JEV vaccines .
Japanese encephalitis virus ( JEV ) , the most important etiological agent of viral encephalitis in Asian countries , causes regular outbreaks in eastern and southeastern Asia , India , and more recently in Australia [1 , 2] . Annually , 30 , 000 to 50 , 000 Japanese encephalitis ( JE ) -confirmed cases are reported in the JEV endemic areas , and 20% to 60% of symptomatic CNS infections are fatal [3–6]; 25% to 50% of symptomatic survivors have long-term neurological sequelae [7] . Asymptomatic JEV infection is about a thousand-fold higher than confirmed cases [8–10] . JEV is transmitted by virus-infected Culex mosquitos from inapparently infected viremic-amplifying hosts such as pigs or aquatic birds to symptomatic accidental hosts , such as horses and humans . Migratory birds have been implicated as the source of virus been introduced into new geographic regions , and associated with JE epidemics and replacement of genotype III ( GIII ) - with genotype I ( GI ) - JEV from southeast Asia to east Asia [11 , 12] . The genome of JEV consists of a ~11-kb , positive-sense , single-stranded RNA , which is translated and processed by viral and host proteases to three structural proteins—capsid , precursor membrane/membrane protein ( prM/M ) and envelope glycoprotein ( E ) —and seven nonstructural proteins ( NS ) —NS1 , 2A , 2B , 3 , 4A , 4B and 5 . The mature virion consists of 180 E proteins forming 90 homodimers and 180 processed M proteins . The immature virion is formed by 60 E and prM hetero-trimers [13 , 14] . E protein is the most critical protein eliciting protective immunity in hosts after viral infection , offering critical protection in mice [15] and inducing protective antibodies in recovering humans [16] . The ectodomain of E protein can be separated into three structural domains: E domain I ( EDI ) to III ( EDIII ) . The fusion peptide in EDII elicits group cross-reactive non- or low-neutralizing antibodies; EDIII , the receptor-binding domain , elicits potent type-specific neutralizing antibodies; and EDI , the center domain connecting EDII and EDIII , elicits complex cross-reactive high- or non-neutralizing antibodies after viral infection [16–18] . Vaccination remains the most effective strategy to control JE epidemics [19] . Live-attenuated and formalin-inactivated JEV vaccines are available for human use , but only live-attenuated vaccines are available for domestic animals , such as swine and horses . The first generation inactivated JEV vaccine , developed by BIKEN in Japan , was the mouse brain-derived , formalin-inactivated GIII Nakayama strain; manufacture of this vaccine has ceased since 2005 because of undesirable adverse effects [20] . Second generation tissue culture-derived , formalin-inactivated SA-14-14-2 vaccines are formulated with aluminum-hydroxide–adjuvant ( IC51 or IXIARO ) . IC51 vaccine has been licensed for use in adult and children older than 2 months [21] . In addition , a live-attenuated JEV SA14-14-2 vaccine , developed in China , is used in some Asian countries such as China , India , and Nepal [22–24] . The vaccine effectiveness has been estimated to be 85% to 90% after two doses of inactivated Nakayama vaccine , and 91% after one dose of the live-attenuated SA14-14-2 vaccine [25–27] . Unlike the live-attenuated vaccine , the formalin-inactivated JEV vaccines require boost immunization to retain the protective neutralizing antibodies [22 , 28] . Significant numbers of JEV endemic countries still depend on the locally produced , mouse brain-derived formalin-inactivated GIII JEV vaccine to control JE epidemics [19] . Formalin is the chemical most commonly used for inactivation to manufacture viral vaccines such as hepatitis A virus , polio , influenza virus , rabies virus , and simian immunodeficiency virus [29–34] . Formalin reacts with amino acids of target proteins to form reversible Schiff-base adducts and non-reversible methylene bridges . It has also been used as isotopic agent to label protein by introducing isotope to specific amino acid and as a cell and tissue fixation agent . Formalin functions chemically when it is used to inactivate virus , and the chemical reaction may modify the antigenic structure of the virion [35 , 36] . It has been shown formalin inactivation alters antigenic properties and reduces the immunogenicity of vaccines , such as hepatitis A and B virus , polio virus , bovine herpes virus 1 and influenza virus in mouse models [37–41] . Formalin-inactivated JEV vaccine remains the most widely distributed vaccine used to control JE epidemics . However , the potential effects of formalin on the antigenic structure of JEV and the antibody profile elicited by this vaccine remain unclear . The use of a low concentration of formalin and short inactivation time can yield antigens capable of inducing high neutralizing titers in mice , but the association between these inactivation procedures and the alteration of antigenic structure of E and the antibody profile elicited by this vaccine remain undetermined [42] . In this study , we used a panel of E-specific , murine monoclonal antibodies ( MAbs ) to analyze the effect of epitope modification of JEV E protein in a formalin-inactivated commercial vaccine ( FICV ) and laboratory grown , formalin-inactivated GIII and GI viruses ( FIV ) . We showed that formalin-inactivation , indeed altered the binding pattern of a JEV-derived , serocomplex cross-reactive neutralizing antibody , T16 . Interestingly , antibodies recognizing formalin-modified epitope were significantly lower in titer and had weaker neutralizing activity in serum from mice vaccinated with FICV and FIV-Nakayama than with untreated control Nakayama virus ( UCV-Nakayama ) . H2O2 inactivated JEV and was a superior approach that retained the antigenic reactivity of the virus with all tested MAbs including T16 as compared to conventional inactivation methods such as formalin and UV .
Animal experiments were approved by the Institutional Animal Care and Use Committee ( IACUC ) of National Chung Hsing University , Taiwan ( Approval No: 101–88 ) , and performed according to a protocol , which adhered to principles in the Guide for the Care and Use of Laboratory Animals ( NRC 2011 ) and meet the requirement in an Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) . The serum samples used in this study were collected from anonymous children who had received JEV vaccination and were without JEV infection during 2010; they were part of an already-existing collection housed at Tungs’ Taichung Metroharbor Hospital in Taichung . The clinical protocol was reviewed and approved by the institutional review board of the hospital ( 99006 ) for serum sample collection . Serum was recovered from blood after clotting and then centrifuged , and stored at -70°C until use . Vero , COS-1 , and C6/36 cells ( kindly provided from Dr . Chang GJ of US CDC , Fort Collins , CO ) were grown in Dulbeco’s modified Eagle’s minimal essential medium ( DMEM , Gibco ) containing 5% , 10% , and 10% heat-inactivated fetal bovine serum ( FBS , Gibco ) , respectively . BHK cells ( kindly provided from Dr . Chen WJ of Chang Gung University , Taiwan ) were grown in Minimum Essential Medium ( MEM , Gibco ) with 10% heat-inactivated FBS . The JEV vaccine strains used were the GIII strains Nakayama and SA14-14-2 , naturally attenuated GIII T1P1 isolate [43] and GI circulating strain YL2009-4 [44] . The FICV used in this study was the mouse brain-derived , formalin-inactivated Nakayama virus vaccine manufactured by ADImmune Corp . in Taiwan . Monoclonal antibodies ( MAbs ) used for antigenic characterization were flavivirus group cross-reactive MAbs ( 4G2 , 6B3B-3 , 6B6C-1 and 23–2 ) , JEV serocomplex cross-reactive MAbs ( T16 , 2B5B-3 , 6B4A-10 , 1B5D-1 and 7A6C-5 ) and JEV-specific MAbs ( 2H4 and 2F2 ) [45–47] . Vero cells , infected with strains of JEV , namely Nakayama , SA14-14-2 , T1P1 , and YL2009-4 , at a multiplicity of infection of 1 ( MOI = 1 ) , were grown in serum-free medium ( SFM4MegaVir; HyClone , Logan , UT ) for 4 days . Supernatant was clarified by centrifugation at 10 , 000 rpm for 30 min; virion particles in the supernatant were pelleted by a second centrifugation at 19 , 000 rpm for 16 hr . Viral pellets from the second centrifugation were resuspended in 1X phosphate buffered saline ( PBS ) . These concentrated viruses were used to derive FIV . An amount of 37% formaldehyde ( Sigma-Aldrich , St . Louis , MO ) was diluted with 1X PBS to 0 . 5% and adjusted to pH 7 . 2 with 10 N NaOH ( Sigma-Aldrich , St . Louis , MO ) . The mixture was added to concentrated JEV viruses give a final formalin concentration of 0 . 05% . The formalin-treated virus ( FIV ) or untreated virus ( untreated control virus; UCV ) was incubated at 4°C for 49 days ( the manufacture procedure for FICV provided by Adimmune Corporation in Taiwan ) , or at 22°C for 10 days [42] . FIV and UCV samples incubated at 4°C were collected every week and stored at -70°C for analysis . Nakayama virus specimens were inactivated by short-wavelength UV light at a distance of 3 cm on ice for 30 min or with a final concentration of 3% H2O2 ( Fisher Scientific ) , pH 7 . 2 , at 22°C from 2 to 8 hr , then stored at -70°C . The residual infectious viral titers of FIV , UCV or UV- or H2O2-treated viruses were assessed by micro-plaque assay . Antigen-capture ELISA ( Ag-ELISA ) , described previously [48] , was used to estimate E proteins concentrations in samples with anti-JEV mouse hyper-immune acitic fluid ( MHIAF ) ( immunized with purified and live JEV ) and determine the binding activity of MAbs . Briefly , a 96-well plate ( Sigma-Aldrich , St . Louis , MO ) was coated with rabbit anti-JEV polyclone ( generated from rabbit immunized with pVAX-JEi VLP-expressing plasmid [49 , 50] , and obtained from Dr . Chang GJ of US CDC , Fort Collins , CO ) at 37°C for 1 hr , blocked with StartBlock blocking buffer ( Pierce , Rockford , Ill . ) , then antigen was added at 40 ng per well for incubation at 4°C overnight . Antigen was incubated with MAbs and MHIAF , diluted with 5% skim milk , at 37°C for 1 hr , then peroxidase-conjugated goat anti-mouse IgG ( H+L ) ( Jackson ImmunoResearch , West Grove , PA ) at 37°C for 1 hr . Finally , 3 , 3’ , 5 , 5’-tetramethylbenzidine substrate ( TMB; Neogen Corp . , Lexington , KY ) was added at 100 μl per well for the color reaction and reactions were stopped with 2N H2SO4 added at 50 μl per well; the OD450 values were recorded . The antigen concentration of UCV and FIV was estimated by the OD450 of the MHIAF . The MAb binding activities for UCV or FIV were determined by percentage reactivity estimated by the OD450 of UCV or FIV at the time relative to that at 0-day , respectively . All binding activities were adjusted by fold difference of antigen concentration , estimated by the OD450 of MHIAF against UCV or FIV at the time relative to that at 0-day , respectively and shown as mean±SD of two duplicates of two independent assays . JEVs and JEV virus-like particles ( VLPs ) were mixed with 5X SDS non-reducing sample buffer ( 315 mM Tris , pH 6 . 8 , 50% glycerol , 5% SDS , 0 . 025% bromophenol blue ) , then loaded onto a 10% SDS gel . After separation , proteins were transferred to a nitrocellulose membrane , which was blocked with 5% skim milk . The proteins on the membrane were detected by use of mouse anti-JEV polyclonal antibody , MAbs or MHIAF and visualized by incubation with peroxidase-conjugated goat anti-mouse IgG ( H+L ) ( Jackson ImmunoResearch , West Grove , PA ) ; bands were developed by use of the LumiGOLD ECL Western Blotting Detection Kit ( SignaGen Laboratories , Gaithersburg , MD ) . The intensities of bands were calculated by use of ImageJ version 1 . 44 ( NIH , Bethesda , MD ) . The binding activity of anti-EDII 101/106/107 and anti-EDIII 329/331/389 antibodies against VLP was eliminated by introducing 101/106/107 and 329/331/389 mutations on VLP , respectively . JEV VLPs were produced with the pVAX-JEi plasmid derived from the pCBJE plasmid [50] , which encodes the prM and E protein regions of the SA14 strain genome . This plasmid was also used as the template for introducing mutations into the E protein by use of a site-directed mutagenesis kit ( Stratagene , La Jolla , CA ) as described [50 , 51] . The pVAX-JEi 101/106/107 , 306 , 329 , 331 , 332 and 389 amino acid mutants were introduced by mutagenesis primers ( S1 Table ) , according to the manufacturer’s protocols , and mutation was confirmed by sequencing . The JEV VLP-expressing plasmids were electroporated into COS-1 cells by use of a 0 . 4-cm–electrode-gap cuvette and a Bio-Rad Gene Pulser II ( Bio-Rad Laboratories , Hercules , CA ) at 250 V and 975 μF; electroporated cells were recovered overnight at 37°C and incubated at 28°C to enhance VLP secretion . The secreted VLPs were analyzed by Ag-ELISA and used to evaluate the presence of epitope-specific antibodies . To measure the neutralizing activity of the MAbs or in serum samples , briefly , 2 . 48×104 Vero cells were added into 96-well plates for 24 hr at 37°C with 5% CO2 . MAbs in pre-attachment assay or serum samples were inactivated at 56°C for 30 min , diluted in a two-fold series , mixed with 100 pfu JEV Nakayama strain for 1 hr , then shaken every 20 min . Monolayers of Vero cells were infected with the virus-antibody mixture for 1 hr at 37°C with 5% CO2; in contrast , in post-attachment assay , virus was bound on monolayers of Vero cells at 4°C for 1 hour , then incubated with a two-fold series diluted and inactivated MAbs at 4°C for 1 hour , and then was shift into 37°C incubator with 5% CO2 for 1 hour . After incubation , 1% methyl cellulose in DMEM containing 2% FBS was added to the 96-well plates for incubation for 36 hr at 37°C with 5% CO2 , then plates were washed with PBS , fixed with 75% acetone , and air-dried in a hood . The fixed cells were stained with anti-JEV MHIAF for 40 min at 37°C . After a washing , peroxidase-conjugated goat anti-mouse IgG ( Jackson ImmunoResearch , West Grove , PA ) was added for 40 min , and virus-infected foci were identified by use of a Vector-VIP peroxidase substrate kit SK-4600 ( Vector Laboratories , Burlingame , CA ) . The foci were counted manually under microscopy and used to calculate a sigmoidal dose-response for the focus reduction micro-neutralization test ( FRμNT50 ) titers with use of GraphPad Prism v5 . 01 . Groups of 6-week old BALB/c mice ( n = 5 mice per group ) were vaccinated with three doses of Freund’s incomplete adjuvanted UCV-Nakayama , FIV-Nakayama or FICV . The first booster vaccination was given 2- week after the primary immunization and followed by a final booster at 4 weeks after the first booster vaccination . Serum samples were collected 2 weeks after the final booster vaccination . An IgG antibody-capture ELISA ( GAC-ELISA ) described previously [52] was used to determine the titer of the epitope-specific antibodies in immunized mouse serum . Briefly , goat anti-mouse IgG ( H+L ) ( KPL , Gaithersburg , MD ) was coated on 96-well plates at 37°C for 1 hr , then plates were blocked with StartBlock blocking buffer ( Pierce , Rockford , Ill . ) . Serum samples were serially diluted with wash buffer and added to plates at 37°C for 90 min . After a washing , 40 ng of the JEV antigens , wild-type ( WT ) or mutant JEV VLPs was added and mixtures were incubated at 4°C overnight . The IgG-capture antigens were detected by use of rabbit anti-JEV polyclonal antibody and peroxidase-conjugated goat anti-rabbit IgG ( H+L ) ( Jackson ImmunoResearch , West Grove , PA ) to detect antigen-bound rabbit anti-JEV polyclonal antibodies . The above steps were described for the Ag-ELISA . Total anti-immunogen IgG antibody was determined as endpoint titer . The epitope-specific antibody response was determined by the decreased reactivity titer against mutant VLP compared to WT VLP and was calculated as endpoint titer of ( WT—mutant VLP ) . The epitope-specific neutralizing antibody activity was determined as follows . The pVAX-JEi WT and EDIII 329/331/389 mutant plasmids were electroporated into COS-1 cells , cultured for 24 hr , and cells were resuspended in PBS . Diluted mouse serum samples were mixed with 107 transformed COS-1 cells at 37°C on a shaker for 2 hr . Then the transformed COS-1 cells were removed by centrifugation at 3 , 000 rpm for 10 min and the supernatant containing unbound antibodies was serially diluted and mixed with 100 plaque-forming units ( pfu ) of JEV Nakayama strain at 37°C for 1 hr . The neutralizing activity was determined by FRμNT assay as described previously and neutralizing activity ( % ) calculated by [1- ( plaque numbers of serum mixed with virus/plaque numbers of virus-only control ) ]* 100 . The curves showing the neutralizing antibody activity at different dilutions were fitted by non-linear regression in GraphPad . The percentage of neutralizing antibodies that recognized the JEV EDIII 329/331/389 epitope was calculated by plaque numbers of serum post-adsorbed with [ ( WT VLP—EDIII 329/331/389 mutated VLP ) / ( WT VLP—COS-1 cells ) ]* 100 . The pVAX-JEi WT and EDIII 329/331/389 mutant plasmid-transformed COS-1 cells were seeded into wells of a chamber slide ( Millipore , Billerica , MA ) , and cultured at 37°C overnight , then wells were fixed with 4% paraformaldehyde ( Sigma , St . Louis , MO , USA ) in PBS at room temperature for 20 min and washed with PBS . The fixed cells were made permeable by treatment with 0 . 1% Triton X-100 at 4°C for 5 min and washed with PBS , then wells were blocked with 3% bovine serum albumin ( Sigma , St . Louis , MO , USA ) in PBS at 37°C for 1 hr . Wells were stained with anti-JEV MHIAF and reacted with FITC-conjugated goat anti-mouse IgG ( KPL , Gaithersburg , MD ) in 1% Evans blue . Images were viewed under an OLYMPUS CKX41 microscope . Data are presented as mean±SD from two repeated experiments . Two-tailed Student’s t test was used for all analyses , and statistical significance was set at p <0 . 05 .
Most of the antibodies elicited by JEV infection or immunization are conformation-dependent and , for the most part , recognize the viral E protein and are able to help prevent virus infection [15 , 53] . Formalin inactivation of several human vaccines has been shown to result in antigenic alteration to the viral particles , which can be measured by the binding activity of specific MAbs [38 , 40] , but the effect of formalin inactivation on commercially available JEV vaccines has not been evaluated . Previously , an established Ag-ELISA protocol was successfully used to determine the antigenic structure of JEV using a panel of anti-E protein MAbs [49 , 53] . First , we evaluated the antigenic differences between the live Nakayama virus and FICV by Ag-ELISA using a panel of eleven anti-flavivirus E-protein MAbs [45–47] . The same antigen concentration ( estimated by Ag-ELISA using JEV-specific MHIAF ) of live virus and FICV was used throughout the experiments . Live Nakayama virus and FICV showed similar binding pattern for ten of the eleven tested MAbs , with the exception being T16 MAb ( Fig 1 ) . The binding activity with T16 MAb was significantly lower for FICV than for the live Nakayama virus ( p<0 . 05 ) with end-point titers of 105 . 39 and 103 . 48 for the live Nakayama virus and FICV respectively . The MAb binding pattern suggests that the antigenic structure of FICV differs from that of the live Nakayama virus . The decrease in binding activity of T16 MAb against FICV might be due to procedure variation during vaccine manufacture , which include differences in the formalin inactivation , differences in the virus purification process , changes in the sub-strains used and differences in the passage history of the Nakayama virus used between the vaccine production virus strain and the live Nakayama virus used in this experiment . To rule out the potential influence of sub-strain differences on the E structure and focus on the effect of formalin treatment on antigenic modification of Nakayama virus , we subjected laboratory-grown concentrated Nakayama virus to either formalin inactivation ( FIV-Nakayama ) or without formalin at 4°C for 49 days ( untreated control virus-Nakayama virus , UCV-Nakayama ) . The antigenic reactivity of FIV-Nakayama and UCV-Nakayama , as determined by Ag-capture ELISA with anti-JEV MHIAF , remained constant ( Fig 2A ) . The infectivity of FIV-Nakayama decreased drastically to below the detection limitation after 7 days treatment under these conditions; however , the infectivity of UCV-Nakayama decreased gradually over time and only became undetectable after 49 days ( Fig 2B ) . Ten of the eleven MAbs , with the exception of T16 , showed similar binding activity with FIV-Nakayama and UCV-Nakayama preparations collected at most time points by Ag-ELISA ( Fig 2C ) . The binding activity of the JEV serocomplex cross-reactive T16 MAb against FIV-Nakayama was significantly decreased at the 14-day collection point ( 14-DC ) with this sample having only 83% of the binding activity of the 0-DC sample . This decrease in binding of T16 against FIV-Nakayama was time-dependent; with only 55% binding activity remaining at 49-DC ( Fig 2C , panel e ) . Unlike T16 , 2B5B-3 and 2F2 binding against FIV-Nakayama declined at an early time point compared to UCV-Nakayama , but this was not observed at later time points . This result is consistent with the observation that only T16 exhibiting a decreased binding activity against FICV by Ag-ELISA ( Fig 1 ) . Therefore , we believed that the decreased binding activity of T16 MAb against FICV is the result of formalin inactivation and is not related to potential antigenic differences related to the sub-strain of virus or associated with the passage history of virus . To rule out formalin-induced antigenic modification occurring at strain-specific amino acids [36 , 54] , we prepared three different strains of JEV , the SA14-14-2 GIII vaccine-strain virus , the T1P1 naturally attenuated GIII virus and the YL2009-4 GI virus [44] , and then applied the same formalin inactivation procedures to all three viruses; this was followed by measurement of their MAb binding by Ag-ELISA using a subset of eight MAbs . The pattern of MAb binding activity obtained with these viruses was similar to that obtained with the Nakayama virus with or without formalin treatment ( Fig 3 and S1 Fig ) . Again , at 49-DC , the binding activity of T16 MAb was significantly decreased to 75% , 75% , and 72% for FIV-SA14-14-2 , FIV-T1P1 , and FIV-YL2009-4 , respectively ( Fig 3A ) . To further confirm that the decrease in binding activity of T16 MAb against the E protein was due to formalin inactivation , these viral preparations with or without formalin treatment were analyzed by non-reducing SDS PAGE followed by Western blotting using the T16 . 4G2 and 7A6C-5 MAbs , which have similar Ag-ELISA binding activity against the FIV-JEV and the UCV-JEV antigens , were included for comparison ( Figs 2C and 3 and S1 Fig ) . The intensity of the E protein band , when detected by 4G2 and 7A6C-5 of the various FIV-JEV and UCV-JEVs , including Nakayama , SA14-14-2 , T1P1 , and YL2009-4 , were similar; however , the intensity detected by T16 was lower against the FIV-JEVs than against the UCV-JEVs ( Fig 3B ) . By way of comparison , at 49-DC , the formalin-treated Nakayama , SA14-14-2 , T1P1 , and YL2009-4 viruses were found to have reduced T16 binding intensities of only 36% , 38% , 57% , and 40% of the UCV-JEVs , respectively ( Fig 3C ) . Therefore , formalin inactivation , when it affects the antigenic structure of JEV E protein , would seem not to be viral strain-specific and is likely to occur at the virion level , perhaps affecting the E monomer containing disulfide bonds . To localize the formalin-modified epitope on E protein , we mapped the epitope recognized by T16 MAb . The antigenic structure of non-infectious JEV VLP is similar to that of the virion particle [50 , 51] . T16 MAb is a JEV serocomplex cross-reactive antibody and we previously found that amino acid residues 101 , 104 , and 106 , which are present in EDII , and amino acid residues 315 , 331 and 389 , which are present in EDIII , are important for the binding of JEV serocomplex cross-reactive MAbs [49 , 53] . Thus we used a VLP-expressed plasmid to locate the formalin-modified epitope recognized by T16 MAb . JEV VLPs with EDIII amino acid substitutions S329A , S331K , and D389G , but not JEV VLPs with amino acid substitutions W101G/G106K/L107D , E138K , E306G , A315G and D332R , showed decreased binding to T16 MAb ( Fig 4A ) . Amino acids 329 and 331 are located within the BC loop of the EDIII of JEV and amino acid 389 is located within the FG loop of the EDIII of JEV; these three amino acids are likely candidates to undergo modification during formalin inactivation ( Fig 4B ) . The BC loop of EDIII contains critical residues recognized by neutralizing MAbs against JEV and West Nile virus , while the FG loop of EDIII is involved in host tropism [55–57] . Therefore , we analyzed the neutralizing ability of T16 by FRμNT assay ( Fig 5A ) . It was found that the FRμNT50 potency of T16 MAb was 21 . 8 μg/ml . 4G2 MAb neutralizes and inhibits flaviviral infection at the post-attachment step [58] . Thus , we used both 4G2 and T16 MAbs to determine the mechanism of viral neutralization . MAb was used to bind to JEV before infecting Vero cells in order to carry out a pre-attachment assay . Alternatively , MAb was added to JEV-bound cells in order to carry out a post-attachment assay . The neutralizing patterns of 4G2 and T16 MAbs were similar regardless of whether either MAb was added before or after viral attachment ( Fig 5B ) , which indicates that T16 MAb seems to inhibit JEV at a post-attachment step . To evaluate the influence of the T16 epitope ( EDIII 329/331/389 ) on the immunogenicity of formalin-treated JEV antigens , we further investigated IgG antibody responses and the properties of antibodies against EDIII 329/331/389 in vaccinated mice . Female BALB/c mice were vaccinated with UCV-Nakayama , FIV-Nakayama , or FICV and then post-vaccination serum samples were analyzed by the IgG-capture ELISA using wild-type , EDIII 329/331/389-mutated and EDII 101/106/107-mutated VLPs . The EDII 101/106/107-mutated VLPs eliminate the immunodominant B-cell epitope , conserved in all flaviviruses as well as inducing cross-reactive , non-neutralizing and/or low-neutralizing antibodies [49 , 59] . The total JEV-specific IgG elicited by all three immunogens were similar ( p>0 . 05 ) with the average titer end-points being 8 . 5×103 , 1 . 5×104 , and 1 . 1×104 for UCV-Nakayama , FIV-Nakayama , and FICV-immunized mice , respectively ( Fig 6A , panel a ) . We determined the antibody responses that recognized the EDII 101/106/107 epitope and the EDIII 329/331/389 epitope by calculation the decreased reactivity titer against EDII 101/106/107-mutated and EDIII 329/331/389-mutaed VLP compared to the wild-type VLP . The titer of antibodies recognizing the EDII 101/106/107 epitope was similar ( p>0 . 05 ) for all serum from mice vaccinated with UCV-Nakayama ( 103 . 9 , range 103 . 2–104 . 3 ) , FIV-Nakayama ( 104 . 1 , range 103 . 7–104 . 5 ) , and FICV ( 103 . 9 , range 103 . 1–104 . 5 ) ( Fig 6A , panel b ) . In contrast , the titer of EDIII 329/331/389 epitope-specific antibodies was significantly lower ( p<0 . 05 ) for serum from mice vaccinated with FIV-Nakayama ( 102 . 7 , range 102 . 4–103 ) and FICV ( 103 , range 102 . 5–103 . 4 ) compared to UCV-Nakayama ( 103 . 6 , range 103−104 . 1 ) ( Fig 6A , panel c ) . Based on the above , we suspected that FICV-immunized children might produce a similarly lower proportion of EDIII 329/331/389 epitope-specific antibody . Twelve FICV-immunized children serum samples were found to show a lower level for the EDIII 329/331/389 epitope-specific IgG antibodies , namely 23% ( 10–46% ) ( S2 Table ) ; this result closely resembles the antibody reactions elicited in the FIV-Nakayama–immunized and FICV-immunized mice . To confirm the results obtained by epitope-specific IgG ELISA , the FIV-Nakayama and FICV immunized mouse serum samples were examined by Western blot analysis using the same concentration of WT , EDII 101/106/107-mutated VLP and EDIII 329/331/389-mutated VLP ( Fig 6B ) and the results quantified against standardized protein concentrations ( Fig 6C ) . The prM protein of all of the JEV VLPs , including the WT antigen , the EDII 101/106/107-mutated antigen and the EDIII 329/331/389-mutated antigen , were equally recognized by the anti-JEV MHIAF . Furthermore , the EDII 101/106/107-mutated VLP and EDIII 329/331/389-mutated VLP could not be recognized or showed significantly decreased recognition with the 4G2 and T16 MAbs , namely <1% and 36% reactivity , respectively . The serum collected from mice vaccinated with UCV-Nakayama was less able to bind to the EDIII 329/331/389-mutated VLP ( 35% ) than FIV-Nakayama and FICV ( 65% and 64% , respectively ) , but this was not true for the EDII 101/106/107-mutated VLP ( 12% , 11% , and 17% , respectively ) ( Fig 6B and 6C ) . The results of the epitope-specific IgG ELISA and Western blot analysis are consistent and indicate a stronger immunogenicity of the EDIII 329/331/389 epitope on UCV-Nakayama than that on FIV-Nakayama and FICV . Therefore , formalin-inactivated Nakayama virus or vaccine in immunized mice was only able to affect the induction of antibodies recognizing the EDIII 329/331/389 , but was not able to affect the induction of antibodies recognizing the EDII 101/106/107 epitope . The protective efficacy of vaccines against JEV infection is positively associated with the presence of neutralizing antibodies in mice [60] . Based on this we evaluated the correlation between the production of neutralizing antibodies binding to EDIII 329/331/389 across various vaccines . The contribution of EDIII 329/331/389-specific antibodies to the viral neutralizing activity was determined by FRμNT using mouse serum specimens elicited by UCV-Nakayama , FIV-Nakayama or FICV . Serum samples were pre-adsorbed with the same number of normal COS-1 cells ( adsorption control ) , or COS-1 cells expressing WT or EDIII 329/331/389-mutated JEV VLPs . The level of VLP-expressing COS-1 cells was estimated by staining with anti-JEV MHIAF at 24 hr after transformation with the JEV WT or EDIII 329/331/389 mutant plasmid . The IFA positive rates were similar at about 85% for COS-1 cells transformed with either of the plasmids ( S2 Fig ) . The pre-adsorption neutralizing antibody titers of mouse serum immunized with UCV-Nakayama , FIV-Nakayama or FICV were similar , with FRμNT50 titers of 52 ( 20–80 ) , 46 ( 20–160 ) , and 52 ( 20–80 ) , respectively ( Fig 7A ) . We then measured the post-adsorption FRμNT50 titers ( Fig 7B ) to determine the contribution of the EDIII 329/331/389-specific antibodies to the viral neutralizing activity . The neutralizing antibody titers were lower for serum post-adsorbed with the WT JEV VLP-expressing COS-1 cells than for serum post-adsorbed with normal COS-1 cells using serum samples elicited by all three vaccines ( Fig 7B ) . However , the post-adsorption serum specimens using JEV EDIII 329/331/389-mutant VLP-expressing COS-1 cells showed a significant reduction in their neutralizing antibody titers activity when serum from either FIV-Nakayama-immunized mice or FICV-immunized mice was used , but not when the serum from UCV-Nakayama–immunized mice was used . The differences in neutralizing activity of the serum samples after adsorption with COS-1 cells expressing the WT VLP or EDIII 329/331/389-mutant VLP may have been due to the contribution made by EDIII 329/331/389-specific antibodies . When the results were fitted using non-linear regression analysis ( Fig 7C ) , this showed that the contribution of EDIII 329/331/389-specific antibodies to neutralizing antibody activity was proportionally higher ( 69% , range 62–78% ) using the serum from UCV-Nakayama-immunized mice than when the serum from FIV-Nakayama-immunized mice ( 38% , range 31–48% ) or FICV-immunized mouse serum ( 44% , range 35–58% ) was used . Thus , formalin modification of the EDIII 329/331/389 epitope would seem to affect the production of neutralizing antibodies . A previous report has suggested that JEV inactivation by formalin at 22°C for 10 days might be more immunogenic than inactivation at 4°C for 49 days [42] . Therefore , we asked if inactivation temperature ( 4°C vs . 22°C ) and inactivation duration ( 49 days vs . 10 days , respectively ) is able to influence T16 modification . We measured the T16 MAb binding activity of the JE Nakayama , SA14-14-2 , T1P1 , and YL2009-4 viruses treated with formalin at 4°C or 22°C for 10 days ( Fig 8A ) . At 10-DC , the T16 MAb binding activity against FIV-Nakayama , FIV-SA14-14-2 , FIV-T1P1 , and FIV-YL2009-4 were all lower at 75% , 77% , 63% , and 43% at 22°C than at 4°C , where the results were 94% , 98% , 120% , and 94% , respectively . Thus T16 epitope modification is present on FIV-JEVs treated either at 22°C for 10 days ( remaining 43–77% of T16 MAb binding activity ) or at 4°C for 49 days ( remaining 55–75% of T16 MAb binding activity ) ( Figs 2C and 3A ) . UV has also been used to inactivate viruses in the past and such UV-inactivated viruses are able to induce protective humoral immunity [61 , 62] . Surprisingly , UV-inactivated Nakayama virus only weakly bound anti-JEV MHIAF and T16 MAb when assessed by Western blot analysis , which indicates that the antigenic structure of Nakayama virus might be severely altered by UV irradiation ( Fig 8B ) . Hydrogen peroxide ( H2O2 ) can be used as a biocide and is known to interact with amino or sulfhydryl groups on antigens . Amanna et al . recently reported that H2O2-inactivated viruses are still able to induce protective cellular and humoral immunity [63 , 64] . Therefore , we followed their protocol and inactivated the Nakayama virus with 3% H2O2 at 22°C from 2 to 8 hr . Two-hours of H2O2 treatment reduced viral infectivity by at least 42000-fold to under the detection limitation ( Fig 8C ) . Ag-capture ELISA revealed that the binding activities of the T16 and other cross-reactive MAbs against the UCV-Nakayama and the H2O2-treated Nakayama virus were the same after 2-hr of treatment at 22°C . This suggests the antigenic structure of the Nakayama virus remained intact after H2O2 inactivation ( Fig 8D ) .
Several countries , including Japan , South Korea , and Taiwan , have successfully reduced the number of JE clinical cases by using inactivated JEV vaccines , but more effective and safe alternative vaccines are still needed [65–67] . The factors that affect the effectiveness and safety of JEV vaccines are the virus strain , method for viral cultivation , vaccine purity and vaccine formulation [68]; however , the effect of formalin inactivation on the quality of vaccine has never been studied . This is important because formalin-induced hypersensitivity has been found associated with risk of enhanced disease during subsequent infection with respiratory syncytial virus ( RSV ) , and formalin inactivation altered the antigenicity of poliovirus [38 , 41 , 69] . Antigenic characterization of formalin-inactivated poliovirus vaccine by using a panel of MAbs revealed that modification of antigenicity is time-dependent [38] . Using a previously collected panel of anti-flavivirus MAbs and the established Ag-ELISA [49 , 53] , we found that only the T16 MAb binding domain was time- and temperature-dependently altered by formalin inactivation . This observation suggested it might be valuable to evaluate the effect of residual formalin in formulated bulk on vaccine shelf life in the future . Importantly , regardless of the JEV strain used , formalin treatment altered the T16 epitope of all tested JE viruses . In contrast , epitopes recognized by 2B5B-3 and 2F2 MAbs on FIV-Nakayama were temporarily modified for specimen collected at early time point . Modification of these two epitopes was Nakayama strain-specific and was reversible since this phenomenon was only observed in the early time point specimen of formalin-treated Nakayama alone . Among anti-flavivirus antibodies , most of the virus-specific , non–cross-reactive , and EDIII-recognizing antibodies have strongly neutralizing activity , and most of the cross-reactive and EDII- or EDI-recognizing antibodies have weak or no neutralizing activity [16 , 70] . T16 MAb is a JEV-derived , JEV-serocomplex cross-reactive antibody . It shows weakly neutralizing activity at the post-attachment step in vitro . However , antibodies that comprise a large portion of the antibody response after WNV infection have only weak neutralizing activity in vitro but still provide therapeutic protection in vivo via the immune complement system [71] . The amino acid residues in both EDII and EDIII of the E protein are important to the binding of JEV serocomplex cross-reactive MAbs [49 , 53] . We determined the binding of T16 MAb to JEV VLPs by the amino acid positions EDII-104 , -329 , -331 , and -389 but used only EDIII 329/331/389-mutated VLPs to analyze epitope-specific antibody responses because EDII-104–mutated VLPs showed reduced secretion . Interestingly , the T16 epitope overlaps with the JEV-specific highly neutralizing E3 . 3 MAb epitope [56] . This result provides additional support that most E-protein epitopes within flaviviruses are overlapping [53] . Formalin is known to mainly react with the amino and thiol groups of amino acids to form methylol groups , which is followed by the formation of Schiff-base adducts; this reaction is reversible . These Schiff-bases adducts can cross-link to functional groups of various amino acids , such as arginine , tyrosine , tryptophan , histidine , glutamine , lysine , and cysteine , forming non-reversible methylene bridges [35] . Thus , the epitope of T16 MAb , namely glycine 104 , serine 329 , serine 331 and aspartic acid 389 , are likely not directly modified by formalin but are possibly influenced by nearby amino acids , including those at 105 , 335 , 336 , 387 , 390 , and 391 , and such cross-linking might directly or indirectly affect the conformational structure of the T16 epitope . Formalin treatment did not alter T16-overlapped epitopes recognized by 4G2 and 6B6C-1 . T16 MAb might be more sensitive to this formalin-generating modification on the non-overlapped residue ( s ) essential for T16 recognition . Currently we still do not know residues specifically reacting with formalin . Structural differences and amino acid variation in flavivirus immunogens , such as whether the virions are mature or immature , VLPs , or EDIII alone , may also affect the immunogenicity , antibody profile , and neutralizing potency elicited [72–75] . For example , EDIII-reacting antibodies show high neutralizing potency , but the recombinant EDIII immunogen induces low avidity and low titers of neutralizing antibodies against the virus [72] . In this study , we found that formalin inactivation altered the structure of the JEV E protein and thus affected the profile of induced antibodies . In this study , T16 epitope was the only epitope affected by the formalin inactivation; however , whether the T16 epitope is the only E structure alteration affecting the profile of antibodies elicited by formalin-inactivated vaccines/viruses is unknown because the T16 epitope , EDIII 329/331/389 , was not directly reactive with formalin . The formalin-modified EDIII 329/331/389 region was found less immunogenic and had less of a contribution to the neutralizing activity , despite non-significant differences in neutralizing antibody titers among UCV-Nakayama–immunized mice and FIV-Nakayama–immunized mice . Weak-neutralizing and non-neutralizing epitopes were located in the fusion peptide , and the introduction of mutations into the fusion peptides of the VLP disrupted the binding activity of anti-fusion loop MAbs . The fusion peptide mutant reduced the immunogenicity of the fusion peptide but retained its ability to evoke neutralizing antibodies [76 , 77] . Thus , the formalin-modified region affects the profile of vaccine-induced antibodies and alters the distribution of neutralizing antibodies . We did not determine the effect of formalin inactivation on the T-cell response , which needs to be addressed because a negative effect of formalin-inactivation on the influenza-virus T-cell response has been documented and T-cell immunity plays a role in how vaccines protect against JEV infection [37 , 78 , 79] . The use of epitope scaffolds or deglycosylation has successfully exposed immunorepressive and cryptic epitopes and enhanced immunogenicity in HIV or redirected the antibody response in simian immunodeficiency virus [80 , 81] . We found the titers of EDIII 329/331/389-reactive antibodies higher among UCV-Nakayama–than FIV-Nakayama–or FICV-immunized mice and use of EDII 101/106/107-reactive antibodies gave similar results . Previously , we found that EDII 101/106/107 and EDIII 329/331/389 form an overlapped epitope for flavivirus group cross-reactive MAbs , such as 4G2 and 6B6C-1 [53] . Thus , the EDII 101/106/107 region may be less likely to cooperate with the EDIII 329/331/389 region in inducing an antibody response when the immunogen been modified by formalin . The formalin-inactivated , lactate dehydrogenase-elevating , virus-elicited antibodies differ from antibodies after natural infection . Formalin-inactivated influenza virus could not induce a T-cell response and was less protective in mice against homologous and heterologous influenza virus challenge as compared with γ-ray–inactivated virus [37 , 82] . However , another study indicated that the use of low formalin concentrations , short inactivation period , and high incubation temperature improved the immunogenicity of formalin-inactivated JEV vaccine and elicited high titers of neutralizing antibodies in mice [42] . Here , we showed that the binding activity of T16 MAb was reduced more by virus inactivation at 22°C than 4°C for the same treatment duration . Surprisingly , UV-inactivated Nakayama virus failed to be recognized by MHIAF and T16 . Adjusting the condition for UV irradiation may maintain the antigenic structure of JEV . UV light inactivates virus by cross-linked viral nucleic acid and viral proteins . Cross-linked by oxidation between the amino acid residues may increase the susceptibility of protease cleavage [83 , 84] , and degradation of aromatic side chain of amino acid and disulfide bond forming cysteine in protein has been indicated after UV treatment [85 , 86] . The loss of viral antigenicity was also observed in UV-inactivated virus including poliovirus ( showing both antigenic and morphologic change ) , and influenza A virus ( exhibiting low hemagglutination activity ) [37 , 87] . Murray Valley encephalitis virus , belonging to JEV serocomplex , inactivated with UV showed lower immunogenicity compared to non-infectious VLP but the UV-induced antigenic change wasn’t described [88] . In conclusion , formalin and UV inactivation alter the antigenic structure of E protein in JEV and reduce the immunogenicity of associated vaccines . H2O2 inactivation seems to be a better alternative for JEV vaccine production . It maintained the antigenic structure of E protein , measured by a panel of MAbs . Further study should focus on identifying an optimal inactivation procedure and testing the immunogenicity of H2O2-inactivated JEV vaccine . Finally , to prevent unexpected modification of the various epitopes on the JEV vaccine during inactivation , a non-infectious JEV VLP or DNA vaccine should be developed . Formalin inactivation introduces an antigenic modification that affects the EDIII of JEV and thus distorts the profile of vaccine-induced neutralizing antibodies . Antigenic-stable inactivation methods are needed to develop better-inactivated JEV vaccines .
|
We demonstrated that formalin inactivation of Japanese encephalitis virus ( JEV ) alters the antigenic structure of the JEV envelope glycoprotein ( E ) , in particular an epitope in domain III , and that this reduces the ability of the inactivated vaccine to elicit protective neutralizing antibodies . Ours and others’ previous studies have highlighted the importance of improving the immunogenicity of genotype III ( GIII ) -derived JEV vaccine in order to provide cross-protection against genotype I ( GI ) viruses , which are emerging and replacing GIII viruses in many JEV-endemic regions . Encouraging the wide use of live-attenuated or chimeric vaccines , such as SA14-14-2 or yellow-fever 17D/JEV vaccines , respectively , developing GI virus-derived inactivated or premembrane/E–containing , noninfectious virus-like particle ( VLP ) vaccines are two other possible ways to address this potential problem . In this exploratory study , we highlight an alternative inactivation method , such as H2O2 treatment , which may improve the antigenic stability and immunogenicity of JEV .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Formalin Inactivation of Japanese Encephalitis Virus Vaccine Alters the Antigenicity and Immunogenicity of a Neutralization Epitope in Envelope Protein Domain III
|
The basidiomycete smut fungus Ustilago hordei was previously shown to comprise isolates that are avirulent on various barley host cultivars . Through genetic crosses we had revealed that a dominant avirulence locus UhAvr1 which triggers immunity in barley cultivar Hannchen harboring resistance gene Ruh1 , resided within an 80-kb region . DNA sequence analysis of this genetically delimited region uncovered the presence of 7 candidate secreted effector proteins . Sequence comparison of their coding sequences among virulent and avirulent parental and field isolates could not distinguish UhAvr1 candidates . Systematic deletion and complementation analyses revealed that UhAvr1 is UHOR_10022 which codes for a small effector protein of 171 amino acids with a predicted 19 amino acid signal peptide . Virulence in the parental isolate is caused by the insertion of a fragment of 5 . 5 kb with similarity to a common U . hordei transposable element ( TE ) , interrupting the promoter of UhAvr1 and thereby changing expression and hence recognition of UhAVR1p . This rearrangement is likely caused by activities of TEs and variation is seen among isolates . Using GFP-chimeric constructs we show that UhAvr1 is induced only in mated dikaryotic hyphae upon sensing and infecting barley coleoptile cells . When infecting Hannchen , UhAVR1p causes local callose deposition and the production of reactive oxygen species and necrosis indicative of the immune response . UhAvr1 does not contribute significantly to overall virulence . UhAvr1 is located in a cluster of ten effectors with several paralogs and over 50% of TEs . This cluster is syntenous with clusters in closely-related U . maydis and Sporisorium reilianum . In these corn-infecting species , these clusters harbor however more and further diversified homologous effector families but very few TEs . This increased variability may have resulted from past selection pressure by resistance genes since U . maydis is not known to trigger immunity in its corn host .
Pathogenic microbes secrete hundreds of compounds and proteins into their host as part of the infection strategy . This arsenal of virulence factors , often small proteins with a predicted signal peptide ( SP ) , effectors or candidate secreted effector proteins ( CSEPs ) , functions to facilitate entry , to subdue defense responses that may be triggered through their recognition by the hosts' surveillance system , to divert nutrients and to ensure proliferation [1]–[4] . Plants use a variety of defense mechanisms to avoid pathogen invasion and subsequent disease , including physical barriers , preformed antimicrobial compounds , but also activation of defenses . In particular , defenses can be induced by the recognition of highly conserved pathogen molecules ( Pathogen- Associated Molecular Patterns or PAMPs ) resulting in a broad-based PAMP-triggered immunity ( PTI ) . Certain pathogen effectors , whether secreted into the host apoplast or vessels and taken up , or delivered directly into cells to perform their function , are inadvertently recognized directly or through their action by a highly sophisticated system of which resistance ( R ) genes are a part , to elicit effector-triggered immunity or ETI [5]–[8] . Induced immunity includes cell wall strengthening , the generation of an environment toxic to the pathogen , encasement of the pathogen and localized programmed cell death ( PCD ) to arrest pathogen development [9] . Although the latter affects development of biotrophic pathogens , necrotrophic pathogens and hemibiotrophs at later stages of infection might have evolved to take advantage of triggering PCD [10] , [11] . Earlier studies in several pathosystems demonstrated the presence of genetically dominant avirulence ( Avr ) genes in pathogens , the products of which have been shown more recently to often be effectors , interacting genetically with also often dominant host R genes . This concept was developed by Harold Flor using the flax-rust Melampsora lini pathosystem [12] and simultaneously by his contemporary , Arend Oort who studied the wheat-loose smut Ustilago tritici pathosystem but because of WWII could only publish his results in 1944 in Dutch [13] . In pathogen populations , there is strong selection to avoid recognition resulting in rapidly evolving effectors and , in response , evolving host R genes or effector targets [14] , [15] . This natural arms race is accelerated in agricultural settings where invading pathogens necessitate the introduction of resistant host cultivars from breeding programs , thereby triggering boom-bust cycles . In many cases , Avr genes are present in genomic regions displaying high flexibility , such as telomeres [16] , heterochromatic locations [17] , [18] , or are surrounded by transposable elements ( TEs ) [17] , [19]–[22] which can facilitate effector gene mutation . Basidiomycete smut fungi are important pathogens that cause disease world-wide and are of economic importance on many Poaceae [23]–[25] . Ustilago maydis , the maize smut fungus , has become the paradigm for molecular genetic studies on biotrophic basidiomycete plant pathogens [26] , [27] . The barley covered smut fungus , U . hordei , is closely related but differs in important aspects: in U . hordei , race- and strain-specific virulence compatibility interactions exist whereas no dominant avirulence functions that genetically interact with dominant host resistance genes on a gene-for-gene basis have been identified in U . maydis . Moreover , U . hordei can infect only at the seed germination stage to develop quiescently in the meristematic region until sporulation occurs mainly in the seed heads [28] ( Figure S1 ) , a characteristic shared with many smut fungi , such as the maize-infecting Sporisorium reilianum [29] . In contrast , U . maydis can infect any above-ground part of the maize plant at any plant age , resulting in the proliferation and sporulation of the fungus in tumors it incites . In addition , U . hordei has differently organized mating-type loci affecting its biology [30] , [31] and it has a larger genome due to a much higher content of repeats and TEs [32] . Six Avr genes have been genetically identified in U . hordei which in different combinations constitute 14 different reported races; six corresponding resistance genes have been proposed in barley [33]–[36] . UhAvr1 determines avirulence towards barley cultivar Hannchen which has matching resistance gene Ruh1 which we recently mapped to the short arm of barley chromosome 7H [37] . The UhAvr1 locus was located to an approximately 80-kb region contained on Bacterial Artificial Chromosome ( BAC ) clone 3-A2 , using a marker-based approach in a mapping population of 54 progeny segregating for avirulence towards Hannchen , resulting from a cross between parental lines Uh362 ( MAT-2 Uhavr1 ) and Uh364 ( MAT-1 UhAvr1 ) [38] . We show here that this locus spans a cluster of predicted secreted protein genes on chromosome 18 and we identify through targeted deletions and complementation the gene with the UhAvr1 avirulence function , coding for a predicted secreted effector . The locus is syntenic to cluster 19A in both U . maydis and S . reilianum that also contains small proteins predicted to be secreted [26] , [39] , but has evolved differently . UhAvr1 is located in a transposon- and repeat-rich region and transposon activity seems responsible for breaking the avirulence towards Hannchen . In virulent isolates , it appears that insertion of transposable element sequences in the promoter of UhAvr1 has changed its expression .
BAC clone BAC3A-2 , genetically harbouring U . hordei avirulence gene UhAvr1 , was sequenced by GPS transposon insertion resulting in an assembled sequence of 117 kb . The presence of repeats and TE sequences made assembly challenging . On this BAC insert we identified 47 ORFs ( Figure 1A , Table S1 ) . Hybridization to DNA blots of separated chromosomes located this region to a 667-kb chromosome ( [38] , Figure S2 ) , designated as U . hordei Chr 18 , a homolog of U . maydis Chr19 in our recent comparative genome study [32] . In light of publications in which a number of avirulence gene products were CSEPs , we hypothesized that UhAVR1p could also be a secreted effector . On the sequenced BAC clone , ten predicted CSEPs were identified but only seven were likely candidates: gene 5 and gene 6 were located outside the genetic interval identified previously and gene 44 was very close to RFLP marker 2 which revealed three recombinants ( Figure 1A , Table S1 [38] ) . A change from avirulence to virulence would likely be caused by a mutation in the candidate gene such as a point mutation , leading to an amino acid change or a protein truncation , or a gene deletion or a change in transcription . To identify the UhAvr1 gene , we first checked the presence of the CSEP genes in the virulent parent Uh362 . A PCR-amplification product for all ten genes was obtained from genomic DNA indicating their presence in the genome of the virulent parent . DNA sequence analysis of the seven candidate UhAvr1 genes did reveal point mutations in three of the alleles in Uh362 ( Table 1 ) . Since CSEP 35 displayed an amino acid difference that could have changed its charge between the virulent and avirulent form , gene 35 was deleted in parental avirulent strain Uh364 . However , when crossed with virulent parent Uh362 , it did not result in virulence on cultivar Hannchen harboring Ruh1 . We therefore expanded the sequence comparisons to include a collection of field isolates from different parts of the world , four avirulent and six virulent on Hannchen ( Table S2 ) . Three of the six remaining likely candidate UhAvr1 genes were identical when comparing allelic sequences from virulent or avirulent isolates but in the other three , a few mutations were found ( Table 1 , Figure S3 ) . Unfortunately , none of the revealed mutations could be correlated with the Avr1 or avr1 phenotypes . This indicated that there were other changes outside of the sequences we investigated , that were responsible for the change in phenotype , or that the avirulence function did not reside in the selected effector candidates . Since no likely candidate for UhAvr1 was found , a systematic deletion analysis of the 80-kb region delimited by the markers ( Figure 1 ) was conducted using a marker-exchange method . In a first round , the region was divided into three sections , ranging from 15 to 38 kb in size , taking into account the location of the various predicted CSEP genes in the region ( Figure 1B , Figure S4 ) . No phenotypic differences or abnormal growth were observed for any of the haploid basidiospore deletion mutants and proper mating with compatible haploid basidiospores , such as virulent parental strain Uh362 necessary for pathogenicity tests , occurred . Mated strains were tested for pathogenicity by inoculating them on differential barley cultivars Hannchen ( Ruh1 ) and Odessa ( ruh1 ) . Deletion of fragment C18A2 from avirulent parental strain Uh364 yielded strain Uh1041 ( Uh364 Δ18A2 ) ( Table S2 ) and resulted in disease on Hannchen after mating with compatible virulent wild-type strain Uh362 ( Figure 2A ) , clearly indicating that the 38 . 5 kb fragment C18A2 contained avirulence gene UhAvr1 . When Uh1041 was crossed with avirulent strain Uh365 , a sibling to parental strain Uh364 but of opposite mating type , the resulting dikaryon caused disease on Odessa but not on Hannchen ( Figure 2A ) , indicating complementation with the avirulence function and showing that no other functions in the recognition of the dominant UhAvr1 allele had been inadvertently compromised in the mutant . Fragment 18A2 harboring two CSEPs as the likely UhAvr1 allele , was further divided to make five additional deletion mutants ( sub-sections C18A2-a to C18A2-e; Figure 1B ) . To generate the deletion constructs , primers were designed in such a way that the two CSEPs would be deleted in two different deletion constructs . Sixty-four PCR-positive deletion mutants were obtained for the five deletion constructs , which were further verified by DNA blot analysis ( Figure S5 ) . Nine deletion mutants were selected , two from each , except for C18A2-b for which only one expected deletion mutant was obtained . Among these , the two mutants for C18A2-c and the two for C18A2-d were virulent towards both barley cultivars Odessa and Hannchen in pathogenicity tests after mating with Uh362 ( Figure S5G ) . The overlapping fragments C18A2-c and C18A2-d shared only gene 17 encoding a CSEP ( Figure 1B ) that was a strong candidate for UhAVR1p . To confirm this , another deletion mutant was produced in which the 3′ 319 bp of the ORF of this gene in parental strain Uh364 was deleted ( Figure S6A and B ) . Two independent deletion mutants , Uh1289 and Uh1297 , had this small deletion which resulted in virulence towards both Hannchen and Odessa when crossed with Uh362 , producing disease in 40–50% of the plants ( Figure 2B ) , and confirmed that gene 17 ( UHOR_10022 , GenBank CCF49778 . 1 ) is necessary for UhAvr1 avirulence function . A 11 . 5-kb XbaI fragment cloned in a modified BAC vector ( pUSBAC5 , converted for use in Ustilago species by introducing a specific hygromycin B resistance cassette [38] ) yielded construct BAC1-6 which contained two predicted CSEPs: gene 16 and 17 ( Figure 1B ) . This clone partially overlaps with fragment C18A2 . C18A2 deletion mutant Uh1041 , virulent towards Hannchen , was complemented with BAC1-6 and two stable transformants ( Uh1205 and Uh1207; Table S2 ) were inoculated on barley cultivars Odessa and Hannchen after mating with compatible virulent strain Uh362 . No abnormal growth or defect in mating behavior had been observed in these haploid complemented strains . After mating with Uh362 , the complemented strains caused the same level of disease on Odessa as the wild-type cross and the deletion mutant , infecting from 30–40% of the plants . On Hannchen however , the level of disease was severely reduced compared to the deletion mutant and only ∼2 . 5% of the plants showed infected seed heads ( Figure S7 ) . Incomplete restoration of avirulence could have resulted from the integration of an incomplete fragment at random locations in the genome , affecting transcription . Similar results were obtained for Fusarium oxysporum f . sp lycopersici mutant strains complemented with the Six1 avirulence gene that did not restore complete avirulence towards tomato lines that contained the resistance gene I-3 [40] . This suggested that BAC1-6 contained the functional UhAvr1 gene . To exclude any possible effects from other genes contained on the BAC clone or on the deleted C18A2 fragment , single UhAvr1 deletion mutant strain Uh1289 was complemented with complete wild-type gene 17 sequences , including 659 bp and 630 bp from the upstream and downstream regions , respectively . Three independent transformants , strains Uh1372 , Uh1373 and Uh1374 ( Table S2 ) , completely prevented disease on Hannchen ( Table S3 ) , confirming that gene 17 is sufficient to restore avirulence and hence codes for UhAvr1 ( Figure 2C ) . UhAvr1 codes for a predicted full-length protein of 170 amino acids with a calculated Mw of 21 kDa . SignalP 4 . 1 predicts a 19 amino acids SP resulting in a processed mature protein of 18 . 9 kDa . Mature UhAVR1p has predicted coil , helix and extended beta structures but could not be modeled on any currently existing crystal structures ( Figure S8 ) and no clear similarities could be found to known proteins or other domains . From a BAC library constructed from genomic DNA from virulent strain Uh362 , a BAC clone , BAC1-E2 , was identified using gene 1 sequences as a probe ( Figure 1E ) . We had found that all predicted CSEP ORFs could be amplified by PCR from genomic DNA of strain Uh362 , but amplification of UhAvr1 , gene 21 and subsequent genes further to the right could not be achieved from BAC1-E2; a probe representing gene 18 did not hybridize to a DNA blot from this BAC clone suggesting its insert did not cover this region . Sequencing and assembly of this BAC clone proved challenging due to the presence of many repeats and TEs , as it had been for BAC3-A2 . Comparative analysis revealed indeed that one end of the BAC clone insert extended only 1187 bp past the stop codon of gene 16 . Synteny between the virulent and avirulent parents however , was apparent only up to 115 bp upstream of the start codon of UhAvr1 ( Figure 1C–E , Figure S3B ) . WUBLAST analysis found a 400-bp sequence after this break point , matching to two retrotransposon proteins ( UHOR_14086 and UHOR_14170 ) in the U . hordei genome . To reveal the sequence upstream of the UhAvr1 ORF in the genome of Uh362 , an inverse PCR was conducted with UhAvr1 ORF-specific primers on HindIII-digested and self-ligated genomic DNA . Sequence analysis confirmed the presence of intact UhAvr1 sequences including 115 bases upstream of its start codon . Further 5′ , the sequence diverged revealing no other Uh364-derived sequences and after 166 bases matched sequences with high similarity to the common long-terminal repeat sequence LTR5 from U . hordei Tuh3 , a copia-type retrotransposon also found in the mating-type region ( Figure 3A , [32] , [41] ) . A PCR product of 5 . 8 kb however , was amplified from gDNA from the virulent parent when using primer 1685 at the 3′-end of gene 16 and primer 1815 located at the 5′-end of UhAvr1 ( Figure 1D , Figure 3B , Table S4 ) . Several other primer combinations confirmed that an insertion of approximately 5 . 5 kb had occurred and that the genes to the right were preserved with respect to the organization in Uh364 ( Figure 3B and not shown ) . Consistently , hybridization of three probes representing gene 16 ( left of the breakpoint ) , UhAvr1 and gene 23 ( approximately 12 kb right of the breakpoint ) to separated chromosomes of parental strains Uh364 and Uh362 , clearly revealed that all genes were located to the same Chr18 ( Figure S2 ) . Combined with sequence information from this insertion , the data suggested that in strain Uh362 , TE activity had inserted a seemingly intact TE consisting of gag-pol sequences flanked by LTRs in the intergenic region between gene 16 and UhAvr1 . We speculated that this event separated the UhAvr1 ORF from promoter elements thereby likely changing its expression and hence recognition in Hannchen , making this isolate virulent on this cultivar . UhAvr1 is located in a region of the genome that is , with the mating-type region , among the richest in repeats and TEs , approaching 50% compared to an overall genome content of 8–10% [32] . This elevated presence/retention of TEs and repeats at this effector locus suggests that this region represents a more dynamic part of the genome , enabling evolutionary changes as proposed for other pathosystems [17] , [19]–[22] . If this effector region is under selection pressure , e . g . , to modify the expression of UhAvr1 to avoid triggering immune responses , it is conceivable that TE activity and insertions might have played a role . Since TE activity rates cannot easily be investigated , we assessed possible variation at the UhAvr1 locus in a number of field isolates . In Uh364 , avirulent on Hannchen , 749 bp separate gene 16 and UhAvr1 from each other in the genome and amplification with primers 1685 and 1815 or 1511 ( Figure 1C , Table S4 ) yields PCR products of 898 and 2462 bp respectively ( Figure 3B ) . In five available avirulent field isolates , larger products were obtained for both primer combinations ( results for three isolates are shown in Figure 3B ) and upon sequencing revealed identical 340-bp insertions in the intergenic region in isolates Uh813 , Uh1273 and Uh1283 ( Figure 3C ) . This insertion also matched TE sequences in the U . hordei genome and was flanked by 6-bp repeats ( TGGGTT ) , possibly a footprint of TE activity . This particular insertion , though apparently not affecting avirulence in these isolates , was not found in the virulent parent Uh362 . The region was also analyzed from eight U . hordei field isolates virulent on Hannchen ( Uh805 , Uh811 , Uh815 , Uh818 , Uh820 , Uh822 , Uh1278 and Uh2001-246; Table S2 ) . Primer combination 1685 and 1815 or 1511produced PCR products of approximately 5 . 8 and 7 . 2 kb respectively , similar to the products obtained from Uh362 ( Figure 3C ) . However , upon sequencing , variation was revealed among the TE sequences in the different virulent strains . One predominant mutation found in four virulent strains ( Uh362 , Uh805 , Uh815 , and Uh820 ) was a 10-bp insertion of a repeat ( GAGAGAGAGC ) that was however absent from three other virulent strains ( Uh811 , Uh818 , and Uh822; Figure S3C ) . The 340-bp insertion discovered in three of the avirulent field isolates was not found in these eight virulent field isolates . Overall , the variation found in sequences surrounding UhAvr1 in field isolates both avirulent and virulent on Hannchen , and the similarity of those sequences to various U . hordei-specific repeats and TEs , suggest various transposition events have occurred in different isolates resulting in a variety of combinations upon which selection could act . Previous electronmicroscopy work revealed necrosis in cells immediately surrounding penetration sites early upon infection during an incompatible interaction on Hannchen [42] . We performed a microscopic analysis of the natural infection process by teliospores , previously produced on universal susceptible cultivar Odessa . Infection of coleoptiles of cultivar Hannchen by teliospores from crossed wild-type progenitor strains Uh364 ( MAT-1 UhAvr1 ) ×Uh362 ( MAT-2 Uhavr1 ) caused extensive production of reactive oxygen species as visualized by DAB staining suggesting cell death could have been initiated ( Figure 4A ) , and extensive callose deposition seemingly restricting pathogen development ( Figure 4C and E ) . In stark contrast , teliospores produced from cross Uh1289 ( Uh364 MAT-1 , ΔUhAvr1 ) ×Uh362 ( MAT-2 Uhavr1 ) caused a natural infection of coleoptile epidermal cells of cultivar Hannchen , showing hyphal development , very little oxidative damage ( Figure 4B ) and limited , diffuse callose depositions ( Figure 4D and E ) , illustrating a compatible interaction . Expression analysis of UhAvr1 by quantitative RT-PCR during infection proved challenging . No expression could be detected in haploid avirulent or virulent cells grown in liquid media , or during mating interactions on plates ( data not shown ) . Weak and variable expression was observed in mated cells and teliospores applied to barley coleoptiles but always only when avirulent strain Uh364 was employed ( Figure 5A ) ; linear pre-amplification of cDNA to increase signal strength [43] corroborated these results but introduced variation ( Figure 5B ) . This suggested that the expression of UhAvr1 might be induced only upon direct contact with , or actual infection of coleoptile epidermal cells . The low amount of transcript is likely due to the very few contact and penetration sites present resulting in a very small proportion of responding cells in the biological material ( the inoculated coleoptiles ) from which the RNA was isolated; this resulted in variable qRT-PCR results . From the combined data it was evident that in the virulent parental strain Uh362 ( or its virulent sibling Uh359 ) the level of UhAvr1 mRNA bordered on the limit of detection , but was at most only 10% of the level seen in Uh364 after pre-amplification ( Figure 5B ) . Therefore , to substantiate the tentative expression results and to possibly localize UhAVR1p , a chimeric gene construct was made of UhAvr1 with its native promoter but linked to a green fluorescent protein ( GFP ) moiety at its C-terminal end . This was then used to replace by marker-exchange the ΔUhAvr1 deletion in strain Uh1289 , thereby putting the chimer in its original expression site ( Figure S6C ) . Confocal microscopy of this constructed strain Uh1353 ( Table S2 ) clearly corroborated the qRT-PCR expression results since no fluorescence was observed in haploid or mated cells at the time of inoculation of barley coleoptiles ( Figure 6A ) , whereas bright fluorescence comparable to GFP expressed from the strong U . maydis otef promoter [44] ( Figure 6B ) was apparent after 48 hrs in mated dikaryotic hyphae upon infection ( Figure 6C ) and while extending in the coleoptile later during the infection ( Figure 6D ) . GFP fluorescence was seen in growing hyphal tips , possibly in vesicle-like structures ( Figure 6D and E ) and in older hyphae associated with the cell wall . On cv . Hannchen , a very similar infection by fluorescent hyphae was observed but no obvious HR reaction , such as increased autofluorescence and/or cell collapse , was seen at 72 hrs ( Figure 6F ) . This suggested that the chimeric protein seemed unable to trigger the R gene-based immunity . Indeed , pathogenicity tests with these complemented strains ( Uh1353 , Uh1354 , Uh1355 , Table S2 ) when mated with Uh362 , were causing similar levels of disease on both Odessa and Hannchen ( Table S3 ) . In contrast , as reported above , strains complemented with wild-type UhAvr1 gene sequence including promoter and terminator elements did not cause any disease on Hannchen ( Uh1372 , Uh1373 , Uh1374; Figure 2C , Table S3 ) . These experiments suggested that the C-terminal GFP-moiety interfered with the process that led to resistance triggering . We have not been able to verify in these strains whether or not the intact chimer is produced when infecting and if so whether possibly proper translocation and targeting to the proper location is affected by the presence of a C-terminal moiety . Whether C-terminal extensions interfere with protein structure and/or obstruct proper recognition of the host target ( s ) , R gene or R-gene complex , needs further study but has been shown to occur in the flax rust fungus [45] . When attempting to complement virulent deletion strain Uh1289 ( ΔUhAvr1 ) with the UhAvr1 ORF under the control of the strong constitutive Ustilago Hsp70 promoter , the resulting transformants , when mated with compatible parental strain Uh362 , did not trigger resistance in cultivar Hannchen and yielded levels of disease similar as on Odessa or from control crosses ( Table S3 , crosses 19–21 , compare cross 6 ) . Similar constructs with the Hsp70 or otef promoters driving the UhAvr1 ORF now linked at its C-terminal end to either the HA epitope tag or a GFP moiety , yielded transformants that similarly gave comparable levels of disease on both Odessa and Hannchen ( Table S3 , crosses 4 , 5 , 13–18 ) . Protein blot analysis confirmed the production of the expected chimeric proteins in the transformants ( Figure S9 ) and we assumed from these assays that the wild-type UhAVR1p effector is similarly expressed from the Hsp70 promoter in the transformants mentioned above . In many pathogens studied , cloned avirulence effectors have been shown to assert their avirulence function when reintroduced and expressed from non-native , strong promoters . In U . hordei , the expression of UhAvr1 is finely tuned ( Figure 6 ) and it is possible that this regulation is essential for proper relocation and function , including R-triggered immunity . In several pathosystems , the deletion of avirulence effector genes was shown to affect virulence on host cultivars not harboring the cognate R gene . We tested in the ΔC18A2 deletion mutant whether or not genes 6 to 22 , which included UhAvr1 and two other CSEP genes , have any virulence functions in U . hordei . To this end , an equivalent C18A2 deletion was generated in a MAT-2 mating partner by crossing Uh1041 ( MAT-1 ΔC18A2 ) with virulent parent Uh362 ( MAT-2 Uhavr1 ) on barley cultivar Hannchen . Carboxin-resistant basidiospores of mating type MAT-2 were collected by germinating teliospores from infected heads and lack of fragment C18A2 was verified by DNA blot analysis ( Figure S10A ) . Each of three individual C18A2 deletion mutant progeny ( Uh1116 , Uh1117 , Uh1118 ) was back-crossed with Uh1041 , resulting in virulence towards Odessa that was similar to the wild-type cross ( Figure S10B ) . One cross tested on Hannchen seemed also not affected in virulence compared to the single deletion mutant ( Figure 2A ) . We concluded that genes 6 to 22 do not contribute significantly to virulence on barley . ΔUhAvr1 mutants crossed with Uh362 ( Uhavr1 ) are always included in our pathogenicity tests and over many experiments , virulence , expressed as number of plants infected per total number of plants inoculated , has not differed significantly from wild-type crosses . This suggests that effector UhAVR1p is not contributing significantly to virulence . It is difficult to express virulence in a quantitative manner in this pathosystem and a subtle advantage of expressing UhAvr1 may play out at the population level over time . The sequence analysis of clone BAC3-A2 revealed that the UhAvr1 locus is orthologous to a region on U . maydis chromosome Chr 19 , spanning a cluster of 24 CSEPs , called cluster 19A , the largest of such clusters in the U . maydis genome [26] ( Table S5 ) . A similar cluster is found in S . reilianum , harboring 29 CSEPs [39] . In U . maydis , deletion of this cluster resulted in reduced disease on maize seedlings . SIMAP analysis [46] and two-directional BLASTp searches were used to find orthologs for the U . hordei predicted CSEPs at this region in the U . maydis genome ( Table S1 ) . There is synteny with conserved gene order between these U . hordei and U . maydis genomic regions flanking the predicted CSEPs ( Figure 7A ) . However , the region containing the CSEPs is much diverged and rearrangements , including changes of gene orientation and several translocations of genes within the cluster , are apparent . For example , DigA ( Uh gene 1 and Um gene 4 ) is conserved but a homolog of the adjacent oligosaccharyltransferase gene ( Um gene 5 ) is found 52 kb away in an inverted orientation in U . hordei ( Uh gene 23 ) . On the other end , conserved homologs of U . maydis genes 35 , 36 and 37 are found in a syntenous region in U . hordei ( genes 42 , 43 and 46 , respectively ) , except that a CSEP gene ( Uh gene 44 ) with homology to two Um CSEPs that are however located on a different Um Chr 10 , and repeat sequences have inserted . Overall , the U . hordei cluster region in between the syntenic blocks bordered by Uh genes 1 and 40 , is 40 . 6 kb larger than in U . maydis , in part the result of the presence of TE and repetitive DNA sequences . Other important differences are in the complement of the predicted CSEP genes . In U . maydis , four families of CSEP genes that are arranged in tandem in clusters of several paralogs , were described [26] ( Figure 7A ) . One U . maydis family ( genes um05299 , um05300 and um05301 , genes 11 , 12 and 13 ) is not represented in the U . hordei region or its genome [32] and seems species-specific . A molecular phylogenetic tree was generated and to reveal possible derived family members , we included several family members from the S . reilianum 19A CSEP cluster [39] ( Figure 7B ) . UhAvr1 ( UHOR_10022 , gene 17 ) and its adjacent paralog UHOR_10021 ( gene 16 ) are homologous to U . maydis CSEPs um05294 and um05295 ( genes 6 and 7 ) residing in the tin1-1 to tin1-5 cluster , an expanded family of five adjacent , weakly related paralogous effectors [47]; S . reilianum has 3 homologs , sr10050 , sr10051 and sr10052 . 2 ( Figure 7A and B ) . Such an expansion in U . maydis is also seen for UHOR_10033 ( gene 39 ) and UHOR_13916 ( 38 ) with three related paralogs in U . maydis ( um05317 , um05318 and um05319 , genes 29 , 30 and 31 ) and four in S . reilianum ( sr10073 , sr10075 , sr10077 and sr10079 ) , and for UHOR_08134 ( gene 35 ) with various homologs in both U . maydis and S . reilianum . Overall , in U . hordei , the related families are more dispersed and separated from adjacent genes , sometimes in inverted orientation , by TE and repeat sequences . Virtually no such repeat sequences are present in the U . maydis cluster although U . maydis gene um05316 ( gene 28 ) codes for a transposase indicating possible ( past ) TE activity ( Figure 7A ) .
Previously we showed that the UhAvr1 locus was located to an approximately 80-kb region contained on BAC clone 3-A2 [38] . In this study , we sequenced its insert to discover , among others , ten ORFs encoding small predicted secreted proteins at this genetic locus . RAPD and AFLP markers limited these to seven most-likely candidate avirulence-triggering effector genes . Sequence comparison of these ORFs from the virulent and avirulent parents used to generate the mapping population , as well as from ten additional virulent and avirulent field isolates from a world-wide collection , revealed that the change from avirulence ( UhAvr1 ) to virulence ( Uhavr1 ) is not due to mutations in the ORFs or the presence or absence of ORFs in the two parental strains . We subsequently identified UHOR_10022 through targeted deletions and complementation-based approaches as being U . hordei avirulence gene UhAvr1 . Quantitative RT-PCR analysis to verify expression of UhAvr1 during infection proved challenging because of the very low levels of fungal biomass at this stage relative to the tissue mass in the coleoptile . However , when microscopically investigating single cell events , the substantial fluorescence emanating from the UhAvr1:GFP fusion transcribed from its native promoter in its original genome location , only shortly after contact with barley coleoptiles ( Figure 6 ) , showed that this gene is induced during infection . In RNA samples isolated from immature and mature infected seed heads , no UhAvr1 mRNA could be detected by quantitative RT-PCR whereas high levels of expression were detected for U . hordei actin ( UHOR_08813 ) and eIF-2B ( UHOR_07772 ) genes that were used as references in the analysis ( data not shown ) . This shows that UhAvr1 is expressed only during early infection , being highly regulated and suggests that UhAVR1p is needed only early during infection . In many plant pathogenic fungi and oomycetes , a subset of predicted effectors , some of which trigger avirulence , are expressed only upon infection and sometimes only in specific infection structures such as appressoria or haustoria [26] , [45] , [48]–[51] . Expression of UhAvr1 provides complete immunity in barley cultivars harboring Ruh1 and we show that UhAVR1p harbors an avirulence function which is somehow recognized by RUH1 . Previously we had shown by electron microscopy that this interaction caused hyphal restriction , likely due to the deposition of electron-dense material , and necrosis in cells immediately surrounding penetration sites early upon infection [42] . This correlates well with the timing of expression of UhAvr1 and the accumulation of callose and associated fluorescence just around penetration sites and around restricted hyphae within 72 hours of infection ( Figure 4C ) . Sequence analysis among the limited collection of field isolates virulent and avirulent on Hannchen , suggest that UhAvr1 may encode a rather monomorphic protein; only two point mutations were identified in UhAvr1 . In only one avirulent strain Uh813 that translated into a single amino acid substitution . Whether this points to indirect recognition of this avirulence effector by RUH1 , more in line with the ‘guard model’ stipulating purifying selection as the guard recognizes modifications of the AVR protein on the guardee and imposes selection pressure against its function [52] , [53] and which favours gene inactivation or deletion [54] , whereas direct interaction according to the ‘receptor-ligand model’ tends to result in diversifying selection that generates highly divergent avirulence effector alleles in pathogen populations to escape this recognition by the R gene products [55]–[58] , remains to be investigated . Future experiments are geared towards finding the target and mode of interaction of UhAVR1p . In UhAVR1p , no clear similarities could be found to known proteins or domains . Interestingly , a RxLR tetrad is found in the paralogous U . maydis effectors um05295 ( amino acid positions 99–102 ) and um10554 ( 125–128 ) , and sr10052 ( 89–92 ) from Sporisorium reilianum . When compared , the RxLR motifs line up with a PDFR tetrad in UhAVR1 ( Figure S8 ) . The RxLR motif has been proposed to be involved in binding of specific plant and mammalian cell wall phospholipids ( phosphatidylinositol 3-phosphate or PI3P ) , mediating effector uptake . However , among various fungal and oomycete effectors , this motif has been shown to allow for some variation [59] and its function in uptake has been controversial [60] , [61] . Alternatively , PI3Ps are enriched in intracellular organelle membranes , specifically from early endosomes [62] , [63] and we are investigating possible targeting of UhAVR1 to such locations . Intriguingly , if 20 amino acids were cleaved off , UhAVR1 is predicted to be myristoylated , suggesting a membrane association is involved . Several effectors have been shown to be myristoylated and that this was required for function [64] , [65] . Moreover , amino acid K39 has a high probability of being a sumoylation site ( Figure S8 ) . Sumoylation is a reversible post-translational modification that affects an increasing number of biological processes by altering intracellular localization and protein-protein interactions . We were not able to ascertain the virulence function ( s ) of UhAvr1 in this study . Examples exist of avirulence effectors with a clear role in virulence , such as AVR-a10 and AVR-k1 from Blumeria graminis that enhance fungal penetration in barley epidermal cells [66] . Similarly , AVR3a from P . infestans can suppress necrotic responses in Nicotiana benthamiana induced by INF1 elicitor [67] . Experiments expressing UhAvr1 in Nicotiana leaves did not support a role in the suppression of cell death initiated by several elicitors . This could be due to unavailable or too diverged targets of UhAVR1p in this heterologous system . A homologous system in barley would be needed , possibly in young coleoptiles if timing of expression is essential . In this context it is important to note that infection of barley by U . hordei only occurs early at seed germination , since the fungus needs to reach meristematic tissue; older plants or leaves cannot be successfully inoculated . In the related study by Brefort et al . [47] , a U . maydis strain lacking the Tin1-1 to Tin1-5 effectors ( genes 6–10 in Figure 7 , with UhAVR1p closest related to Tin1-2 ) caused strong induction of endochitinases , SA-binding proteins and the apoplastic peroxidase POX12 in maize , indicative of enhanced defense responses and a possible role for these effectors in suppressing basal host immunity . Whatever function UhAVR1p has , it does not seem to contribute significantly to virulence as shown in Figure 2B ( and Figure S10B , where the paralogous gene 16 is also deleted ) . In U . maydis , deletion of the paralogous tin1-1 to tin1-5 effector family did not cause a statistically significant reduction in virulence [47] . However , it is very difficult to assess relative infection rates in this pathosystem for which no good quantitative measures exist and which relies on the number of infected plants out of a significant number of inoculated plants showing often considerable variation . Functional redundancy may exist in effectors located at other sites in the genome such as effector gene sr13459 , a potential homolog of UHOR_08134 ( gene 35 ) , which is located on a different Chr 20 in S . reilianum . Although not easily measurable as reduced virulence in a few plant experiments , on a population level the UhAVR1 effector may contribute to overall fitness or virulence . It has also been argued that effectors ( alleles ) that contribute to virulence or fitness are maintained in a pathogen population [e . g . 68–70] . A TE insertion inactivated UhAvr1 but in the isolates we investigated , the genetic information of the ORF was still present . It is therefore possible that virulent isolates that have retained the ( inactive ) UhAvr1 ORF sequences may have a selective advantage because subsequent re-activation of the UhAvr1 ORF , i . e . , by hooking it up again behind a promoter through transposition or gene conversion , will again bring about this population-level advantage if the selection pressure , i . e . , plants with Ruh1 , disappeared from the environment the fungal population occupies . Our analysis of the U . hordei genome revealed many TEs and repeats with Repeat-Induced Point mutations , likely inactivating them [32] . However , complete TE ( LTR-like ) sequences with intact conserved predicted ( gag/pol ) proteins were also found indicating that these elements could be active transposons . In addition , comparison of the genomes of the three smuts , U . hordei , U . maydis and Sporisorium reilianum , suggested that a recent expansion had occurred of a few related TEs newly introduced in the U . hordei lineage after separation from a common ancestor , also indicating active elements ( at least in its recent evolution ) . In our study , sequence comparisons between UhAvr1 loci from isolates avirulent and virulent on Hannchen revealed TE sequence variants upstream of UhAvr1 and that virulence towards Ruh1 was the result of TE activity and insertion of TE-derived sequences in the promoter region of UhAvr1 changing expression and likely recognition . TE activity and insertion at avirulence effector loci causing in situ mutations or changes in transcription leading to virulence phenotypes , have been described before in ascomycete pathogens [e . g . 19 , 21 , 71–73] but not in basidiomycetes . At the locus , sequence variation involving TE sequences among various field isolates indicated transposition events , possibly of independent nature suggesting TE activity is an important mechanism to overcome resistance to Ruh1 . In some field isolates , sequence variation was identical such as in virulent strains Uh362 , Uh805 , Uh815 and Uh820 ( Figure S3C ) . Considering the geographic area the latter three were collected from ( Kenya , Canary Island and Tunisia , respectively ) , this likely reflects a common ancestral event and regional spread; Uh362 was derived from a Canadian isolate backcrossed with an African isolate long ago to obtain homozygous material and likely acquired the virulent allele from this region . Similarly , the identical 340-bp insertion found in isolates Uh813 , Uh1273 and Uh1283 ( Figure 3C ) , respectively from Iran , Azerbaijan and Turkey , could have a regional ancestral origin . This illustrates the difficulty of sampling pathogens from a crop plant that is widely traded and grown in certain areas . In order to assess more-comprehensive variation , one would need to sample isolates from truly wild barley populations in remote locations . TEs play important roles in shaping genomes , causing rearrangements such as deletions , inversions , duplications , translocations , but also neo-functionalizations . Recently , genome analyses of several fungal and oomycete pathogens revealed that many effector genes reside in TE and repeat-rich regions ( including at telomeres ) , a feature that may have evolved to allow for variations necessary for parasites under high host selection pressures to quickly adapt when their virulence effectors are triggering defenses [e . g . ] , [18] , [22] , [68] , [74]–[77] . The UhAvr1 gene is located in a region of the genome that sports ten CSEP genes and is , with the mating-type region , among the richest in repeats and TEs , approaching 50% ( Figure 7A [32] , [41] ) . Incidentally , the UhAvr1 locus revealed conserved synteny in regions flanking cluster 19A , the largest cluster of CSEP genes in U . maydis , and to some extend among its coded effectors ( Figure 7 ) . Transcription of the U . maydis CSEP genes is induced after infection of maize and deletion of this whole cluster severely reduces disease [26] . It appears that these species , including related S . reilianum , share some of these likely ancestral genes but that possibly because of their obligate biotrophic interaction with diverse hosts , these effectors have evolved differently . Phylogeny revealed expanded CSEP gene families in U . maydis and S . reilianum . Interestingly , in the U . maydis-maize pathosystem , no effector-R gene interactions involving avirulence and resistance genes have been genetically identified to date [78]–[80] . It is possible that the higher number of paralogs in the U . maydis ( and S . reilianum ) effector gene families represent past diversifying selection acting on these effectors to avoid host recognition and making U . maydis better adapted to host populations . This could have resulted from adaptation to changed effector target molecules or the defeat of major resistance genes over time . While in U . hordei the mechanism to avoid host recognition involves the activity of TEs , U . maydis and S . reilianum have more streamlined genomes with few deleterious repeats and TEs [26] , [32] , [39] . The question arises how the latter organisms have created the needed variation . One scenario could be past TE activity , followed by purging of TEs and repeats brought about by a highly active homologous recombination system known to exist in U . maydis . The numerous small ( 10 bp ) repeats in the U . maydis genome have been suggested to be footprints of past TE activities [32] , [39] and 26 are found exactly in between the effector genes in the U . maydis cluster 19A ( Figure 7A ) . Alternatively , if TE activity did not play role in these organisms , highly active recombination followed by genetic drift may have caused sufficient variability . However , the evolution of these pathogens is more complex and involves sex [81]; U . hordei with its bipolar mating system which promotes inbreeding , may select for the use of TEs as genome modifiers whereas U . maydis and S . reilianum with their tetrapolar mating systems which cause reduced inbreeding potential , can create variation through recombining with outside partners [77] . Undoubtedly , the selection pressure imposed by the host has had a major impact on maintaining the variability among populations , as has been shown for the U . maydis-maize interaction [82] .
Two barley cultivars , ‘Odessa’ ( ruh1 , universal susceptible ) and a differential , ‘Hannchen’ ( Ruh1 ) were used for pathogenicity assays . Fungal strains and mutants generated are listed in Table S2 . U . hordei haploid parental strains Uh364 ( alias Uh4857-4 , MAT-1 UhAvr1 ) and Uh362 ( alias Uh4854-10 , MAT-2 Uhavr1 ) were described previously [38] . Haploid U . hordei strains were grown in liquid Potato Dextrose Broth ( PDB ) , complete medium ( CM [83] ) or YEPS ( 1% yeast extract , 2% peptone , 2% sucrose ) , while 2 . 5 µg/ml carboxin ( Sigma-Aldrich ) , 100 µg/ml Hygromycin B ( Calbiochem , La Jolla , CA , USA ) or 40 µg/ml Zeocin ( Invitrogen , Valencia , CA , USA ) were added when appropriate . Strains were grown at 22°C . For genetic transformation of U . hordei , protoplasts were prepared according to a modified protocol [84] , instead using 384 mg/ml Vinoflow FCE ( Gusmer Enterprises ) as enzyme mix for digesting the fungal cell wall [85] . Protoplasts were transformed with 5 µg DNA mixed with 1 µl of a 15 mg/ml heparin ( Sigma ) in STC ( 10 mM Tris-HCl pH 7 . 5 , 100 mM CaCl2 , 1M sorbitol ) solution and selected on double-complete medium plate ( DCM ) supplemented with 1 M sorbitol and appropriate antibiotic . After 5–7 days incubation at 22°C , colonies from DCM-S were transferred to CM medium and incubated for two days at 22°C before transferring to liquid CM medium for further analysis . Two haploid cultures of opposite mating type ( OD600 of ∼1 , tested in mating assays as described in [86] ) were mixed 1∶1 v/v before inoculation of barley seeds . Seeds were dehulled , surface sterilized for 3 min with 70% EtOH , followed by 10 min with 1% bleach , and rinsed several times with sterile ddH2O . Surface-sterilized seeds were dipped in mated cultures and a vacuum of 20 psi was applied for 20 min . Subsequently , excess inoculum was drained and seeds were kept for 6 hrs at room temperature before sowing in potting mix ( Pro-Mix BX ) at a density of 3 seeds per 3×3″ pot of which 18 were placed in a tray . Plants were grown in controlled-environment chambers with an 18 hour light-6 hour dark cycle at 22°C . Disease ratings were scored at heading , approximately 2 months after planting , by counting infected plants among all inoculated plants . The same inoculum was always applied to both barley cultivars Hannchen ( Ruh1 ) and Odessa ( ruh1 ) simultaneously to verify effectiveness . BAC3-A2 containing the UhAvr1 locus from the avirulent parent Uh364 was sequenced using the GPS-Mutagenesis System ( New England Biolabs ) with a few modifications . In the donor vector , the kanamycin resistance cassette within the transprimer was replaced with a phleomycin resistance cassette driven by both the Em7 bacterial promoter and the U . maydis glyceraldehydes-3-phosphate dehydrogenase ( GAPDH ) promoter and terminator [87] . This generated an insertion that could be used directly as a marker-exchange construct to generate deletions within U . hordei through homologous recombination . After in vitro recombination and transformation in E . coli , BAC clones from 6×96 random bacterial colonies were sequenced using primers N and S , yielding paired sequence reads from the ends outwards of the randomly inserted transprimers . These DNA sequences and several BAC end-sequences covering this region from clones of the source BAC genomic library [41] were entered in the PCAP . REP genome assembly program [88] . To place certain sequences and to verify their location , physical mapping was performed by using the unique Not1 restriction enzyme sites in the transprimer and BAC insert and measuring generated fragment sizes on CHEF gels ( data not shown ) . BAC clone 1-E2 covering the Uhavr1 locus in the virulent parental strain Uh362 , was recovered from a BAC library via hybridization . This BAC clone , as well as BAC3-A2 for confirmation , were sequenced using the 454 technology at the Plant Biotechnology Institute ( Saskatoon , SK ) . The resulting reads were assembled using the Newbler program ( Roche Applied Science ) . Alignment of the BAC sequences from the virulent parent along the avirulent backbone was facilitated by a custom Perl script . The order of contigs was confirmed by PCR and gaps were corrected through manual sequencing . Genes were predicted using FGENESH [89] and VectorNTI ( Invitrogen ) . Predicted proteins were searched for secretion signals using the SignalP 3 . 0 Server ( http://www . cbs . dtu . dk/services/SignalP/ ) , by TargetP v1 . 1 [90] to identify and remove proteins that were predicted to be mitochondrial , and by ProtComP 9 . 0 ( http://linux1 . softberry . com/berry . phtml ) which compares them to proteins in the LocDB and PotLocDB databases which hold proteins with known or reliably predicted localization . The sequence of clone BAC3-A2 was contributed to the Uh364 genome sequencing effort ( [32] http://www . helmholtz-muenchen . de/en/ibis/institute/groups/fungal-microbial-genomics/resources/muhdb/index . html ) and is part of UHOR_scaffold_5 . 00017 , NCBI #CAGI01000148 . 1 with UHOR_10022 , protein ID CCF49778 . 1 , at position 159450–160022; the sequence of the region containing the breakpoint in virulent parent Uh362 on BAC clone1-E2 is accessible under NCBI #KF640593 . To sequence ORFs in Uh362 and field isolates , predicted CSEP genes and intergenic regions were amplified by PCR . Primers were designed 100 bp upstream and 100 bp downstream of the ORFs ( Table S4 ) using the Primer3 software ( http://sourceforge . net/projects/primer3/files/ ) . Sequencing of the purified products was carried out using the Big Dye terminator v3 chemistry ( Applied Biosystems ) . Large PCR products were generated using LongAmp DNA Polymerase ( New England Biolabs , M0323S ) . One gene , UHOR_08134 , was deleted using a double-jointed PCR method [91] and the hygromycin resistance cassette to generate a marker-exchange construct . All other deletion mutants involving individual target genes or clusters of genes were constructed using marker-exchange plasmids generated by the DelsGate method [92] . Briefly , primers were designed separately for each construct to amplify by PCR 1 . 5 to 2 kb of 5′- and 3′- sequences flanking the target region ( Table S4 ) , using Uh364 genomic DNA as template . Primers 5L and 5R were then used for the amplification of a 5′-flanking fragment adding an I-SceI recognition sequence tail upstream and an attB1 sequence tail downstream of the flank sequence . Primers 3L and 3R were used to amplify the 3′-flanking fragment , adding the attB2 sequence tail upstream and the I-SceI sequence tail downstream . The two PCR-amplified fragments were then gel-purified using the QIAquick Gel extraction kit ( Qiagen ) and subsequently recombined into the pDnorCbx vector ( NCBI accession # EU360889 [92] ) using the Gateway BP Clonase II enzyme Mix ( Invitrogen ) . To assess the resulting marker-exchange plasmids , two PCR reactions were performed using 5′- gene-specific primer 5R in combination with the SceIF primer , and 3′- gene-specific primer 3L in combination with primer SceIR primer ( Table S4 ) . SceIF and SceIR primers were designed for the I-SceI enzyme recognition site in the forward and reverse orientation , respectively . The deletion constructs were verified by sequencing and were then linearized with I-SceI enzyme ( New England , Biolabs ) and used directly for U . hordei transformation . Carboxin-resistant mutants were analyzed for proper gene deletion by PCR reactions on purified gDNA . Sixty to as many as 300 carboxin-resistant colonies sometimes needed to be screened ( depending on the region targeted ) to get at least four PCR positive transformants for each construct which were then verified by DNA blot analysis . For DNA blot hybridization , 10 µg of gDNA was digested with selected restriction enzymes and run out in 0 . 8% agarose gels in 1xTAE buffer ( 40 mM Tris-acetate , 1 mM EDTA ) . Blotting to nylon membranes ( Amersham Biosciences , Buckinghamshire , UK ) and hybridization were carried out following standard procedures [93] . DNA probes for either the 5′- or 3′-flanks were amplified using PCR and labeled with [α-32P] dCTP using the random primer labeling system kit ( Amersham Biosciences ) according to manufacturer's recommendations . The efficiency of homologous recombination was different for different constructs and seemed dependent on the size of the deletion fragment; the efficiency was higher for small fragments . Gene expressing constructs were designed to make use of the GateWay technology ( Invitrogen ) . U . hordei ORFs , either with or without the sequence coding for the SP , but without their stop codon , were amplified by PCR with a CACC tetranucleotide sequence at the 5′-end to allow for directional cloning into Gateway entry vector pENTR/D-TOPO ( Invitrogen; Table S4 ) . Cloned inserts were sequenced and were subsequently transferred to a designed GateWay destination vector , pUBleX1Int:GateWay:HA ( a derivative of Ustilago-specific integrative expression vector pUBleX1Int [94] ) , using LR recombineering . For the transient assays and microscopy after bombardment , the above-mentioned pENTR clones ( UHOR_10022-SP-STOP ) were recombined into a modified pMCG161 vector ( ChromDB at http://www . chromdb . org; NCBI accession no . AY572837 ) to create N- or C-terminal GFP-expressing chimers from the maize ubiquitin promoter ( Ubi:GFP:UhAvr1-SP and Ubi:UhAvr1-SP:GFP ) . A control construct expressed just GFP . Details on the constructs and destination vectors can be obtained from the authors . Barley cv . Hannchen coleoptiles were inoculated with mated cell cultures as described above and infection was allowed to proceed for 48 hrs on sterile filter paper in petri dishes in the dark at 22°C . Coleoptiles from 3 biological replicates were dissected from the seed and roots and total RNA from 100 mg of sample was isolated using Trizol Reagent ( Invitrogen , Cat . No . 15596-018 ) . Ten µg of total RNA was then treated with TURBO DNase ( Applied Biosystems , Cat . No . AM22380 . After quantitation , cDNA synthesis was carried using SuperScript III Reverse Transcriptase ( Invitrogen , Cat . No . 18080-093 ) . Quantitative RT-PCR assays were carried out on a CFX96 Real-Time System ( Bio-Rad ) with the following cycling conditions: ( 1 ) 2 min 95°C incubation , ( 2 ) cycling at 95°C 10 sec , 55°C 30 sec for 40 cycles , ( 3 ) melt curve from 65°C to 95°C at 0 . 5 degree increments . Analyses and statistics were carried out with the Bio-Rad CFX Manage Software . To overcome the very low expression levels observed , nested real time PCR was carried out as per [43] . An initial 10 cycle pre-amplification with flanking primers 1689+1249 for UhAvr1 and 1804+1805 for reference gene UheIF-2B ( = UHOR_07772; Table S4 ) was carried out on a Bio-Rad MyCycler ( conditions: ( 1 ) 95°C 2 min . , ( 2 ) 10 cycles of 95°C 30 sec , 55°C 30 sec , 72°C 60 sec ) , followed by the qRT-PCR process above performed with nested internal primers 1798+1799 for UhAvr1 and 1811+1812 for reference gene UheIF-2B ( Table S4 ) . Total protein was isolated from frozen ground cells , as described [95] . Protein samples were boiled for 5 min and spun briefly for 30 sec before being separated by 12 . 5% SDS-PAGE on a Bio-Rad Mini-Protean III apparatus . Protein was transferred from the gel to Sequi-Blot PVDF Western blotting membrane ( Bio-Rad ) using a Bio-Rad liquid transfer apparatus following the manufacturer's protocols . Membranes were probed with 200 ng/ml rat anti-HA ( hemagglutenin ) high affinity monoclonal antibody ( Roche Applied Science ) or anti-GFP ( Clontech Living Colors JL-8 anti-GFP monoclonal ) . For detection of primary bound antibody , membranes were incubated with peroxidase-conjugated AffiniPure Goat Anti-Rat-Ig ( H+L ) secondary antibody according to supplier's instruction . For visualization of bound antibody , the Enhanced Chemiluminescence system ( ECL ) plus Western Blotting Detection Reagents ( Amersham Biosciences/GE Healthcare ) were used . To inoculate barley coleoptiles with teliospores , seed hulls were removed by hand to expose the embryo . Seeds were surface sterilized as above and germinated for 48 hrs in the dark at 18°C on sterile filter paper . Emerged coleoptiles were then dusted gently with a paintbrush with teliospores previously released from an infected seed head by gentle grinding . Alternatively , seeds germinated for 24 hrs with emerged coleoptiles , were immersed in cell cultures of OD600 ∼1 , mated for 24 hrs after mixing MAT-1 and MAT-2 strains in a 1∶1 ratio , under a vacuum of 20 psi for 20 min , after which the inoculum was drained . After inoculation , seedlings were kept moist and were further incubated in the dark at 18°C . Observation of GFP-expressing fungal infection was done on a Leica SP2-AOBS laser scanning confocal microscope at 488 nm excitation and detection at 499–552 nm . For light microscopy , seedlings were sampled at 72 , 96 , 120 and 144 hrs following inoculation . Plants were gently washed and crown tissues consisting of a 1 to 2 cm section of the coleoptile surrounding the crown region were excised , split longitudinally in half and both halves were mounted in lactophenol-cotton ( aniline ) blue to stain for callose [96] . Sections were viewed with a Zeiss Universal microscope using the 330–385 nm and 460–490 nm excitation and emission filters , respectively , and a HBO103W/2 light source . For detection of the oxidative burst , hydrogen peroxide was detected by vacuum infiltrating dissected coleoptiles for 10 min with 1 mg/ml 3 , 3′- diaminobenzidine tetrahydrochloride ( DAB , Sigma ) in 10 mM Na2HPO4 , pH 7 and 0 . 05% v/v Tween 20 , incubation for 6 hrs , and subsequent bleaching in a 3∶1∶1 ethanol ∶ acetic acid ∶ glycerol solution . The numbers of DAB stained sites and their relative size on both halves of 1 cm coleoptile sections were counted from a minimum of 5 seedlings per replication . Three replications were employed and the study was repeated two times . For quantitation of callose , average fluorescence associated with penetration sites was measured on 5 ( compatible interaction ) to 11 ( incompatible interaction ) TIF images imported into ImageJ software ( National Institutes of Health , Bethesda , Maryland ) and the average background fluorescence was subtracted . Data were analyzed using PROC GLM with SAS software ( SAS Institute , Cary , NC , USA ) and means were separated using Duncan's multiple range test ( P≤0 . 05 ) .
|
Upon host infection , plant pathogens secrete suites of virulence effectors to suppress defense responses and support their own development . In certain cases , hosts evolve resistance genes that recognize such effectors or their actions to initiate defense responses . By deleting candidate genes , we identified the immune-triggering effector UhAvr1 from Ustilago hordei , a barley-infecting basidiomycete smut fungus . We show that this effector is expressed only when hyphae sense and infect barley coleoptile epidermal cells . Its presence in the fungus causes a necrotic reaction immediately upon penetration resulting in complete immunity in barley cultivars having resistance gene Ruh1 . We show that fungal isolates that have mutated to change the expression of this non-crucial protein are avoiding recognition by the host , hence overcoming restriction by its immune response . In virulent isolates , transposable elements , known as genome modifiers , have separated the UhAvr1 coding region from its transcription signals . UhAvr1 is located in a larger cluster of ten effectors and is similar to clusters with more and further diversified effectors in the related maize pathogens U . maydis and Sporisorium reilianum . This study should lead us to discovering a mechanism by which this major cereal crop protects itself against this pathogen .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"genome",
"evolution",
"pathology",
"and",
"laboratory",
"medicine",
"microbiology",
"gene",
"function",
"cereal",
"crops",
"mutation",
"fungi",
"crops",
"molecular",
"genetics",
"mycology",
"medical",
"microbiology",
"microbial",
"pathogens",
"comparative",
"genomics",
"crop",
"diseases",
"pathogenesis",
"agriculture",
"host-pathogen",
"interactions",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"computational",
"biology",
"organisms"
] |
2014
|
An Immunity-Triggering Effector from the Barley Smut Fungus Ustilago hordei Resides in an Ustilaginaceae-Specific Cluster Bearing Signs of Transposable Element-Assisted Evolution
|
The increasing incidence of dengue among adults in Malaysia and other countries has important implications for health services . Before 2004 , in order to cope with the surge in adult dengue admissions , each of the six medical wards in a university hospital took turns daily to admit and manage patients with dengue . Despite regular in-house training , the implementation of the WHO 1997 dengue case management guidelines by the multiple medical teams was piecemeal and resulted in high variability of care . A restructuring of adult dengue inpatient service in 2004 resulted in all patients being admitted to one ward under the care of the infectious disease unit . Hospital and Intensive Care Unit admission criteria , discharge criteria and clinical laboratory testing were maintained unchanged throughout the study period . To evaluate the impact of cohorting adult dengue patients on the quality of care and the clinical outcome in a university hospital in Malaysia . A pre ( 2003 ) and post-intervention ( 2005–6 ) retrospective study was undertaken . Cohorting all dengue patients under the care of the Infectious Disease team in a designated ward in 2004 . The number of patients enrolled was 352 in 2003 , 785 in 2005 and 1158 in 2006 . The evaluation and detection of haemorrhage remained high ( >90% ) and unchanged throughout the study period . The evaluation of plasma leakage increased from 35 . 4% pre-intervention to 78 . 8% post-intervention ( p = <0 . 001 ) while its detection increased from 11 . 4% to 41 . 6% ( p = <0 . 001 ) . Examination for peripheral perfusion was undertaken in only 13 . 1% of patients pre-intervention , with a significant increase post-intervention , 18 . 6% and 34 . 2% respectively , p = <0 . 001 . Pre-intervention , more patients had hypotension ( 21 . 5% ) than detected peripheral hypoperfusion ( 11 . 4% ) , indicating that clinicians recognised shock only when patients developed hypotension . In contrast , post-intervention , clinicians recognised peripheral hypoperfusion as an early sign of shock . The highest haematocrit was significantly higher post-intervention but the lowest total white cell counts and platelet counts remained unchanged . A significant and progressive reduction in the use of platelet transfusions occurred , from 21 . 7% pre-intervention to 14 . 6% in 2005 and 5 . 2% in 2006 post-intervention , p<0 . 001 . Likewise , the use of plasma transfusion decreased significantly from 6 . 1% pre-intervention to 4 . 0% and 1 . 6% in the post-intervention years of 2005 and 2006 respectively , p<0 . 001 . The duration of intravenous fluid therapy decreased from 3 days pre-intervention to 2 . 5 days ( p<0 . 001 ) post-intervention; the length of hospital stay reduced from 4 days pre- to 3 days ( p<0 . 001 ) post-intervention and the rate of intensive care admission from 5 . 8% pre to 2 . 6% and 2 . 5% post-intervention , p = 0 . 005 . Cohorting adult dengue patients under a dedicated and trained team of doctors and nurses led to a substantial improvement in quality of care and clinical outcome .
Dengue is the most rapidly advancing vector-borne disease of public health significance in the tropics . Before 1970 , only nine countries had experienced severe dengue outbreaks . In the last 50 years , the incidence of dengue has increased 30 fold; more than 2 . 5 billion people ( >40% of world’s population ) live in >100 endemic countries [1] . Its more severe form , dengue haemorrhagic fever ( DHF ) was typically a childhood disease in Southeast Asia with deaths occurring mainly in children . Since the 1980’s , however there has been an increasing incidence of DHF among older age groups in Latin America and Southeast Asia [2] . The epidemiology of dengue disease in Malaysia was characterized by a non-linear increase in the number of reported cases from 7 , 103 in 2000 to 46 , 171 in 2010 , and a shift in the age range predominance from children toward adults [3] . The proportion of dengue-related deaths aged > 15 years similarly increased from 49% of total dengue deaths in 1999 , to 86% in 2006 [4] . The phenomenon of predominantly adult dengue population has also been observed in Singapore and Taiwan [5–8] . Other countries that have reported increased dengue infection in adults include Thailand [9–10] , India [11] , Bangladesh [12] , Cuba [13–14] and Brazil [15] . The transition in the epidemiology of dengue from a childhood disease to one that affects the older age group was reflected in hospital admissions throughout urban areas of Malaysia . The shift in the modal age group of DHF from children to the adult population has important implications for health services , resources , and training , provision of treatment and care and ultimately clinical outcome of dengue . To date there is no literature on the impact of this shift in the epidemiological pattern of dengue on healthcare delivery and the clinical outcome of adult dengue patients . University of Malaya Medical Centre ( UMMC ) is a teaching hospital situated in the Klang Valley that has recorded a large number of dengue related outpatient attendance and hospital admissions over the last several years . Patients with suspected dengue who required inpatient care were admitted through the Department of Primary Care Medicine or through the Emergency Department to the Departments of Paediatrics or Medicine respectively . Before 2004 , in order to cope with the surge in adult dengue admissions , each of the 6 general medical wards in UMMC took turns daily to admit and manage patients with dengue . Within each ward , care was delivered by a medical team comprising of the respective consultant , specialists , and several registrars and interns who rotated through the general medical wards . Despite regular in-house training , the implementation of the WHO 1997 dengue case management guidelines [16] by the multiple medical teams was piecemeal and resulted in high variability of care which included the transfusion of blood products . This situation prompted a restructuring of adult dengue inpatient service in 2004 , resulting in all these patients being admitted to one ward under the care of the infectious disease ( ID ) unit , led by an ID consultant , a specialist and 2 to 3 registrars and rotating interns . On-going training in dengue case management was continued as before . Critically ill dengue patients were admitted to the Intensive Care Unit ( ICU ) under the care of Department of Anaesthesiology and Intensive Care . Hospital and ICU admission criteria , discharge criteria and clinical laboratory testing were maintained unchanged throughout the study period . The objective of our study was to evaluate the impact of cohorting of dengue inpatients under the care of the ID unit , on the quality of care and the clinical outcome in adult dengue patients in UMMC .
All data were analysed using SPSS for Windows , version 20 . Data from the post-intervention group ( 2005 and 2006 ) was compared with that of the pre-intervention group ( 2003 ) . The Chi-square and Mann-Whitney tests were used where appropriate . The level of statistical significance for all analyses was set at p<0 . 05 using 2-tailed comparisons .
Results of the study are shown in Tables 1 and 2 . During the study period , the number of adult dengue inpatients increased progressively from 671 in 2003 to 1096 and 1 , 404 in 2005 and 2006 respectively . The number of patients enrolled into the study was 352 ( 52 . 4% of admissions ) , 785 ( 71 . 6% ) and 1158 ( 82 . 5% ) in 2003 , 2005 , 2006 respectively . The slight predominance of male patients increased significantly from pre-intervention to post-intervention period and this pattern was also observed in the corresponding total dengue inpatients . There was no difference in the median age of the enrolled population and the median duration of illness at the time of admission throughout the study period . The documentation of haemorrhagic manifestations by physicians was consistently high ( <90% ) during the pre and post-intervention periods , with no significant difference in the observed incidence of any bleeding ( 57% ) or the more serious gastro-intestinal bleeding ( 4 . 5% to 5 . 8% ) throughout the study period . In contrast , only 35 . 8% of patients’ medical records had any documentation of plasma leakage during the pre-intervention period; but any documentation of this phenomenon by doctors increased significantly and progressively during the 2 years post-intervention to 71 . 7% , p<0 . 001 , being 59% and 78 . 8% in 2004 and 2005 respectively . Together with increased detection for plasma leakage , so its presence increased , from 11 . 4% pre-intervention to 34 . 4% , post-intervention , p<0 . 001 , ( 13 , 8% in 2005 and 47 . 6 in 2006 ) . Likewise , examination for peripheral perfusion was undertaken in only 13 . 1% of patients pre-intervention , with a significant increase to 27 . 6% post-intervention , p = <0 . 001 ( 18 . 6% and 34 . 3% respectively ) . The presence of peripheral hypoperfusion increased significantly together with an increase in documentation of this physical sign , from 11 . 4% pre-intervention to 23 . 5% post-intervention , p<0 . 001 , ( 14 . 1% and 29 . 9% respectively ) . Conversely , the incidence of hypotension and narrowed pulse pressure decreased significantly post-intervention , 9 . 8% , from 21 . 5% pre-intervention , p<0 . 001 , ( 10 . 2% and 9 . 5% respectively ) . The ratio of patients observed to have peripheral signs of hypoperfusion to those with hypotension increased from 0 . 53 pre-intervention to 2 . 41 , post-intervention . The highest haematocrit ( HCT ) and the HCT before the start of intravenous fluid was significantly higher post-intervention . There was no significant difference in the lowest total white cell count and lowest platelet count throughout the illness during the pre and post-intervention periods . The duration of intravenous fluid therapy showed a slight but significant decrease from a median of 3 days pre-intervention to a median of 2 . 5 days ( p<0 . 001 ) post-intervention . Similarly , the length of hospital stay , reduced from a median of 4 days pre- to that of 3 days ( p<0 . 001 ) post-intervention . In addition , a significant reduction in the use of platelet transfusions was observed , from 21 . 7% pre-intervention to 9% post-intervention , p<0 . 001 ( 14 . 6% in 2005 and 5 . 2% in 2006 ) . Likewise , the use of plasma transfusion decreased significantly from 6 . 1% pre-intervention to 2 . 6% post-intervention , p = 0 . 003 ( 4 . 0% and 1 . 6% in 2005 and 2006 respectively ) . Although there was no difference in the lowest platelet counts there was , however , a significant difference in the lowest platelet count among patients who received platelet transfusions , 13 x109/L pre-intervention to 9 . 5% post-intervention ( 10 x109/L and 8 x109/L in 2005 and 2006 respectively ) , p<0 . 001 . The lowest platelet counts in patients who did not receive platelet transfusions decreased significantly during the post-intervention , p = 0 . 001 . There was a decrease in use of blood transfusion and antibiotics during the post-intervention period , the latter being a significant reduction , p = 0 . 046 . The rate of ICU admissions decreased significantly from 5 . 8% pre to 2 . 5% post-intervention , p = 0 . 002 . The overall mortality rate decreased from 0 . 75% pre-intervention to 0 . 56% post-intervention , this difference was however not significant , p = 0 . 857 . The rate of positive IgM for dengue increased significantly post-intervention , from 69 . 5% pre to 93% post intervention , p<0 . 001 , with significantly more patients , 34 . 3% and 56 . 7% respectively having repeated serology tests , p<0 . 001 .
Our study demonstrates that cohorting dengue patients under the care of a trained team of clinicians were associated with several changes in clinical practice . The pre and post-intervention groups of patients were comparable in their median age and the day of illness at the time of admission . They were comparable in the severity of illness as represented by the presence of any haemorrhage , gastrointestinal haemorrhage , and the lowest platelet and white blood cell counts . While the majority of clinicians were well aware of the haemorrhagic manifestations of dengue , only a minority were knowledgeable of plasma leakage before the intervention . The higher level of awareness of clinicians to haemorrhagic manifestations may be due to the term “dengue haemorrhagic fever” which placed emphasis on haemorrhage rather than plasma leakage as the main pathophysiology of severe dengue . Under-recognition of plasma leakage was identified as one of the difficulties causing the inappropriate classification of dengue using the 1997 WHO dengue case classification [17 , 18] . With repeated exposure over a period of time , a progressive increase in awareness among clinicians to the unique feature of plasma leakage and the ensuing shock led to a practice where these physical signs were actively elicited . Clinical detection for plasma leakage increased from 2003 to 2006 resulting in a higher number of patients identified with this complication . A similar but smaller increase in clinicians’ awareness to peripheral signs of shock was also noted . As a result of an increase in the detection of peripheral hypoperfusion , which precedes hypotension , the incidence of central hypoperfusion among dengue inpatients decreased . More patients were detected to have hypotension ( 21 . 5% ) than hypoperfusion ( 11 . 4% ) pre-intervention , indicating that clinicians recognised shock only when patients developed hypotension , a finding similar to that of an earlier study in 2002 [17] . In contrast , during the post-intervention period , clinicians recognised peripheral hypoperfusion as an early sign of shock . A slight but significant increase in the highest haematocrit from the pre-intervention to post-intervention period suggested that the latter patients could have more severe plasma leakage . It is possible that a few hours of delay in blood sampling during the plasma leakage phase of illness , might result in an increase in haematocrit . This delay might be due to a larger volume of inpatients during the post-intervention phase . However , earlier detection and correction of hypovolaemia might have prevented patients from progressing to hypotensive shock necessitating ICU management , resulting in a significant decrease in ICU admission post-intervention . There was a non-significant reduction on overall mortality observed in the post intervention compared to the pre-intervention period . A marked decrease in the use of platelet and plasma transfusions was one of the most noticeable effects of cohorting dengue patients . During the study period , the threshold to transfuse platelets for thrombocytopenia increased significantly , from 13 x109/L pre-intervention compared to 10 x109/L and then 8 x109/L during the post-intervention years of 2005 and 2006 respectively . Over time , experiential knowledge of thrombocytopenia being not associated with bleeding in dengue [19–21] led clinicians to develop confidence to gradually abandon the practice of platelet or plasma transfusions . The duration of administration of intravenous fluid was reduced which was associated with a significant decrease in length of hospital stay post-intervention . Experience plays a central role in the learning process whereby knowledge and concepts are derived from and continuously modified by experience [22] . Clinical experience , based on personal observation , reflection , and judgment is needed to translate scientific results into treatment of individual patients [23] . Personal experience is often a more powerful persuader than scientific publication in changing clinical practice [24 , 25] . Cohorting dengue patients under a trained team provide an environment that favours the experiential learning process . Theoretical knowledge , for example , of plasma leakage and its complications and clinical skills in detection could be continually validated and internalized through repetitive clinical exposure . This is shown in the incremental improvement of clinical practices and clinical outcomes over time . Other benefits of cohorting dengue patients include transfer of experiential knowledge from paediatricians to adult physicians , focused training of nursing staff which , in turn , contributed to better monitoring of dengue patients for symptoms and signs of plasma leakage and shock and oral and intravenous fluid balance . Teams dedicated to the case management of dengue have been successfully formed in countries such as Thailand and Viet Nam [26–28] , where case fatality rates for children with DHF have been the lowest in the region . Our study has several limitations . It was a retrospective study entirely dependent on the availability of medical records of randomly selected patients . This might have an effect on the enrolment of our patients and resulting in a selection bias . The failure to record clinical information reflected either a lack of awareness of clinicians or incomplete documentation . The fact that a high proportion of patients with documented data on haemorrhage throughout the study period suggests a high level of awareness of bleeding as a complication of dengue . Thus , a lack of documentation of evidence of plasma leakage or shock suggests a lack of awareness of these complications . Not all dengue inpatients had laboratory confirmation of dengue , perhaps again reflecting the low level of knowledge pre-intervention , that a negative IgM for dengue will need to be repeated . The rate of seropositivity for dengue IgM increased when repeated later in the disease . It could be that changes in practices between 2003 and 2005–6 reflect overall temporal trends rather than an effect of cohorting patients , as there were progressive changes over the two years post-intervention . However , a study conducted in 2002 in the same institution and in the same setting of dispersed placement of adult dengue inpatients showed that only 7% of adult dengue inpatients with thrombocytopenia and evidence of plasma leakage were correctly classified as DHF , thus reflecting a low level of awareness among clinicians of the main pathophysiology of severe dengue [17] . Improving dengue case management is a significant priority in ensuring improved clinical outcomes . Cohorting dengue inpatients under the care of a dedicated and trained team of doctors and nurses enhanced the acquisition of experiential knowledge and clinical experience , which is translated into improved clinical practice and clinical outcome . The changes in clinical practices during the post-intervention period were the higher rates of elicitation of evidence of plasma leakage and peripheral hypoperfusion and repeat of dengue serology when negative . The observed changes in clinical outcomes were a significant reduction in duration of intravenous fluid therapy , incidence of ICU admissions and transfusions of platelets and plasma . Our study also provides evidence that preventive transfusions of blood products , widely practised in countries endemic in dengue , are unnecessary [29 , 30] . By diverting the focus of management away from preventive transfusions , more attention could be given to the major problem in severe dengue , increased capillary permeability . All these would translate into substantial reduction in costs of treatment . In addition to training a dedicated team , allocating a designated ward for dengue patients in endemic countries or during a dengue outbreak should be considered as one of the effective strategies in improving clinical outcome .
|
The epidemiology of dengue disease in the tropical regions is characterized by a rapid increase in the number of reported cases , and in some Asian countries , a shift in the age range predominance from children toward adults . This has important implications for health services , resources , and training , provision of treatment and care and ultimately clinical outcome of dengue . There is no literature on the impact of this shift in the epidemiological pattern of dengue on healthcare delivery and the clinical outcome of adult dengue patients . Before 2004 , each of the 6 general medical wards in a university hospital took turns daily to admit and manage patients with dengue . Despite regular in-house training , the implementation of the WHO 1997 dengue case management guidelines by the multiple teams was piecemeal with high variability in clinical practice . In 2004 all dengue patients were admitted to one ward under the care of the infectious disease unit . In this pre ( 2003 ) and post-intervention ( 2005–6 ) retrospective study we demonstrated that cohorting dengue patients under a specific team enhances the experiential knowledge of clinicians in managing dengue patients from the perspective of clinical evaluation and detection of increased vascular permeability and the ensuing hypovolemic shock with improvement in clinical outcomes .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Cohorting Dengue Patients Improves the Quality of Care and Clinical Outcome
|
Plasmacytoid dendritic cells ( pDC ) produce type I interferons ( IFN-I ) and proinflammatory cytokines in response to viruses; however , their contribution to antiviral immunity in vivo is unclear . In this study , we investigated the impact of pDC depletion on local and systemic antiviral responses to herpes simplex virus ( HSV ) infections using CLEC4C-DTR transgenic mice . We found that pDC do not appear to influence viral burden or survival after vaginal HSV-2 infection , nor do they seem to contribute to virus-specific CD8 T cell responses following subcutaneous HSV-1 infection . In contrast , pDC were important for early IFN-I production , proinflammatory cytokine production , NK cell activation and CD8 T cell responses during systemic HSV-2 and HSV-1 infections . Our data also indicate that unlike pDC , TLR3-expressing cells are important for promoting antiviral responses to HSV-1 regardless of the route of virus administration .
Most cells are able to produce type I interferons ( IFN-I ) in response to viruses , however , some cell types such as plasmacytoid dendritic cells ( pDC ) are more efficient than others . pDC detect RNA and DNA viruses through two endosomal sensors , Toll-like receptor ( TLR ) 7 and TLR9 , respectively , which induce secretion of IFN-I through the MyD88-IRF7 signaling pathway [1]–[3] . Because of their capacity to produce IFN-I , as well as proinflammatory cytokines , and their ability to present antigens to T cells , pDC are thought to be important for promoting immune responses , particularly to viruses [4] , [5] . In order to evaluate the contribution of pDC to innate and adaptive antiviral responses in vivo , depletion studies are warranted . Several mouse models to eliminate pDC have been described . First attempts used antibody ( Ab ) -mediated depletion of pDC [6] . Within the past few years , genetically modified mouse strains have become available that lack pDC either constitutively [7] , [8] or by inducible depletion [9] , [10] . CLEC4C-DTR transgenic ( Tg ) mice have been generated that express the diphtheria toxin receptor ( DTR ) under the control of the CLEC4C promoter [9] . CLEC4C , also known as blood dendritic cell antigen 2 , is a type II C type lectin that is uniquely expressed by human pDC [11] , [12] . Injection of diphtheria toxin ( DT ) into CLEC4C-DTR Tg mice selectively eliminates pDC [9] . Recently , a SiglecH-DTR knockin mouse was described that has an IRES-DTR-EGFP cassette inserted into the SiglecH locus [10] . These mice not only lack SiglecH expression , but can also be depleted of pDC after DT administration . SiglecH is a member of the sialic acid-binding immunoglobulin ( Ig ) -like lectin family that is routinely used to discriminate pDC from other cell types in mice [13] , [14] . Herpes simplex virus ( HSV ) -1 and HSV-2 are large double-stranded DNA viruses that infect epithelial or epidermal cells before establishing a latent infection in sensory neurons [15] . Both innate and adaptive immune responses are necessary for limiting viral replication and maintaining latency [16] . pDC detect HSV and produce IFN-I and proinflammatory cytokines via TLR9 [17]–[20] . Ab-mediated depletion studies have suggested a critical role for pDC in promoting immunity to HSV both locally and systemically . Because the available pDC-depleting Abs also cross-react with other cell types , we decided to investigate the impact of pDC depletion on local and systemic antiviral responses to HSV infections using CLEC4C-DTR Tg mice . We found that the absence of pDC did not appear to influence antiviral responses to local HSV-2 and HSV-1 infections . In contrast , pDC were important for IFN-I production , NK cell activation and CD8 T cell responses following systemic HSV-2 and HSV-1 infections . Our findings suggest that previous studies highlighting a protective role for pDC during local HSV infections may be related to the depletion of other cell types . Our data also corroborate previously published findings that TLR3-expressing cells , unlike pDC , are critical for antiviral CD8 T cell responses to HSV-1 regardless of the route of administration [21] and are a major source of IFN-I during systemic infection that promotes NK cell activation at later time points post-infection ( p . i . ) . Altogether , our study demonstrates that the cellular sources of IFN-I and antigen-presenting cells that mediate antiviral responses to HSV infections vary depending on the route of infection .
It has been reported that in humans pDC infiltrate the dermis of recurrent genital herpes simplex lesions and are closely associated with T cells and NK cells and thus , may contribute to control of recurrent herpes virus infection in vivo [22] . pDC have also been detected in the vaginal mucosa of mice infected with HSV-2 [23] , [24] . Ab-mediated depletion of pDC using 120G8 [25] or mPDCA1 [26] during vaginal HSV-2 infection impairs innate antiviral responses [23] . pDC-depleted mice showed higher viral titers on days 2 and 3 p . i . , reduced IFN-I responses and survival following HSV-2 challenge [23] . Since the available antibodies to deplete pDC , which include 120G8 [25] , mPDCA1 [26] , 927 [27] and Gr-1 [28] , [29] , also react with other cell types we decided to evaluate the involvement of pDC in vaginal HSV-2 infection using genetically modified mice where pDC are selectively eliminated after DT administration [9] . To this end , CLEC4C-DTR Tg mice were injected with phosphate buffered saline ( PBS ) or DT then infected intravaginally ( ivag ) with HSV-2 . We found that pDC in the lumbar lymph nodes , which drain the vagina , were efficiently depleted at the time of infection ( Figure 1A ) . Depletion of pDC did not impact viral burden or IFN-α in vaginal tissues on day 2 p . i . ( Figure 1B and 1C ) . We next determined whether pDC ablation impaired survival to HSV-2 infection . Mice were depleted of pDC or not and challenged with two different doses of HSV-2 ivag ( Figure 1D ) . Survival was monitored for 15 days . Both groups of mice infected with 1×104 pfu of virus had a survival rate of approximately 20% by day 15 . When a lower dose of virus was administered ( 2×103 pfu ) , around 50% of control and pDC-depleted mice succumbed to infection by day 15 . These results suggest that pDC do not strongly influence viral burden or survival following vaginal HSV-2 infection and that previous findings describing a major role for pDC in antiviral responses to HSV-2 may be related to the depletion of other cell types that are critical for controlling vaginal HSV-2 infection . It has also been described , using Ab-mediated ablation , that pDC are imperative for optimal HSV-1-specific CD8 T cell responses after cutaneous infection [30] . In this study , pDC depletion impaired DC functions and cytotoxic T cell-mediated virus eradication in draining lymph nodes ( DLN ) [30] . Given that Ab administration may potentially deplete non-pDC in this model as well , we decided to evaluate local anti-HSV-1 CD8 T cell responses in CLEC4C-DTR Tg mice depleted or not of pDC . First , we confirmed that pDC are efficiently depleted in the popliteal lymph nodes of DT-treated mice ( Figure 2A ) . We then infected control and depleted mice with HSV-1 in the footpad . On day 7 p . i . we measured total numbers of pDC , CD8 and CD4 T cells in DLN; the ability of cells from DLN to lyse target cells pulsed with HSV gB peptide; frequencies of gB-specific CD8 T cells in DLN and spleen; and viral burden in DLN ( Figure 2 and data not shown ) . Our results show that pDC were largely reduced in the DLN of DT-treated mice on day 7 p . i . suggesting their absence throughout the course of infection ( Figure 2B ) and that absolute numbers of CD8 and CD4 T cells in DLN were comparable between PBS and DT-treated mice ( Figure 2C ) . We also found using standard chromium release assays that cells from DLN of PBS and DT-treated mice were equally capable of lysing peptide-pulsed target cells ( Figure 2D ) . Furthermore , when cells from DLN and spleen were restimulated with HSV gB peptide both PBS and DT-treated mice had approximately 2% of total CD8 T cells in DLN and 12% of total CD8 T cells in spleen that produced IFN-γ by intracellular staining ( Figure 2E and 2F ) . Finally , we attempted to measure viral burden in DLN by plaque assay on day 7 p . i . but were not able to detect HSV-1 in either PBS or DT-treated mice ( data not shown ) . This is in contrast to the study by Yoneyama et al . which detected HSV-1 in DLN on day 7 p . i . in pDC-depleted mice [30] . Taken together these data suggest that the contribution of pDC to antiviral CD8 T cell responses during local HSV-1 infection is negligible in the CLEC4C-DTR Tg mouse model . Previous studies using either Ab-mediated depletion of pDC or SiglecH-DTR knockin mice injected with DT to eliminate pDC have demonstrated a major role for pDC in innate and/or adaptive responses to systemic HSV-2 or HSV-1 infections [10] , [31] , [32] . Depletion of pDC with Ab in mice infected systemically with HSV-2 results in reduced serum IFN-I levels , augmented IL-17 production and higher viral burden in liver [31] , [32] . In the case of HSV-1 , SiglecH-DTR knockin mice depleted of pDC have impaired systemic IFN-α levels and virus-specific CD8 T cell responses [10] . Therefore , we next decided to investigate whether pDC depletion in our mouse model impacted antiviral responses to systemic HSV infections . To this end , we injected CLEC4C-DTR Tg mice with PBS or DT then infected mice intravenously ( i . v . ) with either HSV-2 ( Figure 3 ) or HSV-1 ( Figure 4 ) . In the absence of pDC , there was a dramatic reduction in serum IFN-α and IFN-γ levels 8 h after systemic HSV-2 infection ( Figure 3A ) . IL-12p70 levels also appeared to be reduced in pDC-depleted mice but numbers did not reach statistical significance ( Figure 3A ) . Serum IFN-α levels were still decreased in pDC-depleted mice at 12 h p . i . ( Figure 3B ) and NK cell activation , as measured by intracellular IFN-γ and CD107a staining , was impaired in the absence of pDC ( Figure 3C–3E ) . However , in contrast to the study by Stout-Delgado et al . [32] , serum IL-17 levels were undetectable in our PBS or DT-treated mice ( data not shown ) . Finally , we performed a dose response experiment to determine whether pDC-depleted mice were more susceptible to systemic HSV-2 infection ( Figure 3F ) . Mice were injected with PBS or DT and infected with different doses of HSV-2 i . v . At a dose of 1×107 , mice in both groups died on days 4 and 5 , while at the lowest dose ( 1×103 ) all mice survived . Infection with 1×105 pfu per mouse revealed a strong difference in survival between control and pDC-depleted mice ( 80% versus 20% ) . Taken together , these data indicate that pDC promote antiviral responses and play a protective role during systemic HSV-2 infection . Similar to our findings with systemic HSV-2 infection , CLEC4C-DTR Tg mice depleted of pDC had impaired antiviral responses to systemic HSV-1 ( Figure 4 ) . Analyses of serum IFN-α ( Figure 4A ) and IFN-γ ( Figure 4B ) levels revealed that pDC-depleted mice had lower amounts of both cytokines compared to PBS-treated mice at 6 h p . i . Although pDC-depleted mice had ∼3 fold less serum IFN-γ at 12 h p . i . ( Figure 4B ) , they had a modest reduction in levels of IFN-α ( Figure 4A ) compared to control mice , suggesting that other cell types and/or viral sensors are involved in sensing systemic HSV-1 infection and producing IFN-I . Examination of spleens from HSV-1-infected mice at 12 h p . i . revealed that NK cells from pDC-depleted mice expressed lower levels of CD69 ( Figure 4C ) . Additionally , spleens from DT-treated mice had lower frequencies of CD107a+ and IFN-γ-producing NK cells compared to PBS-treated mice ( Figure 4D–4F ) . We also evaluated whether CD4 T cells and NKT cells produced IFN-γ in infected control and pDC-depleted mice ( data not shown and Figure 4G ) . Although we did not detect IFN-γ production by CD4 T cells at this time point , we did observe similar frequencies of IFN-γ-producing NKT cells from PBS and DT-treated mice . On day 7 p . i . we measured virus-specific CD8 T cell responses in spleens by intracellular IFN-γ staining after restimulation with HSV gB peptide ( Figure 4H ) . pDC-depleted mice showed a 25–30% reduction in IFN-γ-producing CD8 T cells compared to PBS-treated mice . We have previously shown that pDC do not prime virus-specific CD8 T cells but appear to promote their survival during vesicular stomatitis virus infection [9] . To determine how pDC impact virus-specific CD8 T cells during systemic HSV-1 infection we injected CFSE-labeled CD8 T cells from gBT-I Tg mice [33] into CLEC4C-DTR Tg mice . Following administration of PBS or DT and infection with HSV-1 we measured accumulation and proliferation of gB-specific CD8 T cells in spleens on day 3 p . i . ( Figure 4I ) . Transferred T cells accumulated in control and pDC-depleted mice and exhibited similar levels of CFSE dilution and CD25 , suggesting that the absence of pDC may alter the functional ability of virus-specific CD8 T cells to produce IFN-γ rather than their proliferation or accumulation . Although our IFN-I and CD8 T cell results during systemic HSV-1 infection are similar to those obtained using pDC-depleted SiglecH-DTR knockin mice [10] , we were not able to detect viral replication in spleens of either PBS or DT-treated CLEC4C-DTR Tg mice on day 7 p . i . ( data not shown ) in contrast to the Takagi et al . study . This discrepancy may reflect differences in the strains of HSV-1 used ( KOS versus strain F ) or differences in the promoter used to drive DTR expression ( CLEC4C versus SiglecH ) . It has been shown that SiglecH is expressed by specialized macrophage subsets and progenitors of pDC and classical DC [14] , [34] , [35] . Thus , it is possible that certain SiglecH+ subsets other than pDC are functionally altered and/or depleted in SiglecH-DTR knockin mice thus yielding stronger phenotypes ( i . e . major reductions in IFN-I and/or CD8 T cell responses and increased viral replication ) than what we observe in CLEC4C-DTR Tg mice . Although we detected reduced IFN-α responses in pDC-depleted mice 6 h post-HSV-1 infection , at 12 h p . i . there were modest differences in IFN-α levels between control and pDC-depleted mice ( Figure 4A ) . These data indicate that pDC are not the only source of IFN-α during systemic HSV-1 infection and prompted us to examine which sensor was responsible for IFN-α production in the presence or absence of pDC at later time points . TLR3 has a prominent role in HSV-1 recognition and IFN-I production [36]–[38] . In human , TLR3 has proven essential for the recognition of HSV-1 in the brain most likely though the generation of RNA intermediates that occur during viral replication that gain access to the endosomal compartment by phagocytosis or endocytosis [36]–[39] . Thus , we infected wildtype ( WT ) and TLR3−/− mice i . v . with HSV-1 and measured serum IFN-α levels at 6 and 12 h p . i . We found that at 6 h p . i . TLR3-deficiency did not impact IFN-α levels; however , at 12 h p . i . IFN-α production was greatly diminished in TLR3−/− mice ( Figure 5A ) . These data indicate that pDC are important for the early burst of IFN-α during HSV-1 infection while TLR3 signaling is essential for the later burst of IFN-α . We also found that serum IFN-γ levels were decreased in TLR3−/− mice at 6 and 12 h p . i . ( Figure 5B ) , suggesting a defect in NK cell activation . Examination of spleens at 12 h p . i . revealed that NK cells from TLR3−/− mice expressed lower levels of CD69 ( Figure 5C ) . Moreover , spleens from TLR3−/− mice had reduced numbers of CD107a+ and IFN-γ-producing NK cells compared to infected WT mice ( Figure 5D–5F ) . We also looked at IFN-γ production by CD4 T cells and NKT cells in WT and TLR3−/− mice 12 h p . i . and observed a reduction in activated NKT cells but not CD4 T cells ( Figure 5G and data not shown ) . Prior work has demonstrated that in mice TLR3 expression and CD8α DC , which express high levels of TLR3 , are important for virus-specific CD8 T responses following subcutaneous , intravenous and flank HSV-1 infections [21] , [40] . To this end , we infected WT and TLR3−/− mice i . v . with HSV-1 and measured virus-specific CD8 T cells in spleens on day 7 p . i . by intracellular IFN-γ staining after restimulation with HSV gB peptide ( Figure 5H ) . TLR3−/− mice showed a ∼50% reduction in IFN-γ-producing CD8 T cells compared to WT mice . Next , we measured serum IFN-α levels and CD8 T cell responses in Batf3−/− mice , which have a defect in CD8α DC numbers [41] , [42] , after systemic HSV-1 infection . Interestingly , Batf3−/− mice produced lower levels of IFN-α compared to WT mice 12 h p . i . , which were similar to those observed in TLR3−/− mice ( Figure 5I ) . Analyses of spleens on day 7 p . i . revealed that Batf3−/− mice had fewer numbers of total CD8 T cells ( Figure 5J ) and had reduced frequencies and numbers of IFN-γ-producing CD8 T cells after restimulation with HSV gB peptide ( Figure 5K and 5L ) . Thus , we conclude that while pDC contribute modestly to late IFN-α production and CD8 T cell responses during systemic HSV-1 infection , TLR3-expressing cells , most likely CD8α DC , are required for late IFN-α production and efficient priming and/or expansion of virus-specific CD8 T cells [21] , [40] , [42] . Finally , to confirm that TLR3 expression is critical for antiviral responses to HSV-1 regardless of the route of infection we examined T cell numbers and responses in DLN of TLR3−/− mice following footpad infection . DLN from TLR3−/− mice contained fewer numbers of total CD4 and CD8 T cells ( Figure 6A and 6B ) as well as reduced numbers of IFN-γ-producing CD8 T cells after restimulation with HSV gB peptide ( Figure 6C ) . Taken together , these data corroborate previous findings and illustrate the importance of TLR3 expression in antiviral responses to HSV-1 .
In this study , we evaluated antiviral responses during local and systemic HSV infections in the presence and absence of pDC using CLEC4C-DTR Tg mice . We found that pDC have a negligible impact on viral burden , local IFN-I production and survival during vaginal HSV-2 infection and virus-specific CD8 T cell responses after subcutaneous HSV-1 infection . In contrast , pDC were essential for optimal IFN-I production , NK cell activation , CD8 T cell responses and survival following systemic HSV-2 and HSV-1 infections . These findings suggest that the contribution of pDC to antiviral immunity is dependent on the route of virus entry . Although pDC are often recruited to sites of infection and inflammation [43] , [44] , in homeostatic conditions they are mainly found in circulation and lymphoid tissues , which may explain why they have a more prominent role in controlling systemic viral infections . Indeed , it has been reported that following local infections with viruses such as Newcastle disease virus and vesicular stomatitis virus , macrophages are major sources of IFN-I and promote antiviral responses [45] , [46] . Previous studies concluding a major role for pDC in controlling local viral infections , such as HSV , may have been a consequence of depleting other cell types in addition to pDC . This is a likely scenario given that the currently available antibodies to deplete pDC cross-react with other cell types . In early studies , pDC were depleted with a monoclonal antibody ( mAb ) specific for the antigen Gr-1 [28] , [29] , which is shared by Ly6C and Ly6G . Ly6C is expressed on pDC , but also on plasma cells , activated T cells and inflammatory monocytes while Ly6G is expressed on granulocytes [47]–[49] . Several other pDC-depleting mAbs have been developed , including 120G8 , mPDCA1 and 927 [25]–[27] . These mAbs recognize bone marrow stromal cell Ag 2 ( BST2 ) . Although BST2 is selectively expressed on mouse pDC in naïve mice , it is also expressed by plasma cells and is upregulated on most cell types , including classical DC , B cells , T cells , myeloid cells , NKT cells and stromal cells following viral infections or stimulation with IFNs [27] . Therefore , it is important to corroborate these conclusions in mice that specifically lack pDC either transiently [9] , [10] , constitutively [7] , [8] or in mice that have a functional defect of pDC [50] , [51] . In contrast to pDC-deficient mice , TLR3−/− mice display impaired antiviral responses to HSV-1 regardless of the route of virus entry . It has previously been reported that TLR3 is required for the generation of CD8 T cell immunity to HSV-1 after flank infection [21] and our results indicate that TLR3−/− mice have reduced virus-specific CD8 T cell responses after systemic and footpad infections . CD8α DC , which express high levels of TLR3 , are essential for inducing T cell-mediated immunity to HSV following subcutaneous and systemic infections [40] , [42] . The significance of TLR3 in sensing HSV has been corroborated by studies showing that TLR3 deficiency in humans is associated with increased susceptibility to herpes simplex encephalitis , although this is mainly due to impairment of local responses rather than a defect a CD8 T cell responses [36]–[38] . The importance of IFN-I in controlling systemic HSV-1 infections has been confirmed in mice lacking IRF7 and STING , which have major defects in IFN-I responses [52] , [53] . Our data and others [19] suggest that although pDC contribute to IFN-I production during systemic HSV-1 , they are not the only source . While pDC are mainly important for early IFN-I secretion , TLR3-expressing cells , such as CD8α DC , are essential for IFN-I production at later time points . Other TLR3-expressing cells that might contribute to IFN-I during HSV-1 infection may include both hematopoietic and non-hematopoietic cells such as marginal zone macrophages , fibroblasts and stromal cells [36]–[38] , [54] . Taken together , our study indicates that some cell types are more important than others for establishing antiviral immunity to a given virus and that cell types expressing different viral sensors promote IFN-I production in a time-dependent manner during systemic HSV infections .
For mouse studies at Washington University , the principles of good laboratory animal care were carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and following the International Guiding Principles for Biomedical Research Involving Animals . The protocols were approved by the Animal Studies Committee at Washington University ( #20100280 ) . C57BL/6 , CLEC4C-DTR Tg and TLR3−/− mice were bred in house . Batf3−/− mice were kindly provided by K . Murphy and W . KC ( Washington University School of Medicine , St . Louis , MO ) . All mice were on a C57BL/6 background and used at 6–12 weeks of age . Diphtheria toxin ( DT , Sigma-Aldrich ) was injected i . p . into CLEC4C-DTR Tg mice ( 100–120 ng/mouse ) 24 h prior to infection and every other day after . Seven days prior to vaginal infections , female CLEC4C-DTR Tg mice were injected subcutaneously with 2 mg of Depo-Provera or medroxyprogesterone acetate in PBS as previously described [55] . HSV-1 KOS strain , provided by D . Leib ( Dartmouth College , Hanover , NH ) , was injected i . v . at 1×107 pfu or in the footpad at 1×105 pfu . HSV-2 strain 186TKΔkpn was provided by A . Iwasaki ( Yale University , New Haven , CT ) . Mice were infected with 5×106 pfu i . v . or 1×103 pfu ivag of HSV-2 . For susceptibility studies , mice were infected with different doses of HSV-2 either i . v . or ivag ( specified in figures/legends ) . Vaginal washes and vaginas and cervix from HSV-2-infected mice were removed on day 2 p . i . weighed and stored at −80°C in 1 ml of plain DMEM . Spleens and DLN from HSV-1-infected mice were removed on day 7 p . i . and stored at −80°C in 1 ml of plain DMEM . On the day of titering , samples were thawed in a 37°C water bath then homogenized with a Roche MagNA Lyser . Samples were centrifuged for 5 min at 5000 rpm then homogenates were serially diluted in plain DMEM . Dilutions were applied to Vero cell monolayers in 12 well plates for 1 h at 37°C . After the incubation period , an overlay containing complete DMEM and 1% methylcellulose was added and plates were incubated for 2–3 days . A second overlay with neutral red was applied 8 h before counting plaques . Spleens and lymph nodes were minced and digested for 45 min at 37°C in RPMI 1640 with collagenase D ( Sigma-Aldrich ) . Single cell suspensions were prepared by passage through nylon mesh cell strainers ( BD Biosciences ) . Red blood cells were lysed with RBC lysis buffer ( Sigma-Aldrich ) . Serum was separated from whole blood in serum collection tubes and stored at −20°C until analysis . CD8 T cells from gBT-I Tg mice [33] were isolated by negative selection using the CD8α+ T cell Isolation Kit II according to the manufacturer instructions ( Miltenyi Biotec ) . Cells were labeled with 5 µM CFSE and injected i . v . into CLEC4C-DTR Tg mice at 1×106 cells per mouse . One day after transfer , mice were injected with PBS or DT . One day later , mice were infected with 1×107 pfu HSV-1 i . v . Spleens were analyzed for CD8+CFSE+ T cells on day 3 p . i . EL4 cells were grown in complete RPMI: RPMI 1640 ( Gibco/Invitrogen ) with 10% bovine calf serum ( BCS ) , 1% glutamax , 1% nonessential amino acids , 1% sodium pyruvate and 1% kanamycin sulfate ( Gibco/Invitrogen ) . Vero cells were grown in DMEM with 10% BCS , 1% glutamax , 1% hepes and 1% pen/strep ( Gibco/Invitrogen ) . Primary cells were cultured in complete RPMI with 10% fetal calf serum ( FCS , Hyclone ) . The following reagents were from BD Biosciences , eBioscience or Biolegend . Fluorochrome labeled anti-CD3 ( 145-2C11 ) , anti-NK1 . 1 ( PK136 ) , anti-CD8α ( 53–6 . 7 ) , anti-CD4 ( MZ3 ) , anti-IFN-γ ( XMG1 . 2 ) , CD69 ( H1 . 2F3 ) , CD107a ( 1D4B ) , CD25 ( PC61 ) , and Streptavidin . Fc receptors were blocked before surface staining with supernatant from HB-197 cells ( ATCC ) . Propidium iodide was used to gate out dead cells . CD107a staining on NK cells was performed as previously described [56] . Briefly , splenocytes from control and infected mice were cultured for 1 h with FITC-conjugated CD107a . After the 1 h incubation , Brefeldin A and monensin were added to wells for an additional 4 h . Cells were stained with NK1 . 1 . and CD3 then fixed with 2% paraformaldehyde . All flow cytometry was conducted on a dual laser FACSCalibur flow cytometer ( BD Biosciences ) and analyzed with FlowJo software ( Tree Star , Inc . ) . Spleen or lymph node cells from HSV-1-infected mice were incubated in complete RPMI alone or with HSV gB peptide ( Anaspec , 10 µg/ml ) in the presence of Brefeldin A . After 6 h , cells were intracellularly stained for IFN-γ using the Cytofix/Cytoperm kit from BD Biosciences . IFN-γ-producing NK and NKT cells were also detected by intracellular staining using the Cytofix/Cytoperm kit . For Ag-specific lysis assays , lymph nodes from HSV-1-infected mice were resuspended in complete RPMI and serially diluted . EL4 cells were pulsed or not with HSV gB peptide ( 10 ng/ml ) and labeled with 1 mCi/ml 51Cr for 2 h then incubated with effector cells at 37°C for 4 h . 51Cr release in supernatants was measured with a γ-counter . Serum IFN-α and proinflammatory cytokine levels were determined by ELISA ( PBL Interferon Source ) or Cytometric Bead Array ( BD Biosciences ) , respectively . For measurements of IFN-α in tissue homogenates from HSV-2 infected mice , vaginas and cervix were removed and weighed on day 2 p . i . then stored at −80°C in tubes with silica beads and lysis buffer: PBS , PMSF , protease inhibitors , Triton X-100 . Samples were thawed in a 37°C water bath then homogenized with a Roche MagNA Lyser . Samples were centrifuged for 5 min at 5000 rpm and supernatants were diluted and used for ELISA . The statistical significance of differences in mean values was analyzed with unpaired , two-tailed Student's t-test using GraphPad Prism software . P values less than 0 . 05 were considered statistically significant .
|
Herpes simplex viruses ( HSV ) cause a variety of diseases in human from the common cold sore to more severe illnesses such as pneumonia , herpes simplex keratitis , genital herpes and encephalitis . HSV are large double-stranded DNA viruses that infect epithelial or epidermal cells before establishing a latent infection in sensory neurons . Both innate and adaptive immune responses are necessary for limiting viral replication and maintaining latency . Viral detection through distinct pathogen recognition pathways triggers several signaling cascades that lead to the production of proinflammatory cytokines and type I interferons , which establish inflammation , confer an antiviral state and promote immune responses . Our study provides new insights into the cell types and pathogen recognition pathways involved in antiviral defense to HSV at local and systemic barriers and thus , might facilitate the development of novel strategies to treat HSV infections .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Plasmacytoid Dendritic Cells Contribute to Systemic but Not Local Antiviral Responses to HSV Infections
|
Zika virus ( ZIKV ) is an important flavivirus infection . Although ZIKV infection is rarely fatal , risk for severe disease in adults is not well described . Our objective was to describe the spectrum of illness in U . S . Veterans with ZIKV infection . Case series study including patients with laboratory-confirmed or presumed positive ZIKV infection in all Veterans Health Administration ( VHA ) medical centers . Adjusted odds ratios of clinical variables associated with hospitalization and neurologic complications was performed . Of 1 , 538 patients tested between 12/2015-10/2016 and observed through 3/2017 , 736 ( 48% ) were RT-PCR or confirmed IgM positive; 655 ( 89% ) were male , and 683 ( 93% ) from VA Caribbean Healthcare System ( VACHCS ) . Ninety-four ( 13% ) were hospitalized , 91 ( 12% ) in the VACHCS . Nineteen ( 3% ) died after ZIKV infection . Hospitalization was associated with increased Charlson co-morbidity index ( adjusted odds ratio [OR] 1 . 2; 95% confidence interval [CI] , 1 . 1–1 . 3 ) , underlying connective tissue disease ( OR , 29 . 5; CI , 3 . 6–244 . 7 ) , congestive heart failure ( OR , 6; CI , 2–18 . 5 ) , dementia ( OR , 3 . 6; CI , 1 . 1–11 . 2 ) , neurologic symptom presentation ( OR , 3 . 9; CI , 1 . 7–9 . 2 ) , leukocytosis ( OR , 11 . 8; CI , 4 . 5–31 ) , thrombocytopenia ( OR , 7 . 8; CI , 3 . 3–18 . 6 ) , acute kidney injury ( OR , 28 . 9; CI , 5 . 8–145 . 1 ) , or using glucocorticoids within 30 days of testing ( OR , 13 . 3; CI 1 . 3–133 ) . Patients presenting with rash were less likely to be hospitalized ( OR , 0 . 29; CI , 0 . 13–0 . 66 ) . Risk for neurologic complications increased with hospitalization ( OR , 5 . 9; CI 2 . 9–12 . 2 ) , cerebrovascular disease ( OR 4 . 9; CI 1 . 7–14 . 4 ) , and dementia ( OR 2 . 8; CI 1 . 2–6 . 6 ) . Older Veterans with multiple comorbidities or presenting with neurologic symptoms were at increased risk for hospitalization and neurological complications after ZIKV infection .
Zika virus ( ZIKV ) is a flavivirus transmitted primarily by Aedes species mosquitoes . Since the first reported primate ZIKV infection in 1947 , sporadic human cases have occurred in Africa and Asia , followed by outbreaks in Micronesia and French Polynesia , culminating in widespread infection in the Americas in 2015–2016 [1–4] . In May 2015 , locally transmitted infection in the Western Hemisphere was first reported in Brazil; the predominant strain was related to the Asian genotype [5] . ZIKV disseminated among this largely immunologically naïve population , where the World Health Organization estimates >534 , 000 confirmed or suspect cases , involving the majority of Western Hemisphere countries by the end of 2016 [6] . As of September 2017 , 5 , 431 cases have been reported in the continental United States ( U . S . ) , of which 5 , 155 were travel-associated , 225 were locally acquired mosquito-borne cases , and 36 , 644 cases were reported in Puerto Rico and the U . S . Virgin Islands [7] . ZIKV infection is often asymptomatic and usually self-limited , with most symptoms resolving in 7–10 days [4] . Patients typically present with rash , arthralgia , conjunctivitis , or fever [1 , 3 , 4] . More serious complications include congenital syndrome ( microcephaly and other fetal abnormalities ) , Guillain-Barré syndrome ( GBS ) and other neurological disorders [8–23] . ZIKV is detectable for approximately 1 week in blood and 2 weeks in urine [24–26] . The Veterans Health Administration ( VHA ) has health care facilities throughout the U . S . and territories . We perform ongoing surveillance for emerging pathogens , and reported the first ZIKV case in Puerto Rico in December 2015 [27] . Since 46% of all U . S . Veterans and 62% of Veterans in Puerto Rico are aged ≥65 years and have significant comorbidities , they could be at higher risk for severe ZIKV infection compared to those in the general U . S . population exposed to the virus ( i . e . , returning travelers and those living in areas with local ZIKV transmission ) [28–30] . Herein , we describe characteristics of ZIKV-infected Veterans and investigate risk factors for hospitalization and neurological complications .
ZIKV testing and surveillance were conducted as part of VHA operations and public health activities . As such , the VHA Office of Research Oversight considers public health investigations as operational and not research in VHA [31] . Since only ZIKV-positive cases required reporting to public health and U . S . federal agency authorities , negative cases were not reviewed . Patient data were anonymized after data abstraction for analyses . We identified patients from all VHA facilities with ZIKV test results for specimens collected between December 1 , 2015–October 31 , 2016 , utilizing VHA national data sources . Additional case finding was performed by querying inpatient and outpatient encounter data for Zika-specific International Classification of Diseases , Clinical Modification , 10th Revision ( ICD-10-CM ) code A92 . 5 and from VHA facility communications with VA leadership . Testing for ZIKV was performed at the VHA’s Public Health Reference Laboratory ( PHRL ) , and public health , federal or commercial laboratories . Testing and confirmation of ZIKV infection in our patient population is summarized in Fig 1 . Initially , PHRL utilized a ZIKV reverse transcriptase PCR ( RT-PCR ) assay described previously [32] . Following U . S . Food and Drug Administration approval in March 2016 , the Centers for Disease Control and Prevention ( CDC ) Trioplex RT-PCR assay for ZIKV , Dengue virus ( DENV ) and chikungunya virus ( CHIKV ) ( in serum , whole blood , urine and spinal fluid ) and ZIKV MAC IgM enzyme-linked immunosorbent assay ( ELISA ) ( Anti-Zika Virus IgM Human MAC-ELISA kit , CDC ) ( for serum only ) were used according to manufacturer’s recommendations [33 , 34] . DENV ( DENV Detect , InBios , Seattle , WA ) and CHIKV ( Abcam , Cambridge , MA ) serum IgM ELISA assays were also performed per manufacturer’s recommendations . Testing methods performed by non-VA laboratories were unable to be confirmed . Samples with presumptive positive , equivocal or inconclusive ZIKV IgM results with an RT-PCR result that was either negative or not performed were sent to CDC for confirmatory testing using PRNT for DENV and ZIKV IgM [25] . Coinfection was defined as CHIKV and ZIKV-positive RT-PCR or IgM assays . In addition , patients with positive DENV RT-PCR and ZIKV RT-PCR assays would be considered coinfected . Potential cross-reaction of tests was defined as positive DENV IgM or RT-PCR and ZIKV IgM with plaque reduction neutralization test ( PRNT ) positive for DENV and ZIKV IgM results within 30 days of testing . Patients who were ZIKV RT-PCR and DENV IgM positive without PRNT were unable to be categorized as coinfected or cross-reactive . Since laboratory testing was performed based on clinician orders , not all assays were performed in all patients . In addition , CDC-recommended testing strategies changed over the course of 2016 [25 , 26 , 35] . For all patients with available ZIKV-positive diagnostic test results , we extracted demographics , clinical symptoms during acute illness and travel history from clinical notes , laboratory results ( leucocyte , lymphocyte and platelet count; creatinine , alanine [ALT] and aspartate [AST] aminotransferases ) , concomitant medications ( including 3-hydroxy-3-methylglutaryl-coenzyme ( HMG-CoA ) reductase inhibitors , antidiabetic , nonsteroidal anti-inflammatory drugs ( NSAID ) , antidementia , oral glucocorticoids , antineoplastics , antivirals ( for HIV-1 treatment ) , immunosuppressants , and intravenous immunoglobulin ) , hospitalizations , and outcomes from VHA electronic health records ( EHR ) . Comorbidities were identified by extraction of discharge and encounter ICD-10-CM codes and provider notes . Age-adjusted Charlson co-morbidity indices ( CCI ) were calculated as a measure of a patient’s health status [36] ) . Laboratory cutoffs were as follows: leukopenia ( <4 , 500 white blood cells [WBC]/μL ) ; leukocytosis ( >11 , 000 WBC/μL ) ; lymphopenia ( <1 , 000 lymphocytes/μL ) ; and thrombocytopenia ( <155 , 000 platelets/μL ) . Acute kidney injury ( AKI ) was calculated based on serum creatinine levels that were collected at presentation and the most recent serum creatinine prior to ZIKV infection ( used as baseline ) . AKI was categorized into stage I ( 1 . 5–1 . 9 times baseline ) , stage II ( 2 . 0–2 . 9 times baseline ) , and stage III ( ≥3 . 0 times baseline ) [37] . Abnormal hepatic function was determined by elevated transaminasemia 1–1 . 9 times upper limit of normal ( ULN ) ( AST 34 , ALT 40 ) , 2–2 . 9 times ULN , and ≥3 times ULN [38] . Neurologic complications were determined by review of encounter ICD-10-CM codes after ZIKV diagnosis through 3/31/2017 . Positive cases were categorized as laboratory-confirmed which was defined as a patient with detectable ZIKV RNA by RT-PCR in serum or urine or a patient with positive ZIKV IgM ELISA result and confirmatory PRNT positive for ZIKV IgM only . A presumed positive case was a patient with a positive serum ZIKV IgM and negative DENV IgM result or not tested and a PRNT result positive for ZIKV and DENV IgM . Characteristics of patients with ZIKV infection diagnosed in VA Caribbean Health Care System ( VACHCS ) were compared to elsewhere in the U . S . In addition , factors associated with 1 ) hospitalization and 2 ) timing of diagnosis ( in relation to infection ) was assessed . Patients with a positive RT-PCR result for ZIKV ( regardless of their IgM laboratory test results ) were assumed to have been diagnosed earlier during their infection than patients with only an IgM-positive result for ZIKV ( i . e . , early vs . late where the latter was used as the referent group ) . Student’s t-test and χ2 test were used to estimate associations between continuous and categorical variables , respectively . Logistic regression was used to estimate crude and adjusted odds ratios ( OR ) and 95% confidence intervals ( CI ) for factors associated with hospitalization and timing of diagnosis . For all clinical and medication-related data , “no” and “unknown” responses were combined as “no” and served as the referent group for all logistic regression models . A multi-stage backwards model building approach was used to develop a parsimonious main effects model ( S1 Text and S1 Fig ) . That is , age group ( in 10-year age categories ) , age-adjusted CCI , and the individual comorbidities used in the CCI were included in stage I . Non-significant comorbidities were removed from the model . Clinical findings , laboratory findings , and medications prescribed as an outpatient prior to ZIKV infection were added and subsequently removed ( per non-significance ) in stages II , III , and IV , respectively . Age group and age-adjusted CCI remained in the model , regardless of statistical significance until the completion of stage IV . The Kaplan-Meier log-rank test was used to estimate differences in length of stay . Non-parametric tests ( e . g . , Wilcoxon , Mann-Whitney ) were used for non-normally distributed data ( e . g . , age among those who died ) . Death among patients with laboratory-confirmed ZIKV infection was analyzed in separate age-adjusted models . An alpha of 0 . 05 was used to determine statistical significance . Analyses were performed by using SAS 9 . 4 ( SAS Institute , Inc . , Cary , North Carolina ) .
We identified 1 , 538 VHA patients with ZIKV test results during December 2015– October 2016 ( Fig 1 ) . PHRL performed 1 , 424 ( 93% ) of these tests and the remainder were performed at non-VA laboratories . Seven hundred thirty-six ( 48% ) patients were RT-PCR-positive or serum IgM presumed positive confirmed with PRNT . Of these , 585 patients were laboratory-confirmed by RT-PCR ( n = 569 ) or positive IgM with PRNT positive for ZIKV IgM only ( n = 16 ) . Per CDC guidelines and since there was a lack of active dengue cases seen by PCR or dengue specific IgM testing , the remaining 151 patients were presumed positive for ZIKV as their ZIKV IgM PRNT was positive for both ZIKV and DENV [24] . Demographic and clinical factors are summarized in Table 1; cumulatively , there were 655 ( 89% ) male patients , with the majority ( 93% ) of patients diagnosed at VACHCS , and the remaining 7% of patients diagnosed at 24 other VHA medical centers . Documented travel for those diagnosed in the continental U . S . included non-Puerto Rico Caribbean ( 18 ) , Puerto Rico ( 16 ) , Central America ( 9 ) , South America ( 2 ) , Indonesia ( 1 ) , and Senegal ( 1 ) . Six patients had exposure only in Florida . Mean age of all patients was 58 . 8 years ( range 20–99 ) . Patients from VACHCS versus returning travelers with ZIKV infection were older ( mean age 60 versus 47 years; p< 0 . 001 ) . Four hundred seventy-four of 736 ( 65% ) patients presented to an emergency department . Most common documented symptoms in patients with ZIKV infection were arthralgia/myalgia 92% , rash 90% , conjunctivitis 75% , and reported fever 66% ( Table 2 ) . Documented fever and myalgia/arthralgia or rash was reported for 378 ( 51% ) patients and subjective fever and rash for 315 ( 43% ) patients . The distribution of laboratory findings during their ZIKV illness is shown in Table 2 . Among ZIKV-positive patients , and of those who had hematology and chemistry testing performed , at their nadir , 37% had leukopenia ( median , 3 , 700 WBC/μL; range , 800–4 , 400 ) , 30% had lymphopenia ( median , 740 lymphocytes/μL; range , 0–990 ) , and 25% had thrombocytopenia ( median , 126 , 500 platelets/μL; range , 17 , 000–149 , 000 ) . Eleven percent had leukocytosis ( median , 15 , 000 WBC/μL; range , 11 , 000–39 , 200 ) . Twenty-five ( 5% ) patients had acute kidney disease and 121 ( 32% ) patients had elevated serum transaminases . Concomitant use of HMG-CoA reductase inhibitors was the most frequently observed medication class ( 232 [32%] ) , followed by antidiabetics ( 124 [17%] ) , NSAID ( 83 [11%] ) , antidementia ( 40 [5%] ) , glucocorticoids ( 11 [2%] ) , antineoplastics ( 8 [1%] ) , antivirals ( 5 [0 . 7%] ) , and immunosuppressants ( 3 [0 . 4%] ) . No patients received intravenous immunoglobulin . At VACHCS , 91 ( 12% ) of 683 patients were hospitalized with median acute care length of stay ( LOS ) of 6 days ( range 1–214 days ) , including 20 ( 3% ) who were admitted to intensive care ( ICU ) with median LOS of 4 days ( range 1–30 days ) ; at VHA hospitals elsewhere in the U . S . , 3 of 53 ( 6% ) returning travelers with known hospitalization status were hospitalized with median LOS of 4 days ( range 1–6 days ) but none were admitted to an ICU . The length of stay between these two groups was not significantly different ( p = 0 . 13 ) . The average age among hospitalized patients was higher among the 91 patients at VACHCS than the three returning travelers in the continental U . S . ( 75 vs . 60 years , p<0 . 001 ) . Crude measures of association with hospitalization are shown in Table 3 . Adjusted ORs , 95% CI and p-values controlling for all significantly associated factors are presented in Table 4 . The odds of hospitalization significantly increased with CCI , connective tissue disease , congestive heart failure , dementia , neurologic symptoms , GBS , leukocytosis , thrombocytopenia , AKI , and glucocorticoid steroid use within 30 days of ZIKV testing . Patients presenting with a rash were less likely to be hospitalized . In additional adjusted analyses reported in Table 4 , only having rash , conjunctivitis , leukopenia or lymphopenia at presentation were significantly associated with a positive RT-PCR test . Forty-six ( 6% ) patients with ZIKV infection ( 37 confirmed , 9 presumed positive ) also had neurologic complications as summarized in Table 1 . Five patients had cerebrospinal fluid ( CSF ) tested for ZIKV , all of whom had at least one of the identified neurologic complications . One patient with altered mental status , meningitis and viral encephalitis was positive for ZIKV by RT-PCR in CSF and serum . CSF findings for this patient were consistent with a viral etiology , demonstrating mild pleocytosis ( WBC 12/cm3 , 88% polymorphonuclear leukocytes , 12% lymphocytes ) and normal CSF protein level ( 30 . 2 mg/dL ) . As shown in Table 4 , neurologic complications were significantly more likely in patients with a prior history of cerebrovascular disease ( CVD ) and dementia as well as those who had been hospitalized . Of 81 women with a positive ZIKV test , 50 were of childbearing age ( 18–52 years old ) and four were pregnant at the time of infection . Two of these patients were from VACHCS and two were identified as returning travelers . Three patients delivered their babies ( further details on the outcome of the babies is unknown ) and one patient miscarried at 9 . 5 weeks . Fourteen patients were positive for DENV IgM and ZIKV RT-PCR alone , of whom one was hospitalized . These 14 were unable to be categorized as coinfection or cross-reaction as they did not have PRNT performed . Three additional patients were positive for all three viruses ( CHIKV IgM , DENV IgM and ZIKV RT-PCR ) , of whom one was hospitalized . Fifty-six patients ( 8% ) , all diagnosed at VACHCS , were positive for ZIKV ( RT-PCR [n = 43] or IgM [n = 13] ) and CHIKV IgM coinfection , of whom nine ( 16% ) were hospitalized . In adjusted analysis , age was significantly associated with coinfection and arthralgia/myalgia was significantly less common in these patients . There was no increased risk of hospitalization or neurologic complications associated with coinfection . Nineteen ( 3% ) patients died post-ZIKV infection , all of whom presented to VACHCS with ZIKV related symptoms of which 16 were hospitalized ( Table 5 ) . Fourteen of 19 had viremia at presentation ( Table 5 ) . The mean age of ZIKV patients who died was 82 years ( range , 50–99 years ) , compared to 59 years ( range , 20–98 years ) for those at VACHCS who survived ( p<0 . 001 ) . The mean time from ZIKV testing until death was 39 days ( range 3–104 days ) . Eighteen ( 95% ) had at least one CCI condition . Thus , it was difficult to determine whether ZIKV infection contributed to death or not .
Our study is the first to characterize U . S . Veterans with ZIKV infection . Testing varied based on test availability , provider preference , or presenting symptoms . The majority received a diagnosis in Puerto Rico , although 53 were returning travelers or had locally acquired infection elsewhere in the U . S . Among returning travelers , three were hospitalized , whereas in Puerto Rico , where patients were older and had more comorbidities , approximately 13% of patients with ZIKV infection were hospitalized , of whom 3% were admitted to ICU , and 3% died post ZIKV infection . Although we cannot directly link the deaths with ZIKV , the number of deaths was higher among VHA patients compared with a report from Puerto Rico in December 2016 that described only 5 deaths identified by surveillance on the island [39] . CCI was associated with increased risk for hospitalization which was possibly related to a lower threshold for hospitalization in those with significant chronic illness . After adjusting for CCI , connective tissue disease , CHF , dementia as well as presenting with neurologic symptoms , leukocytosis , thrombocytopenia , AKI or being prescribed glucocorticoids 30 days prior to ZIKV diagnosis was associated with increased risk for hospitalization . However , presenting with a rash made hospitalization less likely and no patients receiving a NSAID were hospitalized . Hospitalization and deaths are reported to be uncommon in ZIKV infection [3 , 40–44] . During the 2007 ZIKV outbreak in Micronesia , among 49 confirmed and 59 probable cases , patients presented with typical symptoms described here , but none were hospitalized and none died [3] . Although patients in that study were on average >10 years younger and fewer had comorbidities than U . S . Veterans . In Brazil , among 119 ZIKV confirmed patients only one hospitalization and no deaths were reported [41] . Hospitalizations and death ( <1% ) were noted in Puerto Rico from November 2015–July 2016 [40] . In our Veteran population , 3% died after a ZIKV diagnosis and 13% were hospitalized which is higher than other ZIKV studies and may be related to Veterans increased comorbidities [30] . Several studies have documented coinfection with ZIKV and CHIKV [45–47] . A prior study identified patients from Nicaragua with positive CHIKV and ZIKV [47] . Since cross-reaction is unlikely between these viruses , these patients were noted to have coinfection . In their study , 16/263 ( 6% ) ZIKV-positive patients were noted to have coinfection with CHIKV and ZIKV [47] . In the Nicaraguan cohort , patients with coinfection trended toward more hospitalization and had similar symptoms to those monoinfected [47] . In our study , 56/736 ( 8% ) patients were identified as being positive for ZIKV and CHIKV IgM ( with or without positive DENV IgM ) . No patients were identified with ZIKV and CHIKV or DENV by RT-PCR . In patients with coinfection , there was no increased risk in hospitalization or neurologic complications but there was an increased risk of coinfection with advanced age . Symptoms were similar between groups except there was a decrease in documented arthralgia/myalgia in coinfected patients . ZIKV has been documented to have congenital as well as neurologic complications [8–23] . Forty-six patients in our cohort were also noted to have neurologic complications after ZIKV infection . While these neurologic complications are quite broad , they identify potential complications post-ZIKV infection . Prior history of CVD and dementia as well as being hospitalized with ZIKV increased risk of neurologic complication . It was difficult to confirm whether these other neurologic complications were the result of ZIKV infection . Since neonatal and pediatric care was not provided by VA , the status of the infants exposed to ZIKV is unknown . There are several limitations to our retrospective study . Cases not tested or with results not documented in VA’s EHR could not be identified; asymptomatic and mild cases were unlikely to have testing performed; early dated cases were not tested for ZIKV IgM as it was not available at the time of clinical testing , so some cases that were RT-PCR negative may have been missed; samples from VACHCS prior to December 2015 that were tested for DENV and CHIKV were not tested for ZIKV; samples from facilities in the continental U . S . were only tested for those ordered by the provider and complete testing may not have been ordered by provider depending on timing of symptoms and possible exposure . Some health departments had strict testing criteria and submitted Veteran samples may have been rejected or not tested . Across the VA system , and specifically at VACHCS , testing was not restricted , particularly for testing performed in VA , as there was an ongoing outbreak . We were unable to determine if deaths or neurologic complications were directly related to ZIKV infection . No ICD-10-CM diagnosis code was available for ZIKV until 10/1/2016 for additional case identification purposes . Medications not obtained within the VA could not be identified . Patients receiving care outside of the VA were unable to be reviewed for neurologic complications . Only Veterans who presented to VHA facilities and had appropriate diagnostic testing completed were included . Since asymptomatic individuals were unlikely to be tested for ZIKV , overall burden of disease was unable to be determined . Given the nature of the investigation , primary focus was placed on ZIKV-positive patients . Sample sizes among certain subgroups limited inferences from statistical analyses . Veterans represent a unique group of patients who tend to have increased age and comorbidities compared to the general population [30] . Among returning travelers , many of whom presented for care in the U . S . during the convalescent period , when diagnosis is dependent upon serology , some diagnoses could have been missed as ZIKV IgM typically declines after several weeks to months [26] . Clinicians practicing in areas with ZIKV transmission should be aware that ZIKV infection among elderly patients and patients with comorbidities , including connective tissue disease , dementia and CHF , those on glucocorticoids , and those presenting with neurologic symptoms , leukocytosis , AKI , and thrombocytopenia may have more severe disease . In addition , patients hospitalized and those with prior history of CVD and dementia were more likely to have neurologic complications . Larger studies are required to determine risks associated with atypical complications , intensive care utilization and death associated with ZIKV infection; and whether prevention strategies or closer monitoring for those at greatest risk for such complications after ZIKV infection should be targeted .
|
Zika virus ( ZIKV ) infection has become an important flavivirus infection that affected over a half of a million people in the Western Hemisphere by the end of 2016 . Here we show risk factors for hospitalizations and neurologic complications in a US Veteran population . Over 700 Veterans with confirmed or presumed positive ZIKV were included . Our study showed that older Veterans with multiple comorbidities and those presenting with neurologic symptoms were more likely to be hospitalized , while if a patient presented with a rash they were less likely to be hospitalized . Neurologic complications were more likely in those hospitalized or those with a prior history of a cerebrovascular disease or dementia . Better understanding of those patients most at risk for severe disease can help providers when evaluating and treating patients with ZIKV infection .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"reverse",
"transcriptase-polymerase",
"chain",
"reaction",
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"pathogens",
"microbiology",
"cancers",
"and",
"neoplasms",
"alphaviruses",
"viruses",
"organisms",
"oncology",
"hematologic",
"cancers",
"and",
"related",
"disorders",
"chikungunya",
"virus",
"rna",
"viruses",
"nosocomial",
"infections",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"myeloproliferative",
"disorders",
"infectious",
"diseases",
"leukocytosis",
"artificial",
"gene",
"amplification",
"and",
"extension",
"medical",
"microbiology",
"microbial",
"pathogens",
"dementia",
"mental",
"health",
"and",
"psychiatry",
"molecular",
"biology",
"hematology",
"flaviviruses",
"viral",
"pathogens",
"neurology",
"co-infections",
"biology",
"and",
"life",
"sciences",
"polymerase",
"chain",
"reaction",
"zika",
"virus"
] |
2018
|
Zika virus infection in the Veterans Health Administration (VHA), 2015-2016
|
Lytic or lysogenic infections by bacteriophages drive the evolution of enteric bacteria . Enterohemorrhagic Escherichia coli ( EHEC ) have recently emerged as a significant zoonotic infection of humans with the main serotypes carried by ruminants . Typical EHEC strains are defined by the expression of a type III secretion ( T3S ) system , the production of Shiga toxins ( Stx ) and association with specific clinical symptoms . The genes for Stx are present on lambdoid bacteriophages integrated into the E . coli genome . Phage type ( PT ) 21/28 is the most prevalent strain type linked with human EHEC infections in the United Kingdom and is more likely to be associated with cattle shedding high levels of the organism than PT32 strains . In this study we have demonstrated that the majority ( 90% ) of PT 21/28 strains contain both Stx2 and Stx2c phages , irrespective of source . This is in contrast to PT 32 strains for which only a minority of strains contain both Stx2 and 2c phages ( 28% ) . PT21/28 strains had a lower median level of T3S compared to PT32 strains and so the relationship between Stx phage lysogeny and T3S was investigated . Deletion of Stx2 phages from EHEC strains increased the level of T3S whereas lysogeny decreased T3S . This regulation was confirmed in an E . coli K12 background transduced with a marked Stx2 phage followed by measurement of a T3S reporter controlled by induced levels of the LEE-encoded regulator ( Ler ) . The presence of an integrated Stx2 phage was shown to repress Ler induction of LEE1 and this regulation involved the CII phage regulator . This repression could be relieved by ectopic expression of a cognate CI regulator . A model is proposed in which Stx2-encoding bacteriophages regulate T3S to co-ordinate epithelial cell colonisation that is promoted by Stx and secreted effector proteins .
Bacteria such as Escherichia coli that colonise the mammalian gastrointestinal tract are exposed to high levels of bacterial viruses , bacteriophages [1] . Many bacteriophages inject their genetic material and use the bacterial host simply to produce more phage in a lytic cycle . Other , temperate phages , can insert their genetic material into the bacterial genome [1]–[4] . In this lysogenic state the phage genome is amplified along with the dividing bacteria [4] . Lysogenic phages have evolved to initiate their replication and escape from the bacterial host by using the stress or SOS response of the bacterium [5] . The integration of phage genomes and their subsequent degeneration is important in the evolution of many bacterial genera , for example Salmonella , Staphylococcus and Escherichia [6] , [7] . Typical enterohemorrhagic E . coli ( EHEC ) are defined by the presence of Shiga toxin ( Stx ) -encoding lambdoid-like bacteriophages in the chromosome and the ability to form attaching and effacing lesions on epithelial cells of the gastrointestinal tract due to the activity of a type III secretion ( T3S ) system [8]–[11] . This secretion system is expressed from the locus of enterocyte effacement ( LEE ) and is controlled by multiple regulatory inputs , many through the LEE-encoded regulator ( Ler ) . The T3S system translocates a cocktail of effector proteins into eukaryotic cells , many of which were introduced originally into the EHEC chromosome encoded on prophages along with regulators that co-ordinate effector expression with the T3S and may allow these non-LEE encoded effectors to compete for translocation [12]–[15] . While many of these prophages are considered cryptic their presence in strains with active prophages generates the constant possibility of new bacteriophage generation and transduction [16] . The Stx bacteriophages also introduce further variation into the strains , including the expression of different variants of Stx [17] . All Stx types belong to a family of A1B5 exotoxins comprising of a single A subunit that is non-covalently associated with pentameric B subunits that are responsible for toxin binding to its receptor , the glycosphingolipid Gb3 ( globotriaosylceramide; CD77 ) [17]–[19] . The A subunit of Stx is an N-glycosidase which cleaves the N-glycosidic bond of a specific adenine residue of the 28S rRNA in the 60S ribosomal subunit inhibiting protein synthesis [17] , [20] . Severe EHEC infections in humans are characterised by bloody diarrhoea and capillary damage in the kidneys and brain as a result of Stx activity with potential fatal consequences or long-term morbidity [1] , [21] , [22] . Humans are generally considered to be an incidental host for the majority of EHEC strains and it is apparent that ruminants [23] , [24] , in particular cattle are the main reservoir for the EHEC serotypes associated with human infection , in particular EHEC O157:H7 and EHEC O26:H11 [25] , [26] . The pathogenesis of the organism therefore needs to be considered in the context of the ruminant host in terms of selective factors that drive its evolution . In contrast to the situation in humans , there is little evidence for pathology associated with EHEC infection in mature , immuno-competent cattle or other reservoir hosts which typically carry EHEC asymptomatically [27] , [28] . This is related , in part , to differences in Gb3 receptor distribution in cattle versus humans and differences in how internalised toxin is trafficked in the cell [29]–[31] . If phage insertion can confer an advantage to the bacterium then this will also increase the survival chances of the integrated phage DNA . Studies on Shiga toxin activity have provided some insight into its possible benefits for EHEC in the ruminant host , even though a subset of the bacteria must lyse to release the toxin . Advantages for colonisation include the redistribution of nucleolin to the epithelial cell surface where it aids bacterial attachment via interaction with the bacterial outer membrane protein intimin [32] . Stx has also been shown to have immuno-modulatory functions , including the suppression of inflammatory responses and repression of B cell/T cell proliferation [33] , [34] , which could complement T3S effector repression of innate responses [35]–[39] . Epidemiological studies have shown that particular strain types emerge that are more likely to be associated with human disease . In the USA , Clade 8 is associated with more serious human disease [40] and in Europe particular phage types are more commonly associated with human infections [41] . In a previous study , Chase-Topping et al [42] investigated fecal pat samples from 481 farms throughout Scotland for the presence of E . coli O157:H7 . Three main phage types were identified , 21/28 ( 46% ) , 32 ( 19% ) , and 8 ( 12% ) . Previous research had shown that cattle colonized at the terminal rectum can shed bacteria at high levels leading to the concept of ‘super-shedders’ [43] , [44] . Low or high-level shedding was defined by whether the E . coli O157 counts were below or above 103 CFU g−1 feces respectively . PT21/28 strains were more likely to be associated with high shedding and PT32 with low-level shedding samples . PT21/28 strains are the predominant type associated with human infection in the UK [44] . The initial aim of this study was to identify genetic differences between PT21/28 and PT32 strains . The research indicated that Stx bacteriophage carriage was different between the two as was the regulation of T3S . This led us to determine the impact of Stx phage lysogeny on T3S regulation in EHEC with the conclusion that Stx2-encoding bacteriophages usurp control of this essential colonization factor .
In order to investigate genomic differences between Phage types 21/28 and 32 , twelve strains ( six of each type ) were analysed by comparative genomic hybridization ( CGH ) as detailed in Materials and Methods . These strains were chosen to include strains originating from fecal pats with both high and low bacterial counts ( table S1 ) . While multiple differences in gene content were identified between the phage types ( GEO accession number GSE28838 ) the most consistent difference was in hybridization to the Stx2 bacteriophage immunity region . Only one PT32 strain from the six analyzed showed hybridization to this immunity region in contrast to all six PT21/28 strains that were investigated ( figure 1A ) . The array design is based on the E . coli K-12 , EDL933 and Sakai sequences ( figure 1A ) . While EDL933 and Sakai contain both Stx1- and Stx2-encoding prophages they do not contain an Stx2c-encoding prophage . Previous characterization of the strain collection by the Scottish E . coli reference laboratory had determined that they all contained at least one type of Stx2 phage ( table S1 ) , so to investigate whether Stx2 vs Stx2c prophages could account for the hybridization differences , genomic DNA preparations from strains ( n = 62 ) were analyzed for these prophages by PCR ( Materials and Methods ) . While the majority of all the strains analyzed contained the Stx2c prophage ( 87% ) , the main difference was in the distribution of Stx2 prophages . Overall , examining both human and bovine-derived strains ( table S1 ) , 27/30 PT21/28 strains contained both an Stx2 and Stx2c-encoding prophage , compared with 9/32 of the PT32 strains analysed ( P<0 . 01 ) . This difference was still significant when examining bovine strains only ( P<0 . 01 ) ( table S1 ) . It was interesting to note however that PT32 strains from humans were more likely ( p = 0 . 01 ) to contain a Stx2-encoding prophage than the PT32 bovine isolates ( 75% versus 25% ) , while the majority of 21/28 strains from either source contained a Stx2-encoding prophage ( table S1 ) . So overall , PT21/28 strains , which are the main human disease associated phage type in the United Kingdom , generally harbor both Stx2 and Stx2c phages whereas PT32 strains were more likely to contain only a single Stx2 or Stx2c-encoding prophage , but with some further selection for Stx2-encoding prophages in the human-derived isolates . The secretion profiles of thirty strains of each phage type were analysed , including western blotting for the translocon protein EspD ( a subset of strains is shown in figure 1B ) . From these initial profiles it was evident that more PT32 strains were secreting higher levels of EspD . The strains also produced variable levels of EscJ , with higher levels of the T3S apparatus protein EscJ correlating with higher levels of EspD secretion ( figure 1B ) . In order to quantify any differences in the expression of the T3S system in the two phage types , a subset of the strains ( n = 31 ) were transformed with a LEE1::GFP reporter construct that measures transcription initiation of the first LEE operon including ler that encodes a key regulator of the T3S system [45] . Fluorescence levels were determined for the different strains during growth curves in MEM-Hepes as defined in the Materials and Methods . For statistical analyses , fluorescence values were determined from multiple experiments for OD600 = 1 . 0 ( table S1 ) . PT32 strains had a significantly higher median level of expression compared to the PT21/28 strains ( p<0 . 001 , figure 1C ) . As an alternative way of analysing potential differences in LEE1 expression between the PTs , the number of strains exhibiting expression levels above or below the median value of all the readings ( 13 , 846 ) was compared . On this basis , PT32 strains were more likely to express the LEE1 fusion at higher levels than the PT21/28 strains ( p<0 . 01 ) , consistent with the secretion profiles and Western blotting data . There was no association between T3S expression and single point shedding levels of the strain obtained from the fecal pat at the time of sampling ( table S1 ) . Based on the observation that PT 21/28 was more likely to contain two types of Stx2-encoding prophage ( Stx2 and Stx2c ) and generally demonstrated lower levels of T3S , LEE1 expression levels were compared with the presence of Stx2 and Stx2c prophages . Strains containing both Stx2 and Stx2c prophages were significantly more likely ( p<0 . 05 ) to have lower LEE1 expression levels than strains containing just one Stx2 prophage ( figure 1D ) . The differences in T3S and LEE1 expression appeared consistent with differences in Stx2 and Stx2c prophage content , with more repression of T3S evident in strains containing both Stx2 and Stx2c prophages . As only one of the PT21/28 and PT32 strains analysed contained a Stx1 phage ( table S1 ) it was not possible from this data to examine any impact of Stx1 prophage integration on T3S . In order to investigate whether the expression of the T3S system is repressed by the presence of Stx prophages , secretion profiles and LEE1 expression levels in EHEC strains , with and without integrated Stx prophages , were analyzed . The sequenced isolate EDL933 was compared with a published Stx1 and Stx2 prophage-cured derivative TUV93-0 ( figure 2A ) . Higher levels of EspD secretion were detected in bacterial supernatants of TUV93-0 compared with EDL933 ( figure 2A ) . This correlated with higher levels of the apparatus protein , EscJ , in the whole cell fractions compared with RecA as a control ( figure 2A ) . Again , to quantify this difference , the LEE1 promoter fusion was introduced into both strains . The expression in TUV93-0 was higher than EDL933 throughout the growth curve with a significance difference for values determined at OD600 = 0 . 9 ( figure 2B–C ) . As both the EDL933 and TUV93-0 strains used have been cultured in a number of different laboratories and EDL933 was also subject to genetic manipulation during which phage deletion occurred [46] , there may be other genetic changes , in addition to the absence of the Stx phages , that may account for the increased level of T3S in TUV93-0 . To investigate this , two approaches were taken . In the first , a Stx2 bacteriophage ( Sp5 ) marked with a kanamycin resistance cassette from a derivative E . coli O157 Sakai strain ( Sakai stx2A::kan , table 1 ) was moved in the TUV93-0 background . Attempts to transduce this marked Sp5 phage into this background were not successful . However , conjugation of the prophage into TUV93-0 worked in the presence of a cloned and induced copy of the Sp5 cI ( pBAD-CI , table S2 ) to restrict prophage induction . EspD secretion level , EscJ expression and LEE1-GFP levels were then determined in the absence of arabinose-induced cI induction ( figure 2A–B ) . The conjugation led to the repression of T3S but not to the level demonstrated for EDL933 . This repression was confirmed by analysis of both ler and espD transcript levels in the TUV strain pair , with TUV93-0 ( Sp5 ) showing a significant reduction in these T3S-associated transcripts in comparison to the parent strain ( TUV93-0 ) , while there was no significant change in gapA transcript levels between the two strains ( figure S1 ) . As a second approach , another series of strains were generated from EDL933 in which the Stx2-encoding prophage and then the Stx1-encoding prophage were deleted by allelic exchange to generate another Stx prophage negative derivative of EDL933 . This strain exhibited increased levels of T3S and LEE1 expression by comparison with the isogenic EDL933 parent , but at lower levels than shown for TUV93-0 ( figure 2A–C ) . Of note is that deletion of just the Stx2-encoding prophage resulted in a significant increase in T3S and LEE1 expression levels but that the further deletion of the Stx1-encoding prophage had no further impact on expression , indicating that the repression is associated with the Stx2-encoding prophage only in EDL933 . The impact of Stx phage deletion was further confirmed by RT-PCR ( figure S1 ) , with significantly higher espD transcript levels associated with either the single or double prophage deletions . Taken together , it is evident that the presence of Stx2 prophages in the EHEC O157 chromosome can limit T3S by repressing levels of LEE1 transcription . However , other uncharacterized changes between our laboratory stocks of EDL933 and TUV93-0 may also contribute to T3S differences between these strains . To examine if Stx2 prophage deletion has an effect on T3S in another EHEC serotype , two pairs of published E . coli O26:H11 strains were obtained ( table 1 ) . Both pairs are considered isogenic apart from the presence and absence of Stx2 prophage [47] , the prophage were lost during culture of the bacteria . For one pair , T3S increased significantly in the absence of Stx2-encoding prophage ( figure 2A ) , while for the other no T3S was detectable for either strain under the conditions tested ( data not shown ) . For the pair with detectable T3S , the LEE1-GFP reporter was transformed into both strains and levels of expression measured throughout the growth curve ( figure 2D ) . Again the presence of the Stx2 prophage correlated with significant repression of LEE1 expression in agreement with the Western blotting profile for EspD secretion and production of the T3S apparatus protein , EscJ ( figure 2A–C ) . In order to determine whether the repression of T3S expression by Stx2 prophages could also be detected in a distinct genetic background , the marked Stx2 bacteriophage ( Sp5 ) induced from E . coli O157:H7 Sakai was used to transduce E . coli K12 MG1655 . This transduction was achieved and the establishment of a lysogen in E . coli K-12 confirmed by selectable markers and PCR assays ( Materials and Methods ) . This E . coli K-12 Stx2 prophage lysogen and the parental MG1655 strain were then transformed with the LEE1::GFP promoter fusion plasmid and fluorescence levels measured throughout the growth curve ( figure 3A ) . While there was some detectable expression from the LEE1 promoter there was no significant difference in LEE1 expression levels between the two strains ( figure 3A ) . However , neither of these strains contains the LEE encoded regulator ( Ler ) which is known to activate T3S by relieving H-NS repression of LEE1 and other LEE promoters . To examine the effect of the Stx2 prophage on Ler-induction of LEE1 expression , ler was expressed from an IPTG-inducible promoter on pWSK29 in the same pair of E . coli K-12 strains containing the LEE1 reporter . Induction of Ler was carried out using IPTG and fluorescence levels measured throughout the growth curve . As anticipated expressing Ler in trans significantly increased expression from the LEE1 promoter fusion in the E . coli K-12 strain ( figure 3A ) . However , this ler-dependent activation did not occur to the same extent in the E . coli K-12 derivative containing the Stx2 prophage ( figure 3A ) . This result implied that Stx2 prophage-based repression of T3S involved inhibition of Ler-mediated LEE1 promoter activation . To confirm that Ler was induced by IPTG induction to a similar level in the two backgrounds , a modified Ler was expressed from pWSK29 with a carboxy-terminal six Histidine tag ( 6×His-Ler ) to allow levels to be determined by Western blotting . This construct was also able to induce LEE1 expression in E . coli MG1655 and again this activation of transcription from the LEE1 promoter was repressed by the presence of the marked Stx2 prophage ( Sp5 ) from Sakai ( figure 3B ) . Whole cell protein samples were blotted for detection of 6×His-Ler and the isogenic K-12 strains containing the LEE1-GFP reporter ( figure 3E , lanes 6 & 8 ) contained equivalent levels of the tagged Ler protein when compared to RecA levels as a control . This implied that although Ler was induced and present its capacity to act on the LEE1 promoter was limited by the presence of the integrated Stx2 phage in the chromosome . One explanation for the failure of Ler to induce LEE1 expression in the presence of the Stx2 lysogen is that a regulator or regulators expressed from the Stx2 lysogen repress Ler activity on the LEE1 promoter . Several of the key lysis/lysogeny regulators characterized for phage lambda were cloned from the Sakai Stx2 phage ( Sp5 ) into the arabinose-inducible vector pBAD myc HisA ( table S2 ) . These clones were sequenced and confirmed as identical to the published genes from Sp5 . The clones were electroporated into the Stx phage negative E . coli O157:H7 strain TUV93-0 and the T3S profiles determined from supernatants following arabinose induction ( figure 4a ) . Only the clone of cII showed any evidence of T3S repression . However , it is established that CII induction can be toxic for cells and we observed reduced growth rates when inducing expression of cII . It is also apparent that in this case , even though EspD levels were reduced , EscJ expression was not , indicating a difference over the Stx2 prophage repression phenotype . Of note was the same reduced growth rate when cII was induced in EPEC O127 E2348/69 , but in this case there was no repression of T3S ( data not shown ) , indicating a regulatory difference in this Stx phage negative background . To determine the relevance of cII to repression of T3S at physiological levels , the cII gene was deleted sequentially from the Stx2 prophage and then from the Stx1 prophage in E . coli O157 EDL933 . Western blotting for T3S proteins as well as fluorescence levels from the LEE1-GFP reporter transformed into the strain demonstrated significantly increased LEE1 expression and T3S in the absence of cII from the Stx2 prophage ( figure 4A–B ) , but with no further increase in secretion following the deletion of the Stx1-prophage cII regulator . This result is consistent with the results obtained following sequential deletion of the Stx2 and Stx1 prophages from EDL933 ( figure 2 ) . Available database cII sequences from Stx1- and Stx2-encoding bacteriophages were aligned and clustered ( figure 4E ) . This indicated that while cII sequences can group to some extent with Stx type , identical cII sequences can also be associated with both Stx1 and Stx2-encoding phages . In the case of EDL933 , the BP933W ( Stx2 ) -associated CII peptide ( Z1449 ) shares 94 . 6% amino acid identity to the predicted CII peptide ( Z3357 ) sequence from CP-933V ( Stx1 ) which could account for the differences in activity observed in this study . Both EDL933 and Sakai also encode an additional gene within syntenic , non-stx associated lambdoid phages ( z0310 in EDL933 CP-933H prophage and ecs0276 in Sakai Sp1 prophage , respectively ) , with low predicted identity to the Stx-associated CII homologs ( <30% amino acid identity with the EDL933 Stx2 prophage CII peptide sequence ) . Table S4 summarizes the results of these comparisons . In the case of classic lambda , CII and the majority of prophage regulators are silenced by CI once lysogeny has been established [4] , [5] . Given that CII and potentially other regulators on Stx2/2c prophages are involved in the repression of T3S when the prophage is in a lysogenic state , we wanted to determine whether increased expression of a homologous cI could alleviate this prophage-based control by repressing the expression of these controlling factors such as cII . This was tested in a number of ways: Firstly , when the marked Sakai Sp5 Stx2-prophage was conjugated into TUV93-0 ( figure 2 ) this required that cI from this phage was cloned , transformed and expressed in TUV93-0 in order to prevent zygotic induction of phage-mediated lysis on receipt of the transferred prophage . It had already been demonstrated that this clone had no impact on T3S in TUV93-0 in the absence of a Stx2 prophage ( figure 4 ) , but when this cI-encoding plasmid was subsequently displaced from the TUV93-0 ( Sp5 ) background , there was further suppression of T3S as determined by a reduction in EspD secretion levels ( figure 5A ) . The presence of the CI-encoding plasmid was therefore limiting the capacity of the bacteriophage to repress T3S . Secondly , the same Sakai ( Sp5 ) cI construct was then transformed into both E . coli O157 Sakai and EDL933 backgrounds . While there was no detectable impact of the regulator on T3S in EDL933 ( data not shown ) , the expressed regulator increased levels of T3S in the Sakai background ( figure 5B ) . This was consistent with the fact that the Sakai Sp5 produces a CI regulator that is markedly different ( 52/217 identity , 24% ) to that encoded by the Stx2 prophage ( BP-933W ) of EDL933 . Finally , in order to test whether the prophage-based regulation contributed to repression of T3S in the original PT21/28 isolates that contain both Stx2 and Stx2c prophages ( figure 1 ) , the Sakai Sp5 cI clone was transformed into the PT21/28 strain 15602 ( table S1 ) . Sequencing of amplified cI alleles from this strain ( Materials and Methods ) indicated that it contains an identical cI to that present in Sakai Sp5 . The presence of the Sp5 cI clone and further induction with arabinose increased EspD secretion in this PT21/28 strain ( figure 5C ) , supporting the concept that increased expression of a homologous cI can counteract the activity of Stx2 prophage-based regulators , including cII , that repress T3S .
Enterohemorrhagic Escherichia coli strains are characterised by the integration of lambdoid-like phages into their chromosome that encode Shiga toxins [1] , [8] , [47] , [48] . EHEC O157 strains can be lysogenized by different types of Stx phage , but those associated with Stx2 and Stx2c are considered the most important in terms of human virulence [49] , [50] . Some strains carry multiple Stx phages indicating differences in phage exclusion control beyond that characterized for classical lambda [51] , [52] . A key question is whether Stx phage lysogeny confers an evolutionary advantage on the recipient EHEC strain , and whether this might underpin the epidemiology of EHEC O157 strains ? A large epidemiological study of farms in Scotland demonstrated that PT21/28 strains were more likely to be associated with higher level fecal counts compared with PT32 strains . In addition , PT21/28 strains over that period were the most prevalent phage type associated with human infection [41] , [42] . The initial aim of this study was to compare PT21/28 and PT32 strains in terms of genomic differences to determine how these inform the epidemiology . Comparative genomic hybridisation ( CGH ) and PCR analyses demonstrated that Stx phage carriage differed between the phage types . PT21/28 strains generally carried both Stx2 and Stx2c bacteriophages , irrespective of whether the strains were sourced from cattle or humans . By contrast , nearly all the bovine PT32 strains analysed carried only the Stx2c phage . The situation for the limited number of human PT32 strains was more complex , but these were significantly more likely to contain a Stx2 phage in comparison to the bovine PT32 strains . This data indicates that strains carrying both Stx2 and Stx2c phages are more likely to be associated with human infection . This result is in line with recently published research on Clade 8 strains in the USA that are considered to be more virulent . Clade 8 strains were significantly more likely to carry both the stx2 and stx2c genes in comparison to the other clades containing stx2c [40] . All clade 8 strains tested had stx2 , and 57 . 6% had stx2c . Whether this association for both relates to differences in toxicity of the strains or other phenotypes associated with the strains is not known , although more recent work has indicated that levels of Stx2 production alone are likely to be important , irrespective of lineage , in terms of association with human clinical disease [49] . As type III secretion ( T3S ) is essential for cattle colonisation and variation in regulation may be associated with differences in shedding level from cattle [44] , T3S levels of PT32 and PT21/28 strains was investigated . Levels of T3S varied markedly between strains in agreement with our previous work on T3S profiles of EHEC O157 strains [53] . T3S levels were low in the majority of PT21/28 strains compared with the PT32 strains . Extensive research has defined multiple regulatory inputs for T3S and it is likely that the variation observed will be due to a complex combination of alleles . For example , use of a LEE1 reporter readout in different lineages of EHEC O157 strains has yielded a complex and widely distributed pattern of expression with no clear association with particular integrated O-islands or S-loops [54] . However , the finding in the present study that strains with both Stx2 and Stx phages had on average lower levels of LEE1 expression than strains with a single type of Stx2 bacteriophage prompted us to test strain pairs with and without Stx bacteriophages . Analysis of these strains demonstrated that Stx2 phage insertion into the chromosome leads to repression of T3S , this was also apparent in a pair of O26 strains , including one from which the Stx2 phage had been naturally cured [47] . As stated there are likely to be multiple variable inputs into T3S expression and this was confirmed by analysis of the potentially isogenic EDL933 and TUV93-0 strains . While Stx2 bacteriophage insertion and deletion repressed and activated T3S respectively , other differences between the strains may also account for the marked difference in T3S levels between these strains . To confirm and study the Stx2 bacteriophage repression in another background , the sequenced E . coli K12 MG1655 strain was lysogenised with a marked Stx2 phage from the sequenced E . coli O157:H7 Sakai strain . This set of experiments provided insight into the mechanism of the Stx2 prophage control as the capacity of Ler to induce LEE1 expression was reduced in the presence of the Stx2 lysogen despite equivalent levels of the Ler protein being present . In turn , this will limit expression of the other LEE operons under Ler control reducing T3S levels from the EHEC lysogen . As LEE1 expression is subject to auto-regulation by Ler , we cannot rule out repression of LEE1 transcription independently of the inhibition of Ler activation . Of note is recently published work examining the impact of Stx phage insertion on gene expression in E . coli K-12 by micro-array [55] . There were 166 genes found to be differentially expressed indicating global regulation by the integrating prophage . 62 transcripts were down-regulated and 104 up-regulated , including motility and acid-resistance associated genes . This included increased expression of gadE that is known to be involved in repression of T3S so increased gadE expression following Stx phage lysogeny could account for some of the reduced expression [14] , [56] , [57] . In the present study , this was investigated in E . coli K-12 , in which Stx2 lysogeny was demonstrated to increase gadE expression ( figure S2A ) , but in an E . coli O157 background , the deletion of gadE had no impact on the capacity of the integrated Stx2 prophage to repress T3S ( figure S2B ) , indicating that while this pathway may be relevant for the prophage control , it is not required . Previous work has established that the LEE2/3 operons , but not LEE1 are subject to repression by LexA in EPEC with activation following SOS induction [58] . The repression of Ler activity on LEE1 demonstrated in the present study may be the result of established lambda lysogeny regulators and/or regulators encoded elsewhere in the bacteriophages . To test established lambda phage regulators , cI , cII , cro , anti-terminators N and Q were cloned and induced in TUV93-0 . CI was a potential regulator as this is the main phage encoded-protein known to be expressed constitutively during lysogeny [52] but its induction had no direct impact on T3S in the absence of an integrated Stx prophage . CII was the only induced regulator that had a direct effect in this background , and de-regulated CII expression is known to have effects on growth rate [59] and so the T3S changes may have been due to indirect effects . For example , ClpX activity is increased in stressed bacteria and this is known act predominately on LEE4/5 and so would affect T3S levels of the LEE4-encoded EspD but not EscJ expressed from LEE3 [60] . To test the potential importance of cII , both copies were sequentially deleted from the Stx2- and then Stx1-encoding prophages in EDL933 . T3S was increased following deletion of cIIStx2 but no further increase was measured on further deletion of cIIStx1 , indicating a role of cIIStx2 in the control of T3S regulation . This result is however incongruous with the established regulation and function of CII , which is thought to primarily aid the establishment of the lambda lysogen [52] . While this is the case for bacteriophage lambda there may well be differences in the regulation between classic lambda and Stx lambdoid bacteriophages; for example it is already clear that these phages are tolerant of multiple lysogens [51] . Further confirmation that the repression of T3S is controlled by Stx2 phage-based regulators such as CII that are under the control of CI was shown by increasing expression of a cognate cI in a number of EHEC backgrounds which led to an increase in T3S levels ( Figure 5A–C ) . We cannot rule out that other Stx2 bacteriophage-based regulators may also contribute to repression of T3S following lysogeny but both cI and cII sequence variation , as well as differences in their expression and potential to cross-talk with other Stx-encoding prophages are likely to account for the different repressive capacities of Stx-encoding phages and T3S variation between EHEC strains . So why is this repression of T3S associated with Stx phage lysogeny ? We propose that Stx phages repress LEE1 expression to effectively take control of this critical colonisation factor . The Stx bacteriophage , through this regulation , may ensure that bacterial colonisation and persistence become dependent on the bacteriophage . It is interesting that EHEC strains , by comparison with the sequenced EPEC strain O127:H6 have a greater number of T3S effector proteins expressed from a variety of cryptic prophages [8] , [61] . These effector proteins have their expression co-ordinated with production of the T3S system by the PchA/B regulators and Ler that act on both effector protein expression and LEE1 [12] , [13] . We propose that the repression demonstrated by Stx2 phages could select for co-integration of effector-encoding prophages and their associated LEE regulators ( figure 6A–C ) . As a consequence , it is the network of prophage-based repressors and activators controlling T3S that is important for EHEC strains for which Stx2 prophages must also be taken into consideration [13] . It is also evident that Stx phages provide a selective advantage in EHEC strains for colonisation and persistence through the production of Shiga toxins . SOS induction in the gastrointestinal tract will lead to lysis of a subset of the bacteria with Stx release ( figure 6D ) . Stx allows increased expression and relocation of nucleolin and other receptors to the epithelial cell surface where they bind to the EHEC adhesin , intimin promoting EHEC colonization [32] , [62] . This function of Stx directly links Stx phage regulation and LEE-promoted adherence , as intimin is expressed from the LEE and is necessary for intimate attachment and attaching and effacing ( A/E ) lesion formation by binding to the translocated intimin receptor ( Tir ) also expressed from the LEE and secreted by the T3S system . It is established that EHEC strains vary markedly in their capacity to produce Stx toxins [49] , a phenotype that is intricately associated with Stx phage carriage and levels of lysis induction . EHEC strains also vary in their capacity to secrete T3 translocon and effector proteins [45] , [53] a phenotype that can now also be related , in part , to the combination of Stx phages carried by a strain . The heterogeneous nature of lysis induction means there is a subset of bacteria that do not lyse and which can then induce T3S when appropriate niche signals are detected , including quorum sensing [63]; these bacteria benefit from Stx priming of the epithelium as a result of the lysed population ( figure 6D ) . In this way , the Stx phage and effector prophage repertoire of an EHEC strain are critical factors governing colonization of the gastrointestinal tract and subsequent innate responses that will determine excretion levels from the animal ( figure 6D ) . Future work will examine whether increased excretion of strains from cattle is associated with integration of multiple Stx phages , altered T3S regulation and increased Stx toxin levels , with important ramifications for the selection of strains able to cause more severe infections in humans .
The bacterial strains and plasmids used in the study are described in tables 1 , S1 and S2 . Table S3 lists the oligonucleotide primers used . MEM-HEPES is minimal essential medium with HEPES buffer ( Sigma ) , containing additional glucose to a final concentration of 0 . 2% . LB broth was also used ( Oxoid ) . Antibiotics were included when required at the following concentrations: chloramphenicol ( C ) 12 . 5 µg/ml , Nalidixic acid ( 20 µg/ml ) kanamycin ( K ) 25 µg/ml and ampicillin ( A ) 50 µg/ml , spectinomycin ( 50 µg/ml ) and tetracycline ( 15 µg/ml ) . ler , cII alleles , cI , cro , anti-terminator N and Q genes were cloned from E . coli O157 Sakai and EDL933 strains as explained in table S2 with primer sequences described in table S3 . These clones were confirmed by sequencing ( GATC Biotech ) . The cI alleles from 15602 were sequenced following amplification using CIseq1 and CIseq2 primers ( table S3 ) . Clones in pBAD/HisA were induced with 0 . 2% ( w/v ) L-arabinose when indicated . Genomic DNA was extracted with Invitrogen ChargeSwitch gDNA Mini Bacteria Kit ( Invitrogen ) . Labeling was carried out with Bioprime Plus Array CGH Genomic Labelling System ( Invitrogen ) . Protocols for pre-hybridization , hybridization and washing for ultragaps slides were from UBEC project ( School of Biosciences , University of Birmingham ) . The arrays were 70-mer spotted oligonucleotides based on open reading frames from E . coli K-12 , E . coli Sakai , and E . coli EDL933 . Briefly , array slides were washed twice in Wash Buffer II ( 0 . 1% SDS , 0 . 1× SSC ) , each time for 30S . Then slides were transferred in pre-hybridization buffer ( 0 . 1% BSA , 0 . 1% SDS , 5× SSC ) for 120 min . After washing in Wash Buffer II and III ( 0 . 1× SSC ) , the denatured hybridization probe mixture ( 30% formamide , 5× SSC , 0 . 1% SDS , 0 . 1 mg/ml Salmon sperm DNA , 1× Denhardt's Solution and 80 picomoles Cy3 and Cy5 probe ) was added to slides and incubated overnight . Next day , the array slides were washed in Wash Buffer I ( 0 . 1% SDS , 2× SSC ) , II and III and scanned with an Axon scanner . The hybridization levels were analyzed using Genespring GX 7 . 3 . 1 ( Agilent ) . Genomic DNA for 62 tested strains was extracted with the Invitrogen ChargeSwitch gDNA Mini Bacteria Kit . The concentration of the genomic DNA was measured by a nanodrop . The method of Wang et al . [64] was used to determine the presence of stx2 and stx2c coding sequences . 100 ng of genomic DNA was used for each PCR reaction . An amplicon of 115 bp with the respective primers indicated the presence of an Stx2-encoding bacteriophage while an amplicon of 124 bp with the stx2c-specific primers indicated the presence of an Stx2c-encoding prophage . Bacteria were cultured in 30 ml of MEM-HEPES at 37°C ( 200 r . p . m . ) to an OD600 of 0 . 8 unless specifically stated . The bacterial cells were pelleted by centrifugation at 4000× g for 15 min , and supernatants were passed through filters ( 0 . 45 µm ) . Supernatant proteins were precipitated overnight with 10% TCA , and separated by centrifugation at 4000× g for 30 min ( 4°C ) ; the proteins were suspended in 100 µl of 1 . 5 M Tris ( pH 8 . 8 ) . The bacterial pellet was initially suspended in 50 µl Laemmli sample buffer ( Sigma ) and 50 µl water . Proteins were separated by SDS-PAGE using standard methods and Western blotting performed as described previously [45] , [65] . Primary antibodies for Western blotting were raised against the proteins described with the source provided in brackets: α-EspD ( Gift from Prof . T . Chakraborty , University of Giessen ) , α-EscJ ( Gift from Dr Ando and Prof . Tobe , Osaka University ) , α-RecA ( Stressgen/Enzo Life Sciences ) . Strains were electro-transformed with pAJR70–71 ( table S2 ) and the resultant transformants were cultured overnight in LB broth and then diluted 1∶100 into fresh , MEM-HEPES with appropriate antibiotics . Typically , 30 ml was cultured in Erlenmeyer flasks shaken at 200 rpm , 37°C . The optical density of the cultures was monitored by determination of the OD600 . Cultures were sampled every hour for optical density and GFP measurement . The total fluorescence produced by the population was determined by analyzing 100 µl aliquots of culture with a fluorescent plate reader ( Fluostar Optima; BMG ) . Promoterless plasmid pAJR70 acted as a control for background fluorescence . Bacterial RNA was purified with an RNeasy kit ( Qiagen ) and reverse transcribed with random primers Affinityscript ( Stratagene ) . qPCR was carried out with a PowerSybr mastermix ( Applied Biosystems ) . The qPCR primers used are listed in table S3 . Transcript abundance was normalized to 16S rRNA and relative transcription calculated using MxPro software ( Stratagene ) . In order to construct plasmids for chromosomal exchange , flanking regions of the z3357 , z1449 ( cIIstx2 ) genes and Stx1 and Stx2 phages were PCR amplified and cloned into the temperature-sensitive plasmid pIB307 ( table S2 ) . Primer pairs as described in table S3: z3357 up 5′ , z3357 up 3′ and z3357 down 5′ , z3357 down 3′; z1449 up 5′ , z1449 up3′ and z1449 down 5′ , z1449 down 3′; stx1 up 5′ , stx1 up3′ and stx1 down 5′ , stx1 down 3′; stx2 up 5′ , stx2 up3′ and stx2 down 5′ , stx2 down 3 were used to amplify z3357 , z1449 Stx1 phage and Stx2 phage flanking regions sequences from E . coli O157:H7 EDL933 . These products were cleaned with a Invitrogen PCR purification kit , digested with BamHI and HindIII or SacI , or SalI , re-cleaned , and then ligated with digested pIB307 ( table S2 ) to obtain pXLS10 , pXLS11 , pXLS12 and pXLS15 containing flanking regions of z3357 , z1449 , stx1 phage , and stx2 phage respectively . To produce plasmids for exchange , a sacBkan cassette was cloned into the BamHI sites of pXLS11 , pXLS10 and pXLS12 creating pXLS13 , pXLS14 and pXLS16 that would allow the chromosomal replacement of z1449 , z3357 and stx1 genes respectively with a selectable marker and a counter-selection gene . The method of Emmerson et al . was used for allelic exchange [66] . Plasmid pXLS13 ( table S2 ) was transformed into EDL933 to obtain the intermediate strain with z1449 ( cIIstx2 ) replaced by sacBkan cassette . Briefly , pXLS13 was electroporated into EDL933 ( table 1 ) and cultured at 30°C on LB-C plates . Ten transformants were inoculated into pre-warmed LB-C at 42°C and passaged repeatedly in LB-C broth at 42°C to obtain co-integrates . The culture was further passaged at 30°C in LB-K broth to select for the complete exchange . The kanamycin resistant and chloramphenicol sensitive strain was confirmed by PCR with primer pairs z1449 5′ and z1449 3′ for inside of z1449 , z1449 external 5′ and z1449 external 3′ for outside of z1449 , sacB 5′ and sacB 3′ for sacBkan cassette ( table S3 ) . The resultant strain was termed ZAP1321 . ZAP1321 was then transformed with plasmids pXLS15 and pXLS11 and allelic exchange carried out to generate strains with Stx2 phage and z1449 ( cIIstx2 ) clean deletions respectively . These allelic exchanges were as above except that the cultures at 30°C did not contain any antibiotic selection . The required clones were identified by sensitivity to both kanamycin and chloramphenicol and confirmed by PCR using primer pairs: wrbA and intW plus z1503 and z1504 for the junction sites of Stx2; stx2 5′ and stx2 3′ for a region inside the Stx2 phage; z1449 5′ and z1449 3′ for inside of z1449 ( cIIstx2 ) , z1449 external 5′ and z1449 external 3′ for outside of z1449 , sacB 5′ and sacB 3′ for sacBkan cassette . The clean Stx2 prophage and z1449 deletion strains were termed ZAP1322 and ZAP1323 respectively ( table 1 ) . In order to construct a strain deleted for both Stx1 and Stx2 phages , ZAP1322 was used to generate a strain with the Stx1 prophage replaced with the sacBkan cassette using pXLS16 and the exchange protocol as above . The strain was confirmed by resistance profile and PCR with appropriate primer pairs ( table S3 ) . The resultant strain was termed as ZAP1326 ( EDLStx1&2− ) . To construct a strain with both z1449 ( cIIstx2 ) and z3357 ( cIIstx1 ) deleted , ZAP1323 was used for allelic exchange . pXLS14 was transformed into ZAP1323 and the exchange carried out as described above to produce strain with z3357 replaced by a sacBkan cassette . The strain was confirmed by antibiotic resistance profile and PCR with appropriate primer pairs ( table S3 ) . This strain was termed ZAP1324 . ZAP1324 was electroporated with pXLS10 and the clean deletion of z3357 ( cIIstx1 ) was obtained by allelic exchange and confirmed by resistance profile and PCR . The final z3357 and z1449 clean deletion was termed ZAP1325 ( table 1 ) . A marked Sp5 bacteriophage from an Stx negative variant of E . coli O157:H7 Sakai containing a stx2A::kmr cassette [67] was transferred to other strains by either transduction or conjugation . For transduction , the bacteriophage were prepared by culture of the strain in LB at 37°C followed by addition of 1 µg/ml mitomycin C ( MMC ) until lysed . The culture was centrifuged to pellet debris then chloroform used to extract bacteriophage from the supernatant and to kill any remaining cells . This bacteriophage preparation was used to transduce a spontaneous Nalidixic acid resistant strain of E . coli K12 EDCM367 [68] . Positive transductants were selected on LB agar plates containing kanamycin and then screened on nalidixic acid . The presence of the marked Stx prophage was further confirmed by PCR and MMC-induced lysis . For conjugation into TUV93-0 or TUV93-0ΔgadE ( table 1 ) , Hfr JC5029 ( E . coli Genetic Stock Center ) strain transduced with the marked Sp5 as above was used . Spontaneous Nalr TUV93-0 strains were selected . These and JC5029 ( Sp5 ) were transformed with pBAD-CI ( table S2 ) to prevent zygotic induction following conjugation and to allow the donor to grow in the presence of ampicillin during the mating . Plate mating was carried out on LB agar+0 . 2% arabinose and ampicillin . The bacteria were washed off with LB and then selected on agar plates containing kanamycin and nalidixic acid . The absence of the Hfr strain was confirmed by sensitivity to spectinomycin to which JC5029 is resistant . The final strain was confirmed to contain the marked Sp5 by PCR and lysis following addition of mitomycin C ( 1 µg/ml ) . To remove pBAD-CI ( table S2 ) from TUV93-0 ( Sp5 ) or TUV93-0ΔgadE ( Sp5 ) , the strain was transformed with pBR322ΔScaI-SspI with selection for tetracycline resistance and screening for sensitivity to ampicillin on LB agar plates . Phylogenetic analysis of 16 cII nucleotide sequences from publicly available Stx1- and Stx2/Stx2c-encoding bacteriophages was carried out by maximum likelihood as implemented in PHYLIP ( http://www . phylip . com/ ) . Bootstrap values were calculated from 100 replicates . Sequences were obtained from both complete bacterial genomes and complete phage genomes from the NCBI RefSeq database ( source genome accessions and cII locus tag unique identifiers are listed with square and round brackets , respectively ) ; i ) cII sequences from complete bacterial genomes: E . coli O157:H7 str . EDL933 [NC_002655] Stx1 phage ( Z1449 ) and Stx2 phage ( Z3357 ) , E . coli O157:H7 str . EC4115 [NC_011353] Stx2 phage ( ECH74115_3554 ) and Stx2c phage ( ECH74115_2923 ) , E . coli O157:H7 str . Sakai ( NC_002695 ) Stx1 phage ( ECs2988 ) and Stx2 phage ( ECs1187 ) , E . coli O157:H7 TW14359 [NC_013008] Stx2 phage ( ECSP_3270 ) and Stx2c phage ( ECSP_2739 ) , E . coli O26:H11 str . 11368 [NC_013361] Stx1 phage ( ECO26_1586 ) ; ii ) cII sequences from complete phage genomes: Stx1 phage BP-4795 [NC_004813] from E . coli O84:H4 str . 4795/97 ( phi4795p23 ) ; Stx2 phage Min27 [NC_010237] from E . coli O157:H7 str . Min27 ( pMIN27_28 ) ; Stx1 phage Stx1phi [NC_004913] from E . coli O157:H7 str . Morioka V526 ( Stx1_gp56 ) , Stx2 phage Stx2phiII [NC_004914] from E . coli O157:H7 str . Morioka V526 ( Stx2II_gp57 ) , Stx2 phage Stx2phiI [NC_003525] from E . coli O157:H7 str . Okoyama ( Stx2Ip123 ) ; Stx1 phage YYZ-2008 [NC_011356] from E . coli O157:H7 str . EC970520 ( YYZ_gp27 ) ; and Stx2 phage 1717 [NC_011357] from an unspecified strain of E . coli O157:H7 ( Stx2-1717_gp22 ) . With the exception of pMIN27_28 ( which contains an in-frame 3 nucleotide insertion relative to other cII sequences ) all sequences were 297 nucleotides in length and aligned without gaps . Differences in the level of fluorescence from the PT32 strains compared to PT21/28 strains and strains containing both Stx2 or Stx2c prophages compared to strains only containing one were compared by standard two-sample t-tests on log10 transformed values . Standard two-sample t-tests were also performed to compare relative fluorescence in specific TUV , EDL and K-12 strains . A fisher exact test was carried out to compare in PT32 and PT21/28 strains the proportion exhibiting expression levels above a threshold . All analyses were carried out in R ( © R Development Core Team ( 2009 ) . R version 2 . 10 . 1 ( © Foundation for Statistical Computing , Vienna , Austria . ISBN 3-900051-07-0 , URL http://www . R-project . org . ) . P<0 . 05 was taken to indicate statistical significance . Statistical analysis of qRT-PCR data was carried out using REST 2009 software [69] .
|
Many significant infectious diseases that impact human health evolve in animal hosts . Our work focuses on infections caused by strains of enterohemorrhagic Escherichia coli ( EHEC ) that cause bloody diarrhoea and life threatening kidney and brain damage in humans as an incidental host , while ruminants are a reservoir host . EHEC strains are infected with bacteriophages that can integrate their genetic material into the bacterial chromosome . This includes genes for the production of Shiga toxins ( Stx ) that are responsible for the severe pathology in humans . It has been demonstrated that certain EHEC strains are more likely to be associated with human disease and ‘supershedding’ animals . The current study has shown that these EHEC strains are more likely to contain two related Stx bacteriophages , rather than one , and that the intercalating bacteriophages take control of the bacterial type III secretion system that is essential for ruminant colonization . We propose that this regulation favours co-acquisition of other genetic regions that encode type III-secreted proteins and regulators that can overcome this control . This finding helps our understanding of EHEC strain evolution and indicates that selection of more toxic strains may be occurring in the ruminant host with important implications for human health .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"animal",
"types",
"infectious",
"diseases",
"zoonoses",
"large",
"animals",
"veterinary",
"microbiology",
"genetics",
"biology",
"microbiology",
"evolutionary",
"biology",
"gastrointestinal",
"infections",
"genetics",
"and",
"genomics",
"veterinary",
"science"
] |
2012
|
Lysogeny with Shiga Toxin 2-Encoding Bacteriophages Represses Type III Secretion in Enterohemorrhagic Escherichia coli
|
Kaposi's sarcoma-associated herpesvirus ( KSHV ) stabilizes hypoxia-inducible factor α ( HIF-1α ) during latent infection , and HIF-1α reactivates lytic replication under hypoxic stress . However , the mechanism utilized by KSHV to block lytic reactivation with the accumulation of HIF-1α in latency remains unclear . Here , we report that LANA encoded by KSHV contains a unique SUMO-interacting motif ( LANASIM ) which is specific for interaction with SUMO-2 and facilitates LANA SUMOylation at lysine 1140 . Proteomic and co-immunoprecipitation analysis further reveal that the SUMO-2 modified transcription repressor KAP1 is a critical factor recruited by LANASIM . Deletion of LANASIM led to functional loss of both LANA-mediated viral episome maintenance and lytic gene silencing . Moreover , hypoxia reduced KAP1 SUMOylation and resulted in dissociation of both KAP1 and Sin3A repressors from LANASIM-associated complex . Therefore , the LANASIM motif plays an essential role in KSHV latency and is a potential drug target against KSHV-associated cancers .
Kaposi's sarcoma-associated herpesvirus ( KSHV ) , or human herpesvirus 8 ( HHV-8 ) , is the etiological agent of Kaposi's sarcoma ( KS ) , and is tightly associated with primary effusion lymphoma ( PEL ) and a subset of multicentric Castleman's disease ( MCD ) [1] , [2] . KSHV is a large DNA tumor virus encoding more than 90 open reading frames ( ORF ) . Like all herpesviruses , KSHV has two distinct phases of its life cycle: latency and lytic replication . During latent infection , only a few genes are expressed [3] . The latency-associated nuclear antigen ( LANA ) encoded by ORF73 is one of the dominant latent proteins with multiple functions . These include tethering the viral episome to the host chromatin , inhibiting tumor suppressors , regulating gene transcription and blocking apoptosis for establishment and maintenance of latent infection ( reviewed in [4] , [5] , [6] . The replication and transcriptional activator ( RTA ) encoded by ORF50 is essential for initiation of viral lytic replication [7] , [8] , [9] , [10] . RTA drives lytic replication via activation of a transcription cascade of KSHV genes expressed during the early stages of the viral life cycle [7] , [9] . Studies of KSHV-induced cell transformation in vivo demonstrated that KSHV predominantly enters a latent state [11] , [12] . However , clinical analysis of KS patient samples have shown that a fraction of infected cells can also enter into lytic replication producing infectious virions and expressing virus-encoded paracrine signaling factors [13] . This suggests that spontaneous reactivation from latency in the tumor microenvironment is also important for growth and dissemination of viral infected tumor cells . Hypoxic stress is a common feature of the tumor microenvironment [14] , [15] , [16] , which results from oxygen consumption by successive layers of tumor cells distal to blood vessels or temporary vessel closure [17] . The extent of tumor hypoxia is strongly associated with tumor development and malignant progression . To elucidate the effect of intratumoral hypoxia on KSHV-associated cancers , we and other groups have shown that hypoxic stress reactivates KSHV lytic replication in vitro [18] , [19] . In hypoxia , the major latent antigen LANA together with the key hypoxia responder HIF-1α ( the inducible subunit of the heterodimeric transcriptional factor HIF-1 ) binds to the hypoxia-responsive elements ( HRE , 5′-RCGTGC-3′ ) within the RTA gene promoter inducing lytic replication [18] . Interestingly , under normoxic conditions , HIF-1α is aberrantly accumulated in both KSHV latently infected cells and KS patient tissues in a LANA-dependent manner [20] , [21] . These results indicate that LANA plays a dual role in controlling HIF-1α transcriptional activity for KSHV latent and lytic replication . However , the mechanism utilized by LANA to exert this dual function in normoxia and hypoxia remains unclear . Emerging studies indicate that post-translational modification of proteins by the small ubiquitin-like modifier ( SUMO ) plays an important role in epigenetic control of gene transcription , and in response to hypoxic stress [22] , [23] , [24] , [25] . SUMO is covalently attached to a lysine residue of target proteins through an isopeptide bond by three enzymes: heterodimeric SUMO-activating enzyme E1 ( Uba2-Aos1 ) , SUMO-conjugating enzyme E2 ( Ubc9 ) , and the substrate recognition factors or E3 ligases ( i . e . PIAS protein family ) [26] . Different from ubiquitination , SUMO conjugation often requires a consensus sequence ΨKxE/D ( Ψ , large hydrophobic residue; x , any amino acid ) around the target lysine [27] . The consequence of SUMOylation leads to various facets of protein function , including regulation of gene transcription , cell cycle , DNA repair and subcellular localization . In mammalian cells , there are at least four isoforms of SUMO – SUMO-1 , -2 , -3 and -4 . SUMO-1 is the major SUMO in human cells; SUMO-2 and SUMO-3 are highly similar and responds to different cellular stresses [28] , [29]; and SUMO-4 is tissue specific and so far only found in the pancreas [30] . In addition to attaching covalently to substrates , SUMO can also non-covalently bind to other proteins through a consensus SUMO-interacting motif ( SIM ) , which is identified as sequence h-h-x-S-x-S/T-a-a [31] , V/I-x-V/I-V/I [32] , or K-x3–5-I/V-I/L-I/L-x3-D/E/Q/N-D/E-D/E [33] . Thus , in regards to the biological functions of these modified substrates , the SUMOylated targets may depend on their ability to interact with other effectors containing SIM motifs . In this study , we demonstrate that LANA contains a unique SUMO-interacting motif ( LANASIM ) specific for interaction with SUMO-2 and that LANA can also be SUMOylated on lysine 1140 . Proteomic analysis further showed that LANASIM recruits a transcription inhibitory complex which includes two co-repressors KAP1 and Sin3A . Strikingly , we found that deletion of the LANASIM motif sufficiently abolishes its association with poly-SUMO2 modified KAP1 , and loss of LANA's ability to support KSHV episome maintenance and gene silencing . Furthermore , hypoxia blocks poly-SUMO2 modified KAP1 and leads to dissociation of KAP1 and Sin3A from the LANASIM complex , as well as loss of LANA SUMOylation , which in turn transactivates viral gene expression for lytic replication . This describes the first mechanism by which a viral protein presents a specific SIM-motif platform to selectively recruit a cellular transcriptional complex for viral episome maintenance and gene silencing during latent infection .
To investigate whether LANA is involved with the SUMO-signaling pathway , we performed co-immunoprecipitation ( co-IP ) assays by co-expressing myc-tagged LANA with FLAG-tagged SUMO-1 or SUMO-2 under normoxic conditions . The results showed that LANA dramatically associated with the SUMO-2 not SUMO-1 modified substrate [ ( SUMO1/2 ) n-sb] with high molecular weight ( >170 kDa ) ( Figure 1A , lanes 3 and 4 ) . Interestingly , when hypoxic stress was induced by CoCl2 or 1% oxygen , the association of LANA with ( SUMO2 ) n-sb was significantly reduced ( Figure 1A , lanes 4 , 6 and 8 ) , while a moderate increase in the association of LANA with ( SUMO1 ) n-sb was observed ( Figure 1A , lanes 3 , 5 and 7 ) . In contrast , there was a reduced level of ( SUMO2 ) n-sb [namely more “low” than “high” of ( SUMO2 ) n-sb appearance] in whole cell lysates in hypoxia compared with normoxia ( Figure 1A , compare lanes 4 with 6 and 8 in enlarged region ) , suggesting that the reduced association of LANA with ( SUMO2 ) n-sb in response to hypoxia may be due to less SUMO-2 modification of target proteins . To identify the domain required for LANA to interact with poly-SUMO-2 , we performed similar co-IP assays by using different mutants of LANA . The results show that residues 233 to 340 at the amino terminus are critical for the strong interaction of LANA with the ( SUMO2 ) n-sb ( Figure 1B and supplementary Figure S1 ) , and the central region which contains residues 329 to 925 , a highly repetitive region provides minimal contribution to the interaction of LANA with ( SUMO1 ) n-sb and ( SUMO2 ) n-sb ( Figure 1B and supplementary Figure S1C , upper panel ) . In contrast , as the position of ( SUMO2 ) n-sb size did not change dramatically , even though the molecular weight of LANA truncated mutants ( FL , NC , N3 , N2 and N1 ) was varied from 65 kDa to 250 kDa . This indicates that the majority part of the detected ( SUMO2 ) n-sb is not the SUMO2-modified LANA . To further determine whether a cellular SUMO-interacting motif ( SIM ) -like sequence is located within the residues 233 to 340 of LANA , we aligned the LANA sequence with the three reported cellular SIM sequences [31] , [32] , [33] . The results identified two potential SIM-like motifs ( SIM1 and SIM2 ) with partial homology to cellular SIM motifs both located within the residues 233 to 340 ( Figure 1B and supplementary Table S1 ) . To prove that these SIM motifs ( LANASIM ) are required for LANA to interact with the ( SUMO2 ) n-sb , we generated a series of mutations within the SIM motif of LANA 1–340 ( N3 ) by site directed mutagenesis and again performed similar co-IP assays with SUMO-2 . Strikingly , the results show that deletion of the SIM1 , SIM2 , SIM1/2 , or mutation of three residues within SIM2 motif ( 3IA or 3SA ) resulted in a significant reduction in the interaction of LANA 1–340 with ( SUMO2 ) n-sb ( Figure 1C ) . Therefore , we speculated that the intact 3D structure of the SIM motif within the amino terminus of LANA is important for interaction with the ( SUMO2 ) n-sb . To verify whether the LANASIM motif is specifically required for LANA to preferentially associate with ( SUMO2 ) n-sb instead of ( SUMO1 ) n-sb , we performed similar assays using full-length LANA or its NC truncated mutant with or without deletion of the SIM motifs . Indeed , deletion of the LANASIM motif completely abrogated the interaction of the NC truncated mutant of LANA with ( SUMO2 ) n-sb while no detectable effect with ( SUMO1 ) n-sb ( supplementary Figure S1B , lanes 3 and 6 ) , and also specifically reduced the interaction of full length LANA with ( SUMO2 ) n-sb but not ( SUMO1 ) n-sb ( supplementary Figure S1C , compare lanes 4 , 5 with 2 , 3 ) . In addition , the results from in vitro pull-down assays of His-SUMO-1 or SUMO-2 recombinant protein with radioactive labeled wild type LANA or its SIM-deleted mutant showed that LANA only bound to SUMO-2 not SUMO-1 and this interaction of LANA with SUMO-2 was dependent on the LANASIM motif ( Figure 1D ) , further confirming that the LANASIM motifs are critical for LANA to strongly interact with ( SUMO2 ) n-sb through SUMO-2 . To further confirm that LANASIM is required for LANA to interact with SUMO-2 , we investigated the subcellular localization of LANA and SUMO-2 by using the wild type or SIM-deleted NC truncated mutant of LANA . Consistently , the results showed that SUMO-2 localizes predominantly to the outside compartment of chromosomal DNA when expressed alone ( Figure 2A ) . However , we observed 64 . 8% co-localization of SUMO-2 with LANA on chromosomal DNA in the presence of wild type LANA ( Figure 2B , upper panels ) . In contrast , only 0 . 5% SUMO-2 was co-localized with LANA on chromosomal DNA when the SIM motif of LANA was deleted ( ΔSIM ) ( Figure 2B , lower panels ) . To further demonstrate this point , the sequential ChIP assays were performed in the presence of TR DNA . The results also showed that wild type not SIM-deleted mutant dramatically associated with SUMO-2 on DNA compartment ( Figure 2C ) . This suggests that LANASIM is required for LANA to co-localize with SUMO-2 on chromosomal DNA . In consistence , the results from immunofluorescence analysis of endogenous LANA and SUMO-2 in PEL cells further supported the notion that hypoxia reduces co-localization of LANA with SUMO-2 ( Figure 2D ) . To identify which proteins with SUMO-2 modification interact with the LANASIM motif , we first analyzed the 3D structure of LANA by Robetta prediction software ( http://robetta . bakerlab . org/ ) . Strikingly , we found that the region which contains residues 240 to 300 forms a separate domain with two sub-SIM ( SIM1 and SIM2 ) motifs individually distributed on each side ( Figure 3A ) . This potentially provides a “riding-horseback” like platform important for LANA interaction with poly-SUMO-2 . In addition , we also noticed that the AD domain of the central region is located very close to the LANASIM motif ( Figure 3A ) , which may explain why the deletion of the AD domain decreases the interaction of LANASIM with poly-SUMO-2 . Based on this predicted 3D structure of the SIM domain , we generated a GST-SIM fusion protein by using the residues 240 to 300 region and performed GST pulled down assay with nuclear extracts from 293 cells expressing exogenous SUMO-2 . The results from gel fractionation results showed that at least four unique bands ( b1 , b2 , b3 and b4 ) were pulled down by the GST-SIM protein ( Figure 3B , upper panel ) . Analysis of these bands by mass spectrometry identified five proteins: DNA-PKc , the transcriptional coactivators CBP and p300 , and the corepressors Sin3A and KAP1 , respectively ( Figure 3B , lower panel ) . Interestingly , previous studies have suggested that these five proteins are responsive to hypoxic stress [34] , [35] , [36] , [37] . In addition , except for DNA-PKc , the other four transcription factors were reported to undergo SUMO modification [38] , [39] , [40] , [41] . To elucidate which proteins in this complex are critical for interaction with the LANASIM motif through poly-SUMO-2 , we first performed similar GST pull-down assays by using GST fusion with the wild type ( WT ) or the SIM-deleted ( ΔSIM ) mutant of LANA N terminus 1–329 , following western blot with specific antibodies for each of the five proteins . The results showed that deletion of the LANASIM motif results in a dramatically dissociation of KAP1 and Sin3A with the LANASIM-mediated complex ( data not shown ) . To further verify if KAP1 , Sin3A or both undergo SUMO-2 modification as LANASIM targets are impaired by hypoxia stress , HEK293 and BJAB cells individually transfected with FLAG-SUMO-2 or vector were treated with or without hypoxia and subjected to denature IP of KAP1 and Sin3A . The results showed that KAP1 but not Sin3A was dramatically modified by SUMO-2 ( 1 , 2 , 3 and 4 copies of SUMO-2 according to 19 kDa molecular weight of SUMO-2 ) in the presence of FLAG-SUMO-2 . Furthermore , this modification was blocked by hypoxia ( Figure 3C , compare lanes 2 with 4 , similar results were observed in BJAB cells ) . Intriguingly , although the denatured form of SUMOylated KAP1 ( de-suKAP1 ) was not pulled down by the SIM domain of LANA fusion with GST in vitro ( data not shown ) , the native co-immunoprecipitation assays showed that the LANASIM motif within full length LANA or its NC truncated mutant had a higher affinity with 2×SUMO-2 modified KAP1 than 1×SUMO-2 modified one ( Figure 3D , lanes 1 and 3 ) , and deletion of the LANASIM motif dramatically reduces or completely abrogates the interaction of LANA with 1 or 2× SUMO-2 modified KAP1 ( Figure 3D , compare lanes 1 and 2 with 3 and 4 ) . Unexpectedly , deletion of SIM motif in both full length LANA and its NC truncated mutant led to appearance of a degraded band ( ∼95 kDa ) ( Figure 3D , lanes 2 and 4 indicated by asterisk ) . The result of less association of LANA with native Sin3A once the SIM motif was deleted ( Figure 3D , middle panel , lanes 2 and 4 ) , further supports the notion that the poly-SUMO2 modified chain of KAP1 is the critical one targeted by the LANASIM motif . In addition , the results of in vitro binding assays showed no direct interaction of His-KAP1 with in vitro translated proteins of full length LANA and its different truncated mutants ( Figure 3E ) , suggesting that LANA predominantly interacts with poly-SUMO2 modified KAP1 through the LANASIM motif . To further confirm that LANA does bind with the SUMO-2 modified KAP1 , the wild type KAP1 and its mutant 6KR ( all 6 putative SUMOylated lysines are mutated to arginine ) with FLAG tag were individually used to perform co-IP with myc-tagged LANA in the presence of HA-SUMO-1 , HA-SUMO-2 or vector alone . The results showed that as observed with endogenous KAP1 and SUMOylation , LANA strongly associated with 1 or 2×SUMO-2 modified KAP1 with greater affinity for the 2×SUMO2-KAP1 , although there was more 1× SUMO-2 modified KAP1 seen in the whole cell lysate ( Figure 4A ) . Furthermore , the results of LANA association with 1 or 2×SUMO-2 modified KAP1 with the lysines mutated indicated that LANASIM binds preferentially with poly-SUMO2 modified KAP1 and that these lysines contributed to KAP1 SUMOylation to various degrees with a complete loss of association when all 6 lysines were mutated ( Figure 4B ) . In PEL cells with hypoxia , less association of SUMO2-modified and native KAP1 with endogenous LANA than that in normoxia ( Figure 4C ) , further demonstrated that the association of LANA with KAP1 is sensitive to hypoxic stress . To determine whether the LANASIM motif of LANA contributes to its SUMO modification , we performed native and denatured ( which removes non-covalent binding but retains covalent binding of SUMO modification ) , co-IP assays by using full length LANA or its LANASIM-deleted mutant ( LANAΔSIM ) coexpressed with exogenous SUMO-1 or SUMO-2 . Ubc9 was used as a positive control in this assay . The results showed that LANA can be modified by both SUMO-1 and SUMO-2 ( supplementary Figure S2 , upper panels ) . Importantly , the signals indicating that LANA was SUMO-2 modified were greater than that seen for SUMO-1 ( supplementary Figure S2 , upper panel , lanes 5 and 6 ) . Consistent with our findings that the LANASIM motif associates with SUMO-2 but not SUMO-1 , and the LANASIM motif deletion resulted in a remarkable reduction of SUMO-2 modification of LANA ( supplementary Figure S2 , compare lanes 6 with 8 ) . In contrast , the levels of SUMO1-modified LANA were relatively unchanged in the ΔSIM mutant compared to WT LANA ( supplementary Figure S2 , compare lanes 5 with 7 ) . As the LANASIM-associated poly-SUMO2 modified KAP1 is highly sensitive to hypoxic stress , we wanted to know if SUMO-2 modification of LANA is also impaired by hypoxia . We monitored the SUMO-2 modified levels of endogenous LANA by in vivo SUMOylation assays in PEL cells under normoxic and hypoxic conditions . The KSHV negative ( no LANA ) BJAB cell line was used as a negative control . The results showed that the levels of both SUMO-2 modified LANA ( suLANA ) and native LANA associated with SUMO-2 in BC3 cells are also dramatically reduced under hypoxic conditions ( supplementary Figure S3 ) . This corroborated the above data that the LANASIM motif does contribute to SUMO-2 modification of LANA due to dissociation from hypoxia-sensitive SUMOylated KAP1 . To determine which residues of LANA targeted for SUMOylation , we tested the SUMOylated pattern of different amino or carboxyl truncated forms of LANA ( to avoid the problem of resolution of the SUMO-modified bands linked to full length LANA ) in the presence of SUMO-1 or SUMO-2 by in vivo SUMOylation assays . The results showed that when coexpressed with SUMO-1 or SUMO-2 , only the central region deleted ( NC ) and carboxyl terminus mutants of LANA but not its amino terminus mutant appear to have SUMO-1 or SUMO-2 modified forms of LANA ( with slower migration compared to unmodified LANA ) ( data not shown ) . To identify which specific residues are SUMO modified , we analyzed the carboxyl terminal sequence of LANA according to the SUMOylated consensus sequence ΨKxE/D [27] , and identified two lysine residues 1081 and 1140 which are most potentially targeted for SUMOylation ( supplementary Figure S4 , right panel ) . Follow-up SUMOylation assays in cells indicated that lysine 1140 of LANA is the key residue for both SUMO-1 and SUMO-2 modification ( supplementary Figure S4 , left panel ) . In view of the fact that the LANASIM motif contributes to SUMOylation of LANA , to further verify if the LANASIM-associated complex ( including KAP1 ) acts as a bridge for the amino terminus of LANA to interact with its carboxyl terminus , and in turn enhances SUMO modification of carboxyl terminus of LANA , we performed co-IP assays by expressing FLAG-tagged C terminus of LANA ( LANA-C ) with myc-tagged amino terminus ( WT or ΔSIM ) of LANA ( LANA-N ) in 293 cells . The results showed that the association of the amino terminus with the carboxyl terminus of LANA is dependent on the presence of SIM motif to a large extent ( supplementary Figure S5A , lanes 2 and 3 ) . In addition , co-expression of the amino with the carboxyl terminus of LANA dramatically enhances the SUMOylation at its carboxyl terminus ( supplementary Figure S5B , lanes 1 and 2 ) . Furthermore , the levels of the amino terminus-induced SUMOylation of the carboxyl terminus were significantly decreased when LANASIM-associated KAP1 was knocked down ( supplementary Figure S5B , compare lanes 2 with 4 ) . This further supports the notion that the LANASIM motif and KAP1 play a role in SUMOylation of LANA . Since LANA plays a critical role on maintenance of viral episome during latency , we wanted to know if the LANASIM deletion impairs the ability of LANA binding with the TR and the persistence of the viral episome during cell passage . To this end , we performed chromatin immunoprecipitation assays by using wild type ( WT ) LANA or a series of its mutants ( ΔSIM , ΔAD , ΔAD , SIM and C ) co-expressed with the TR-Puromycin plasmid in 293 cells for 48 hours . The results showed that there were no significant difference between wild type LANA and other mutants ( Figure 5A , upper panels ) . To further determine if the persistence of LANA-binding TR was affected , the transfected cells of LANA and TR-Puromycin were subsequently subjected to colony formation assay where cells with TR plasmid were selected with Puromycin for 2 weeks ( a greater number of colonies would indicate higher efficiency of the TR persistence ) . The results showed that both wild type ( WT ) LANA and its NC truncated mutant ( ΔAD ) presented the highest efficiency of TR maintenance , while both LANASIM-deleted mutants ΔSIM and ΔAD , SIM led to a dramatic loss of TR colonies ( Figure 5A , lower panels ) . In contrast , due to the absence of the amino terminus of LANA for tethering to host chromosomal DNA [42] , there was little or no TR colony formation seen in the carboxyl terminal truncated mutants ( C ) which were similar to vector alone ( Figure 5A , lower panel ) . These results indicate that the LANASIM motif is an important contributor to LANA's role in maintenance of the KSHV episome . To answer if the LANASIM motif also contributes to inhibition of lytic replication , we transiently transfected LANASIM -deficient mutants of LANA as a dominant negative into 293 cells stably carrying the complete KSHV-Bac36 ( viral genome ) , and then monitored for intracellular viral episomal DNA by quantitative PCR analysis . Results with wild type LANA , NC truncated mutant ΔAD or vector alone as control dramatically reduced virion DNA production . However , we also showed that both the LANASIM-deleted mutants ΔSIM and ΔAD , SIM dramatically enhanced viral episome replication ( Figure 5B ) . This strongly suggested that in addition to maintenance of the viral episome , the LANASIM-associated complex including KAP1 and Sin3A are also important for blocking lytic replication . To determine whether this LANASIM -dependent inhibition of lytic replication is due to silencing RTA ( the key activator of the KSHV lytic life cycle ) expression , the RTA promoter was subjected to luciferase reporter assays in the presence of wild type LANA , its mutants with or without LANASIM deletion ( ΔSIM , ΔAD , and ΔAD , SIM ) or vector alone . The results showed that the LANASIM-deletion alone ( ΔSIM ) recovered about 60% of the inhibition of wild type LANA whereas the ΔAD mutants had negligible effect ( Figure 5C , left panel ) . However , the ΔAD , SIM mutant led to an almost complete reversal of the inhibitory activities ( Figure 5C , left panel ) . In contrast , hypoxic stress efficiently reversed the inhibitory function of LANA resulting in transactivation of the RTA promoter regardless of whether the wild type LANA or the NC truncated mutants with or without LANASIM deletion ( ΔAD , ΔAD , SIM ) were used in the assay ( Figure 5C , right panel ) . This further supports the notion that the LANASIM-mediated complex ( including KAP1 ) is involved in the inhibitory function of LANA in RTA expression and is sensitive to hypoxia . To further confirm that LANASIM-mediated KAP1 and Sin3A are indeed required for LANA to repress the RTA promoter through the LANASIM motif , we performed RTA promoter reporter assays in the presence or absence of LANA in both BJAB and 293 cell lines with Lentivirus-mediated constitutive knockdown of KAP1 or Sin3A or nonspecific control . Consistent with our findings , the results showed that knockdown of KAP1 or Sin3A efficiently blocked the inhibitory function of LANA on the RTA promoter in normoxia ( Figure 5D ) . This is consistent with our observation that the dominant negative LANASIM-deleted mutant does in fact enhance KSHV virion production ( Figure 5B ) . To determine whether the hypoxia-induced functional switch of LANA is due to the dissociation of the LANASIM-associated proteins ( like KAP1 and Sin3A ) from the regulatory complex LANA-HIF-1α particularly on the HIF-1α-DNA binding sites , we performed a biotin-labeled DNA oligo pull-down assay in vitro by incubating the nuclear extract with wild type or a mutant of the HIF-binding site- ( mut ) HRE DNA oligo . The specific proteins bound to HRE were pulled-down by Strepavidin agarose beads and were subjected to western blot analysis . The results showed that in normoxia the LANASIM-associated proteins KAP1 and Sin3A in the regulatory complex of LANA and HIF-1α bound to HRE ( Figure 6A ) . However , in hypoxia , KAP1 and Sin3A as well as SUMO-2 ( where the SUMO-2 position corresponds to the pattern of poly-SUMO2 modified KAP1 as calculated by molecular weight ) were dramatically released from the LANA-HIF-1α/HRE complex ( Figure 6A , right panels ) . Notably , we observed that a modified form ( potential SUMO-2 modification based on the pattern of SUMO2 modified LANA ) of LANA was also dissociated from the LANA-HIF-1α/HRE complex while interestingly HIF-1α appears with an additional modified slower migrating band ( Figure 6A , right panels ) . To further determine if SUMOylation is important for DNA binding of the LANASIM-mediated complex , the ChIP assays of HRE region within the ORF50 promoter were carried out in BC3 cells with or without transiently Ubc9 knockdown . The results showed that inhibition of Ubc9-mediated SUMOylation dramatically reduced both KAP1 and Sin3A association with LANA-HIF-1α complex at the HRE DNA ( Figure 6B ) . Taken together , this data above strongly support our hypothesis that KAP1 and Sin3A particularly SUMO-2 modified forms are indeed sensitive to hypoxic stress and are recruited by the LANASIM motif during KSHV latent infection to silence the RTA promoter under normoxic conditions .
Despite the fact that latent infection is the dominant form of KSHV infection which leads to transformation of the host cell , reactivation from latency is also important for the growth and dissemination of viral associated tumor cells [13] . In addition to chemical triggers like phorbol esters and sodium butyrate , reactivation from KSHV latency can also be initiated by various physiological and environmental factors including hypoxia [43] . These agents are thought to directly or indirectly target the transcriptional activation of the promoter which drives the KSHV RTA ( ORF50 ) immediate-early gene . Our previous studies showed that LANA encoded by KSHV cooperates with HIF-1α bound to the HREs of the RTA promoter to induce lytic replication under hypoxic conditions [18] . Interestingly , we also found that LANA stabilizes HIF-1α in normoxia by blocking the negative regulators VHL and p53 and inducing nuclear accumulation [20] , [44] . To investigate how LANA switches from its inhibitory function to transcription activation at the RTA promoter , we now report that LANA has a unique SUMO2-interacting motif ( LANASIM ) and that this LANASIM motif specifically recruits a transcriptional inhibitory complex containing KAP1 and Sin3A for maintenance of the KSHV episome , as well as silencing of lytic gene ( i . e . RTA ) expression under normoxic condition . Of this complex , the poly-SUMO2 modified KAP1 is a critical factor targeted by the LANASIM motif . In hypoxia ( i . e . 1% O2 in this study ) , loss of SUMOylation of KAP1 leads to the LANASIM-associated complex particularly KAP1 and Sin3A dissociation from the LANA-HIF-1α/HRE complex , as well as deSUMOylation of LANA itself , which open the active form of LANA-HIF-1α , and in turn induces lytic gene expression and reactivates lytic replication in response to hypoxic stress ( Figure 6C ) . Interestingly , despite the fact that much dramatically dissociation of KAP1 and Sin3A from HIF-1α/HRE complex in the less SUMOylation status upon hypoxic stress , we did also observe that less native LANA association with HIF-1α/HRE complex to some extent in hypoxia when compared with that in normoxia , which could be due to the less stability of LANA without SUMO modification in hypoxia . The modification of proteins by SUMO conjugation has been demonstrated as a reversible process important for regulating the function of many target proteins [45] , [46] . The consequences of SUMO modification are not completely understood but have been shown to control various facets of protein function , including gene transcription , cell cycle , DNA repairs and protein localization [45] , [46] . The most recent discovery of SUMO-interacting motif leads further interpretation to the roles that SUMO modification can play on protein-protein interaction , conformational changes and subcellular localization [31] , [32] , [33] . In this study , we discovered that LANA has a unique SUMO-2 interacting motif ( LANASIM ) containing two SIM motifs with partial homology to cellular SIM motifs . Furthermore , this LANASIM motif specifically associated with poly-SUMO-2 instead of poly-SUMO-1 . The finding of poly-SUMO2 ( 2×SUMO-2 ) modified KAP1 associated with the LANASIM motif indicates that this interaction requires the intact 3D structure of the LANASIM motif . Notably , since SUMO-2 signaling is only activated due to extracellular stress , LANA may possess a unique SUMO-2 instead of SUMO-1 interacting motif to usurp the host SUMOylation pathway by mimicking stress signaling ( This could be the reason why LANASIM greatly associates with certain target proteins instead of others with SUMO-2 modification ) , thus creating a cellular environment favorable for viral infection . Other instances including cytomegalovirus IE2 [47] , human papillomavirus E2 [48] , Epstein-Barr virus BZLF1 [49] , and KSHV K-bZIP [50] , [51] , further support this notion . Our findings which showed that the AD domain of LANA's central region partially contributes to the LANASIM-associated protein-protein interaction , which subsequently enhances SUMO modification of LANA , supports previous discoveries that acidic residues located downstream from the core SUMO modification sites will enhance the efficiency of SUMOylation of target proteins [52] . In addition , we did not find a similar LANASIM-motif sequence within EBNA1 ( the LANA homologous antigen encoded by Epstein-Barr virus ) , or ORF73 encoded by HVS C488 , A11 or MHV68 . In contrast , distinct from recent report indicating that a SIM motif is located at carboxyl terminus of LANA and required for SUMOylation of histone H2B [53] , the hypoxia-based LANASIM discovered at amino terminus of LANA in our study could mainly target high-molecular-weight proteins ( i . e . KAP1 ) instead of low-molecular weight ones ( i . e . H2B ) . The fact that the LANASIM motif is required for interaction between amino and carboxyl terminal domains of LANA , further suggests that the LANASIM motif could serve as an adaptor for both intra-molecular and inter-molecular interactions of LANA as well as with other SUMO target proteins . The LANASIM motif recruits a specific transcriptional regulatory complex which includes corepressors , coactivators , and DNA binding proteins . KAP1 ( also known as TIF1β or TRIM28 ) was identified as a key component of the complex . It is well known that KAP1 is a universal corepressor protein from the Kruppel associated box ( KRAB ) -domain containing zinc finger proteins , the largest family of transcriptional silencers in the human genome [54] , [55] . KAP1 itself cannot bind DNA directly and recruits or coordinates the assembly of several chromatin-remodeling proteins including the histone deacetylase complex ( NCoR1 ) [56] , and the heterochromatin protein 1 ( HP1 ) family member [57] . Emerging evidence suggests that post-translational modifications , such as SUMOylation and phosphorylation can affect the ability of KAP1 to condense or relax chromatin [41] , [58] . SUMOylation at Lys554 , Lys779 and Lys804 of KAP1 generates a binding platform for SETDB and HADC1 to condense chromatin [58] ( This may explain why there was a lower percentage of KAP1 matched peptide than other proteins like Sin3A in MALDI-TOF-MS assay ) . In response to hypoxia , KAP1 was previously shown to be phosphorylated in an ATM-dependent manner on Ser824 , and phosphorylated KAP1 spreads rapidly throughout the chromatin [59] . In this study , the result of less SUMOylation of KAP1 in hypoxia provides further evidence that SUMOylation of KAP1 is hypoxia-sensitive and is linked to chromosome condensation and gene silencing . Therefore , knocking down KAP1 will lead to constitutive relaxation of cellular and viral chromatin for gene transcription and expression [60] . Taken together , these results provide a better understanding of the mechanism by which hypoxia drives the dissociation of KAP1 from LANA containing complexes for transactivation of target gene promoters . Another key transcriptional co-repressor identified associated with the LANASIM motif was Sin3A . Like KAP1 , Sin3A also functions as a co-repressor by recruiting HDAC1 for gene silencing [61] . Although Sin3A is shown as a substrate of TOPRS SUMO E3 ligase [40] , it was difficult to detect the SUMOylated form of Sin3A directly associated with the LANASIM motif . However , the evidence of Sin3A knockdown results in dramatic attenuation of the enhancement of the transcriptional activity at the RTA promoter , indicating that Sin3A is also a key component in the LANASIM-associated complex . In addition , both transcriptional co-activators p300 and CREB-binding protein ( CBP ) were also pulled down by the LANASIM domain . However , albeit that the LANASIM deletion did not significantly impair their association with LANA , which minimizes a critical role for these co-activators in the LANASIM-associated silencing complex , their contributions cannot be completely ruled out . The aim of LANA targeting p300 and CBP is likely to take advantage of the amount of p300 and CBP in order to compete with other DNA bound transcriptional factors and so is important for balancing the repression and activation at these major lytic promoters . Although a recent report showed that the lytic protein vPK blocks the SUMOylation of KAP1[62] , the fact that deSUMOylation of KAP1 in hypoxia not only occurs in KSHV positive cells but also negative cells , argues against the possibility that hypoxia may up-regulate other viral antigens which in turn lead to the loss of SUMOylation of KAP1 . In addition , the evidence that both KAP1 and Sin3A is released from the LANA-HIFα complex binding to the HRE elements during hypoxic stress provides additional questions for further investigation: 1 ) Is the deSUMOylation enzyme SENP up-regulated by hypoxia stress , and does deSUMOylation of KAP1/Sin3A by SENP lead to loss of KAP1/Sin3A binding with LANA in the silencing complex; 2 ) Does native or modified HIF-1α bind to LANA and block the access of KAP1 and Sin3A to the LANASIM motif ? In summary , our study provides new mechanistic insights into the programmed regulation of the KSHV genome during latent and lytic replication controlled by the major latent antigen LANA for gene silencing or activation based on a specific LANASIM-associated complex . Particularly , the discovery that the LANASIM-associated complex is highly sensitive to hypoxia ( a physiologically relevant environmental condition ) and so provides a potential therapeutic target against KSHV-associated cancers .
Plasmids encoding full length LANA with myc tag and its truncation mutants NC ( 1–329∧925–1162 ) , N1 ( 1–435 ) , N2 ( 1–233 ) , and N3 ( 1–340 ) were previously described [20] . GFP-C1 ( 930–1162 ) -myc and C ( 842–1162 ) -FLAG of LANA were obtained by ligating KpnI/EcoRV PCR fragments into pA3M-GFP and pA3F vector , respectively . GST-SIM ( 240–300 ) , GST-LN1–329 , and GST-LN1–329ΔSIM were generated by ligating KpnI/EcoRI PCR fragments into the derivative of pGEX-2TK ( Introduced a KpnI restriction enzyme site in open reading frame of GST ) . HA-Ubc9 was generated by ligating BamHI/EcoRI PCR fragments into the plasmid of pcDNA3-HA . Plasmids pRta-Luc , HA-SUMO-1/2 , FLAG-SUMO-1/2 , GFP-SUMO-1/2 were described as previously [18] , [22] . His-SUMO-1 and His-SUMO-2 were individually generated by ligating BamHI/HindIII and BamHI/EcoRI PCR fragments into the plasmid of pET-15b . His-KAP1 was generated by NdeI/BglII PCR fragment into pET-28a vector with NdeI/BamHI digestion . Briefly , mutants of full length LANA-myc ( ΔSIM ) , NC-myc ( ΔSIM ) , LANA-N3-myc ( ΔSIM1 , ΔSIM2 , ΔSIM , 3IA and 3SA ) , GFP-LANA-C1-myc ( K1081R ) , and - ( K1140R ) , FLAG-KAP1 with Lysine 779 ( K779 ) , Lysine 804 ( K804 ) , Lysine 779 , 804 ( K779 , 804 ) , Lysine 554 , 804 ( K554 , 804 ) or Lysine 554 , 804 , 779 ( K554 , 804 , 779 ) only were generated by PCR site-directed mutagenesis . All constructs were confirmed by direct DNA sequencing . Wild type FLAG-KAP1 and its 6KR mutant were kindly provided by Dr . Frank J Rauscher III from the Wistar Institute , Philadelphia [58] . The SUMO-1 ( Y299 ) , DNA-PKcs ( 18-2 ) , KAP1 ( 20C1 ) and β-actin ( #8226 ) antibodies were from Abcam . HIF-1α antibodies were from BD transduction laboratory . Sin3A ( AK-11 ) , SUMO-2/3 ( FL-103 ) , GFP ( F56-BA1 ) , p300 ( C-20 ) , and CBP ( A-22 ) were purchased from Santa Cruz Biotech . Inc . GAPDH ( G8140-01 ) antibodies were from US Biological Inc . . Mouse monoclonal antibodies against LANA ( LANA1 , or LNA1 rat mAb from Advanced Biotech Inc . ) , hemagglutinin ( 12CA5 ) , FLAG ( M2 ) , and myc ( 9E10 ) were used as described previously [20] . KSHV-negative ( BJAB ) and positive ( BC3 , BCBL1 ) B lymphoma cells were cultured in RPMI supplement with 7% FBS ( Hyclone ) . HEK293 cells and 293/Bac36 stable cell line ( established by 293 cells transfected with wild type KSHV BACmid [Bac36] and selected by 150 ng/ml Hygromycin B [Sigma] ) were maintained in Dulbecco's modified Eagle's medium ( DMEM ) containing 5% FBS ( Hyclone ) . Transfections were performed by electroporation with a Bio-Rad Gene Pulser in 0 . 4 cm-gap cuvettes at 220 Volts and 975 microfarads . Cells were grown in a humidified atmosphere at 37°C at gas tensions of 21% O2/5% CO2 for normoxic incubation and 1% O2 ( or 21% O2 with 100 µM CoCl2 ) /5% CO2 for hypoxic incubation as described previously [22] . Immunoprecipitation ( IP ) and immuno-blotting assays were performed as described previously [44] . For denatured IP , cells were boiled 10 min in 2% SDS-containing Tris-buffered saline ( TBS ) , followed by sonication and 18-fold dilution with TBS containing 1% Triton X-100 . The cell lysates were immunoprecipitated with specific antibody as indicated , followed by immunoblotting with specific antibodies . The membrane was stripped using stripping buffer ( 200 mM Glycine , 1% SDS , pH 2 . 5 ) for re-immunoblotting . Immunofluorescence assays were performed as previously described [18] . Briefly , cells were washed with ice -cold phosphate-buffered saline ( PBS ) and incubated on polylysine treated coverslips for 20 min ( HEK293 cells were directly grow on sterile coverslips and washed once with PBS ) and then fixed in 3% paraformaldehyde for 20 min at room temperature . After fixation , cells were washed three times in PBS and permeabilized in PBS containing 0 . 2% fish skin gelatin ( G-7765 , Sigma ) , 0 . 2% Triton X-100 for 5 min , and then followed the primary and secondary antibodies staining . DNA was counterstained with DAPI ( 4′ , 6′-diamidino-2-phenylindole ) , and Coverslips were mounted with p-phenylenediamine . Cells were visualized with a Fluoview FV300 ( Olympus Inc . , Melville , NY ) confocal microscope . Overnight starter cultures ( 50 ml ) of BL21 ( DE3 ) transformed with plasmid expressing GST , GST-fused protein , or His-fused protein were incubated into 500 ml of Luria Broth ( LB ) culture medium with specific antibiotic and grown at 30°C to an optical density of about 0 . 6 at 600 nm . After isopropylthiogalactopyranoside ( IPTG ) induction ( 1 mM , 4 hrs at 30°C ) , the bacteria were collected and sonicated in lysis buffer containing 20 mM Tris-HCl pH 8 . 0 , 100 mM NaCl , 0 . 5% NP40 , 1 mM EDTA , 1 M DTT , 5% Sarkosyl and the protease inhibitor cocktail for use with mammalian cell extracts or in vitro translated and radiolabeled proteins . His-SUMO-1 , His-SUMO-2 and His-KAP1 proteins were purified by Ni2+-NTA Agarose chromatography ( Qiagen ) , and GST , GST-SIM , GST-LANA1–329 , or GST-LANA1–329ΔSIM was purified by Glutathione Sepharose chromatography ( Amersham Biosciences ) . For pull-down assay , 35S-methionine-labeled in vitro-translated proteins or cell nuclear extracts were individually incubated with the relevant His or GST-fusion proteins loaded on beads for 3 hrs at 4°C in NETN binding buffer ( 50 mM Tris-HCl pH 7 . 5 , 100 mM NaCl , 10 µm ZnCl2 , 10% glycerol , freshly supplemented with 0 . 1 mM Dithiothreitol ( DTT ) and protease inhibitor ) . After washing , bound proteins were eluted with SDS sample buffer and analyzed by gel electrophoresis followed by direct autoradiography and scan by PhosphoImager ( Amersham Biosciences Inc . ) or immunoblotting with specific antibodies . Gel slices were excised and proteins subjected to tryptic digest followed by peptide identification by lc-ms/ms using a hybrid high resolution quadrupole time-of-flight electrospray mass spectrometer . Results were analyzed suing MASCOT database search tool ( Matrix Science ) . The KAP1 shRNA sequence ( 5′-GCATGAACCCCTTGTGCTG-3′ ) , Sin3A shRNA sequence ( 5′- CAACTGCTGA GAAGGTTGATTCTGT-3′ ) , and the control sequence ( 5′-TGCGTTGCTAGTACCAAC-3′ , non-targeting sequence ) , were individually inserted into pGIPz vector according to the manufacturer's instructions ( Clonetech ) . The pGIPz containing shRNA sequence was cotransfected with Lentivrus package expressing plasmids ( Rev , VSVG and gp ) into Core T cells by Calcium phosphate method to generate virus . The packaged viruses were used to individually transduce target cells ( BC3 , BCBL1 , or HEK293 ) and selection by 1–4 µg/ml puromycin . The RNA interfering efficiency was assessed by western blot analysis with specific antibodies . The chromatin immunoprecipitation ( ChIP ) experiments were done essentially as previously described with some modifications [63] , [64] . Cells ( 3×108 ) were cross-linked with 1 . 1% ( v/v ) formaldehyde , 100 mM NaCl , 0 . 5 mM EGTA , and 50 mM Tris-HCl ( pH 8 . 0 ) in growth medium at 37°C for 10 min , then at 4°C for 50 min . Formaldehyde was quenched by adding 0 . 05 vol 2 . 5 M glycine . Fixed cells were washed with PBS , incubated for 15 min in 15 ml of 10 mM Tris-HCl ( pH 8 . 0 ) , 10 mM EDTA , 0 . 5 mM EGTA , and 0 . 25% ( v/v ) Triton X-100 , followed by 15 min in 15 ml of 10 mM Tris-HCl ( pH 8 . 0 ) , 1 mM EDTA , 0 . 5 mM EGTA , and 200 mM NaCl , and finally sonicated in 1 ml of 10 mM Tris-HCl ( pH 8 . 0 ) , 1 mM EDTA , 0 . 5 mM EGTA , 1% ( w/v ) SDS plus 1 mM PMSF , 1 µg/ml aprotonin , leupettin , and pestatin ) to an average fragment size of 300–500 bp . Solubilized chromatin extracts were clarified by centrifugation at 12 , 000 g , and diluted to 6 OD260 U/ml in IP buffer [140 mM NaCl , 1% ( w/v ) Triton X-100 , 0 . 1% ( w/v ) sodium deoxycholate , 1 mM PMSF , 100 µg/ml salmon sperm DNA , and 100 µg/ml BSA]; pre-incubated for 1 h at 4°C with 10 µl/ml 50% ( v/v ) protein A-Agarose ( Invitrogen Life Technologies ) with normal mouse/rabbit sera; reconstituted in PBS , and washed several times in IP buffer . Aliquots ( 600 µl ) were incubated with 20 µg of each specific antibody for overnight at 4°C . Immune complexes were separated into bound and unbound complexes with protein A-agarose and cross-links were reversed by treatment at 65°C overnight . After treatment with RNase A and proteinase K , samples were extracted once with phenol/chloroform , and the DNA was precipitated with 2 volumes of ethanol plus 10 µg of glycogen as carrier ( Roche ) . Precipitated DNA was pelleted , washed once with 70% ethanol , dried , and resuspended in 100 µl of water . The DNA was analyzed by PCR using specific primer . For sequential ChIP , a second immunoprecipitation was performed using chromatin samples eluted from the agarose beads of the first ChIP by 10 mM DTT in 37°C for 30 min twice . Relative enrichment of DNA binding was shown as a percentage of input . Cells were resuspended in 4 times the volume of the pellet in NE buffer A ( 10 mM HEPES pH 7 . 9 , 10 mM KCl , 1 . 5 mM MgCl2 with protease inhibitors ) after cold PBS wash , and incubated on ice for 1 hour , and then transfer to precold douncer and homogenize with 25 strokes . Homogenized samples were transferred to eppendorf tubes and span at 2000 rpm for 5 mins at 4°C . Aspirate supernatant and resuspend the pellet in 2 times the volume of NE Buffer B ( 20 mM HEPES pH 7 . 9 , 10% glycerol , 420 mM NaCl , 1 . 5 mM MgCl2 , 0 . 2 mM EDTA with protease inhibitors ) . Incubate on ice for 30 mins and Centrifuge at 13000 rpm for 20 mins at 4°C . The supernatants were transferred to fresh eppendorf tube and add an equal volume of NE Buffer C ( 20 mM HEPES pH 7 . 9 , 30% glycerol , 1 . 5 mM MgCl2 , 0 . 2 mM EDTA with protease inhibitors ) . Aliquot and snap freeze samples at −80°C before use . Two complementary 5′-biotinylated oligonucleotides with or without HIF-1α-binding site ( wt: 5′-GGATTCCAAACGTGCCCAGCG GT-3′; mut: 5′-GGATTCCAATTAATCCCAGCGGT-3′ underline indicates HIF-1α-binding site ) were annealed to be double strands DNA and coupled to streptavidin-conjugated agarose beads . Per sample , 3 µg of biotinylated oligonucleotide was incubated with 40 ul of 50% streptavidin–conjugated agarose bead slurry ( Thermo scientific ) in a total volume of 100 ul of a lysis buffer comprised of 50 mM Tris-HCl , pH 8 . 0 , 15 mM NaCl , 0 . 1 mM EDTA , 10% glycerol , 10 mM N-ethylmaleimide ( NEM ) , 1 mM PMSF , 1 mM DTT , 1 µg/ml aprotonin , 1 µg/ml leupeptin , 1 µg/ml pepstatin for 2 h at 4C . Cell nuclear extracts ( 400 µg ) was incubated with 40 µl of DNA-coupled Agarose beads in lysis buffer at a total volume of 500 µl for 3 hr at 4C . The precipitated complexes were washed three times with lysis buffer . Purified DNA-binding proteins were boiled in SDS sample buffer and analyzed by SDS-PAGE and immunoblotting . Equally 5% input amount of biotinylated DNA oligonucleotide were verified by dot blot . Luciferase reporter plasmid pRta-luc ( pGL2 containing full length ( 1–3087 ) Rta promoter ) was used to detect the effect of SUMO-interacting domain of LANA on viral gene expression . The luciferase reporter assays were performed as described previously [44] . TK promoter driven Renilla luciferase was used as control to normalize the transfection efficiency . Total DNA was extracted by lysing buffer ( 10 mM Tris-HCl pH 8 . 0 , 150 mM NaCl , 10 mM EDTA , 1% SDS ) followed by proteinase K digestion . Relative numbers of KSHV episomal copies were calculated by quantitative PCR ( qPCR ) amplification of the terminal repeats ( TR ) ( primers see supplementary Table S2 ) as previously described [65] . For extracellular viral DNA , the viral particles were obtained by following the protocol for the virion purification . Bac36 DNA was used to the standards control . The equal input amount of PEL cells were subjected to induction for KSHV reactivation . After induction , the supernatant of culture medium was collected and filtered through 0 . 45 µm filter , and viral particles were spun down at 25 , 000 rpm for 2 h , at 4°C . The concentrated virus was collected and used for viral DNA quantitation .
|
Hypoxia stress is a common feature of tumor microenvironment and is widely associated with pathogenesis linked to many oncogenic viruses . Kaposi's sarcoma-associated herpesvirus ( KSHV ) , the etiological cause of Kaposi's sarcoma and primary effusion lymphoma , has been reported to encode several proteins that usurp hypoxia signaling during infection . One encoded protein LANA is a latent protein important for regulation of KSHV life cycle and cell transformation . The molecular mechanisms of how KSHV controls life cycle in normoxia and hypoxia is not fully understood . In this study , we show that LANA contains a unique SUMO-interacting motif ( LANASIM ) which is specific for SUMO-2 binding . Importantly , SUMO-2 modified KAP1 , a chromatin remodeling factor recruited by LANASIM is hypoxia sensitive , and plays a critical role in silencing viral gene expression . This discovery not only adds to our understanding of hypoxia-mediated remodeling of the viral episome in a SUMO dependent manner , but also provides a new dimension to development of therapeutic strategies against KSHV-associated cancers .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
A Unique SUMO-2-Interacting Motif within LANA Is Essential for KSHV Latency
|
Phlebotomine sand flies that transmit the protozoan parasite Leishmania differ greatly in their ability to support different parasite species or strains in the laboratory: while some show considerable selectivity , others are more permissive . In “selective” sand flies , Leishmania binding and survival in the fly midgut typically depends upon the abundant promastigote surface adhesin lipophosphoglycan ( LPG ) , which exhibits species- and strain-specific modifications of the dominant phosphoglycan ( PG ) repeat units . For the “selective” fly Phlebotomus papatasi PpapJ , side chain galactosyl-modifications ( scGal ) of PG repeats play key roles in parasite binding . We probed the specificity and properties of this scGal-LPG PAMP ( Pathogen Associated Molecular Pattern ) through studies of natural isolates exhibiting a wide range of galactosylation patterns , and of a panel of isogenic L . major engineered to express similar scGal-LPG diversity by transfection of SCG-encoded β1 , 3-galactosyltransferases with different activities . Surprisingly , both ‘poly-scGal’ and ‘null-scGal’ lines survived poorly relative to PpapJ-sympatric L . major FV1 and other ‘mono-scGal’ lines . However , survival of all lines was equivalent in P . duboscqi , which naturally transmit L . major strains bearing ‘null-scGal’-LPG PAMPs . We then asked whether scGal-LPG-mediated interactions were sufficient for PpapJ midgut survival by engineering Leishmania donovani , which normally express unsubstituted LPG , to express a ‘PpapJ-optimal’ scGal-LPG PAMP . Unexpectedly , these “L . major FV1-cloaked” L . donovani-SCG lines remained unable to survive within PpapJ flies . These studies establish that midgut survival of L . major in PpapJ flies is exquisitely sensitive to the scGal-LPG PAMP , requiring a specific ‘mono-scGal’ pattern . However , failure of ‘mono-scGal’ L . donovani-SCG lines to survive in selective PpapJ flies suggests a requirement for an additional , as yet unidentified L . major-specific parasite factor ( s ) . The interplay of the LPG PAMP and additional factor ( s ) with sand fly midgut receptors may determine whether a given sand fly host is “selective” or “permissive” , with important consequences to both disease transmission and the natural co-evolution of sand flies and Leishmania .
Leishmania are protozoan parasites that cause a spectrum of human diseases that range from self-healing cutaneous lesions to potentially fatal visceral forms . Leishmaniasis is re-emerging as a significant world health problem , with approximately 12 million people presently infected and 2 million new cases diagnosed each year ( www . who . int/leishmaniasis/burden ) . The world-wide distribution of different Leishmania is determined by the availability of transmission-competent sand fly vectors . When a sand fly bites an infected vertebrate host , Leishmania amastigotes residing within macrophages and other cell types are taken up in the blood meal which is surrounded by a midgut peritrophic matrix that lasts for several days . During this time amastigotes differentiate into motile , replicating promastigote forms which reside in the extracellular lumen of the sand fly alimentary tract ( rev . in [1] , [2] , [3] ) . Barriers to Leishmania development in this compartment include the chitin-containing peritrophic matrix which completely encases the blood meal , and the many hydrolytic enzymes and anti-microbial molecules secreted into the gut lumen ( rev . in [2] , [3] , [4] , [5] ) . Eventually the remnants of the digested blood meal are excreted by the sand fly , and this is a crucial juncture for Leishmania promastigotes . In transmission-competent sand flies , parasites attach to the midgut epithelium and go on to establish a stable infection; in a transmission-refractory vector , unattached parasites are expelled when the sand fly defecates ( rev . in [2] , [3] , [4] ) . As the sand fly prepares to feed again , promastigotes transition through several forms that culminate in infectious metacyclic parasites which express a modified surface that cannot bind to the midgut epithelium ( rev . in [1] , [3] , [4] , [6] ) . Thus , a key step in Leishmania transmission is stage-specific midgut attachment which allows Leishmania development to proceed . Two distinct mechanisms for regulating Leishmania attachment to sand fly midgut epithelium have been identified to date ( rev . in [1] , [6] , [7] , [8] ) . One mechanism , utilized in “selective” Phlebotomus papatasi sand flies that support the complete development of only a single Leishmania species , involves a sand fly midgut epithelium receptor that binds the parasite lipophosphoglycan ( LPG ) adhesin . LPG is an abundant glycolipid that covers the entire surface , including the flagellum , of all Leishmania promastigote stages [9] . The basic LPG structure is highly conserved in all Leishmania species , consisting of a glycosyl-phosphatidyl-inositol lipid anchor to which is attached a long polymer of 10–30 phosphoglycan ( PG ) repeating units ( 6Galβ1 , 4-Manα1-PO4 ) , and terminated by a small neutral oligosaccharide cap ( rev . in [10] , [11] ) . The PG repeating units are often modified by strain- , species- , and developmental stage-specific modifications that have been implicated in the midgut attachment and release of several Leishmania species in their respective natural vectors [12] , [13] , [14] , [15] . A second , LPG-independent sand fly midgut binding mechanism was recently identified using LPG-deficient Leishmania and several “permissive” sand flies that support the development of a broad range of Leishmania species in the laboratory [2] , [7] , [8] , [16] . While the precise binding modality is uncertain , the involvement of vector glycans has been suggested [8] , [16] . Leishmania phosphoglycans and/or other LPG2-dependent molecule ( s ) are also required for parasite survival in “permissive” sand flies [8] , [17] . Thus sugars , in the form of surface glycoconjugates , are key players in the productive interactions between Leishmania and sand flies that are necessary for disease transmission . This is in agreement with the general principle that important surface interactions between many microbes and their hosts involve complex glycoconjugates binding to receptors ( rev . in [18] ) . In this study we focused on the interactions between Leishmania major promastigotes and Phlebotomus papatasi sand flies . Phlebotomus papatasi is a “selective” vector which , despite its wide distribution in regions endemic for transmission of several Leishmania species , transmits only Leishmania major in nature and in the laboratory ( rev . in [1] , [2] , [3] , [4] , [6] . In this vector , specificity is controlled by a stage-specific modification in the LPG adhesin [6] , [12] . Midgut attachment is mediated by modified PG repeats bearing side chain β1 , 3 galactosyl residues ( scGal ) , which form the ligand recognized by the midgut LPG receptor PpGalec identified in Jordan Valley strain P . papatasi ( PpapJ ) sand flies [19] . As L . major procyclics develop into infectious metacyclic forms , procyclic form LPG is shed and replaced by metacyclic form LPG , which has increased numbers of PG repeats and scGal residues masked by the addition of terminal arabinose “caps”; these modifications block binding to PpGalec receptors [19] and facilitate detachment from the midgut [19] , [20] . Laboratory infections established the requirement for scGal-LPG in PpapJ midgut survival: L . major mutants or Leishmania species expressing scGal-deficient LPG , or lacking LPG entirely , could not establish stable PpapJ infections [12] , [16] , [21] , [22] , [23] , [24] and bound poorly to isolated PpapJ midguts and recombinant PpGalec receptors in vitro [19] . Notably , geographically diverse L . major strains express very different LPG side chain galactosylation patterns [which we refer to hereafter as scGal-LPG PAMPs ( Pathogen Associated Molecular Patterns ) ] , showing a Southwest-to-Northeast cline across its range from ‘null-scGal’ to ‘poly-scGal’ LPG PAMPs ( [23] , [25] , [26] , [27]; Cardoso et al . , in preparation} . For example , Senegalese strain SD procyclic LPG has such low levels of single βGal modifications that it is effectively unmodified [‘null-scGal’ LPG PAMP; [23] and this report} . In contrast , Israeli strain FV1 procyclic LPG is highly modified with primarily single βGal residues ( ‘mono-scGal’ LPG PAMP; [27] ) , and Central Asian strain LV39 clone 5 ( LV39c5 ) procyclic LPG is highly modified by long polymers of up to 8 βGal residues ( ‘poly-scGal’ LPG PAMP; [28] , [29] ) . Amongst these natural L . major strains , FV1 is sympatric with the “selective” P . papatasi PpapJ sand fly , while SD parasites are sympatric with P . duboscqi , a closely related sibling species of P . papatasi ( rev . in [3] ) . Previously , we hypothesized that different scGal-LPG PAMPs resulted from the combined activity of the seven telomeric SCG ( Side Chain Galactose ) gene family members , which encode PG-side chain-β1 , 3-galactosyltransferases . SCGs exhibit different activities , combining to varying extents ‘initiating’ activities able to attach the first βGal residue to the basic PG repeat , and ‘elongating’ activities able to add additional βGal residues to the initiated βGal side chain ( [30] , [31]; Dobson et al . , in preparation ) . In this work , we made use of this suite of diverse SCG activities to engineer isogenic parasites bearing defined scGal-LPG PAMPs . Using both naturally-occurring L . major strains and SD-SCG transfectant lines , we show that a specific scGal-LPG PAMP is optimal for long-term parasite survival in selective PpapJ sand flies , which preferred highly substituted scGal-LPG PAMPs bearing mono-galactosyl chains , neither “too short” nor “too long” . The “PpapJ-optimal” scGal-LPG PAMP was not sufficient , however , to enhance survival of L . donovani-SCG transfectants in PpapJ sand flies . These findings lead us to propose a two-component model for long-term Leishmania survival in “selective” PpapJ sand flies: 1 ) a specific scGal-LPG PAMP recognized by PpGalec midgut receptors and 2 ) an as yet unidentified L . major species-specific factor ( s ) .
Leishmania major strain Friedlin V1 ( FV1 ) is a clonal derivative of the Friedlin line ( MHOM/IL/80/Friedlin ) , L . major strain LV39 clone 5 ( LV39c5 ) is a clonal derivative of the LV39 line ( RHO/SU/59/P ) , L . major strain SD 75 . 1 ( SD ) is a clonal derivative of the NIH/SD line ( MHOM/SN/74/SD ) , L . donovani Sudanese strain 1S-2D clone Ld4 ( Ld ) is a clonal derivative ( MHOM/SD/00/1S-2D ) , and L . mexicana strain M379 is a clonal derivative ( MYNC/BZ/62/M379 ) . All wild type ( WT ) lines showed good infectivity in animal models and in their natural sand fly vectors [24] , [32] , [33] , [34] . Cells were grown in complete M199 medium containing 10% heat-inactivated fetal bovine serum , penicillin ( 50 units/ml ) , streptomycin ( 50 µg/ml ) , HEPES pH 7 . 4 ( 40 . 5 mM ) , adenine ( 0 . 1 mM ) , biotin ( 0 . 0001% ) , biopterin ( 2 µg/ml ) , and hemin ( 0 . 0005% ) , at 25°C as described [35] . Procyclic promastigotes were harvested from logarithmically growing cultures . Promastigotes were transfected by electroporation , using a low voltage [35] or high voltage [36] protocol . Clonal lines were obtained by plating on semisolid M199 media containing the appropriate selective drug concentration: 50 µg/ml hygromycin B ( HYG ) , 20 µg/ml phleomycin ( PHLEO ) , 15 µg/ml G418 ( NEO ) , or 100 µg/ml nourseothricin ( SAT ) . L . major strain FV1 was the source of all SCG genes used in this study . Episomal expression constructs used here include the cosmid vector cLHYG ( strain B890; [37] ) , pXK ( NEO ) -SCG2 ( B3900; [30] ) , SCG3 cosmid B3979 [30] , and pXG ( NEO ) -LMSAP1 ( B3092; [34] ) . The integrating SCG open reading frame ( ORF ) constructs pIR1SAT-SCG1 ( B5097 ) , pIR1SAT-SCG3 ( B5101 ) , pIR1SAT-SCG4 ( B5103 ) , and pIR1SAT-SCG5 ( B5170 ) were created as follows: SCG ORFs liberated by BamHI digestion of appropriate pXG ( PHLEO ) -SCG ORF constructs ( Dobson et al . , in preparation ) were ligated into the BglII expression site of pIR1SAT ( B3541; [36] ) . Each pIR1SAT-SCG construct was digested with SwaI restriction enzyme , dephosphorylated with calf alkalline phosphatase , and gel-purified to yield linear SSU::IR1SAT-SCG ORF targeting fragments for integration into the ribosomal RNA small subunit ( SSU ) locus by homologous recombination during transfection [36] . Integrated SCG transfectants are referred to by the gene name and location , i . e . SD-SSU:SCG3 has the FV1 strain SCG3 ORF ( SSU::IR1satSCG3 ) integrated into the SD ribosomal SSU locus . We used three Ld-transfectant lines developed previously [30]: a ‘null-scGal’ LPG PAMP line devoid of any LPG side chain modification , transfected with the episomal cosmid vector cLHYG ( Ld-vector ) ; and two different lines exhibiting ‘mono-scGal’LPG PAMPs , one transfected with SCG3 cosmid B3979 ( Ld-cSCG3 ) , and a second transfected with the SCG2 ORF expression construct pXK-SCG2 ( Ld-pSCG2 ) . LPG was prepared from exponentially-growing promastigotes as described [38] . To assess side chain modifications , phosphoglycan repeats were depolymerized using mild acid hydrolysis , dephosphorylated using E . coli alkaline phosphatase , covalently labeled with 1-aminopyrene-3 , 6 , 8-trisulfonate , and analyzed by Dionex HPLC chromatography [14] or capillary electrophoresis [39] , comparing migration distances with oligomeric glucose standards . Procyclic promastigotes ( 5×105/ml ) were grown in complete M199 medium to a final density of 1×107/ml . Culture supernatants were collected and centrifuged for 10 min at 2500 rpm in a Sorvall RT7000 centrifuge to remove cells and debris . 10 microliter samples of clarified culture supernatant were electrophoresed on a non-denaturing polyacrylamide gel and the gel then stained for SAP enzyme activity using α-naphthyl acid phosphate plus Fast Garnet GBC as described [40] , [41] . SAP was quantitated using the AlphaImager version 5 . 5 gel documentation system spot densitometry program ( Alpha Innotech , San Leandro , CA ) . Sand fly colonies were reared at the Division of Entomology , Walter Reed Army Institute of Medical Research and at the Laboratory of Parasitic Diseases , National Institute of Allergy and Infectious Diseases , NIH . The following species were used in this study: Phlebotomus papatasi from colonies originating from the Jordan Valley ( PpapJ ) , Phlebotomus duboscqi from colonies originating from Mali ( PdubM ) , and Phlebotomus argentipes from colonies originating from India ( PargIN ) . Female 3- to 5-day-old sand flies were fed through a chick skin membrane using a feeding device containing a mixture of heparin-treated mouse blood and logarithmic phase promastigotes , as described [24] . The concentration of promastigotes used varied depending on the experiment , from 1×106 to 20×106 parasites/ml . Blood-engorged sand flies were separated and maintained at 28 C with 30% sucrose ( v/v ) . At various times after feeding , flies were anesthetized , their midguts dissected and homogenized , and the number of released midgut promastigotes counted using a hemocytometer as described [24] . Parasite numbers in the midguts of infected flies after blood meal excretion do not follow a Gaussian distribution . This is likely the result of flies within groups having either completely lost their infections or retained parasites that grow exponentially prior to the time of dissection . Therefore , data sets were compared using a nonparametric Mann Whitney test . Mann Whitney calculations were done using Prism 4 ( Graphpad Software , Inc . San Diego , CA ) .
LPGs from three geographically distinct L . major strain procyclic promastigotes were purified , subjected to mild acid hydrolysis and dephosphorylation , and isolated PG repeat structures assessed by capillary electrophoresis ( Methods , Table S1 ) . Side-chain galactosylation can be characterized by two parameters: the fraction of PG repeats that were modified , and the number of βGal residues attached . In these studies we found two general patterns of LPG side chain galactosylation: one in which little or no βGal was added; and a second in which 50–90% of the PG repeats were modified , with varying numbers of βGal residues . From these data we found it useful to calculate a single parameter for comparisons amongst lines , the “average scGal chain length” , obtained by multiplying the fraction of modified PG repeats times the average number of βGal residues added per modified repeat ( Tables 1 , S1 ) . Senegalese strain SD LPG was mostly unmodified ( 0 . 02 avg . scGal chain length ) , consistent with prior studies using specific antisera and lectins suggesting that SD LPG was largely unmodified [23] . In contrast , Israeli strain FV1 LPG was extensively modified with predominantly single βGal residues ( 0 . 8 avg . scGal chain length ) . Central Asian strain LV39c5 LPG was also highly modified , but with longer polymers of up to 8 βGal residues ( 3 . 1 avg . scGal chain length ) . We refer to these three prototypic LPG galactosylation patterns as ‘null-scGal’ , ‘mono-scGal’ , and ‘poly-scGal’ LPG PAMPs ( Pathogen-Associated Molecular Patterns ) , respectively . The results with FV1 and LV39c5 confirmed and extended previous studies [27] , [28] , [29] , and were undertaken to guard against changes in LPG side chain composition occurring during laboratory propagation , as described previously [25] , [42] . PpapJ sand flies were fed on the indicated L . major-infective mouse blood and midgut infections were assessed 48 hr later , a time when parasites remain within the blood meal encased by the peritrophic membrane ( Fig . 1A , “+ blood , d2” ) . At this time all three L . major strains showed high parasite numbers in most flies examined ( >33 , 000 parasites/midgut ) , with the highest numbers observed in flies infected with the SD strain , likely reflecting the faster generation time of this strain . Thus differences in the scGal-LPG PAMPs did not affect the early survival and growth of L . major promastigotes , as expected since even LPG-deficient parasites survive normally in sand flies during this interval [16] , [22] , [24] . By day 5 post-feeding , the sand fly peritrophic matrix disintegrates and the remains of the digested blood meal are expelled . At this time there were clear differences amongst the L . major strains in their ability to survive ( Fig . 1A , “no blood , d5” ) . In agreement with previous studies [22] , [24] , ‘mono-scGal’ FV1 persisted in most PpapJ flies at high levels ( 82% flies infected , 16200±16600 parasites/midgut ) . In contrast , ‘poly-scGal’ LV39c5 survived poorly ( 53% flies infected , 3860±4840 parasites/midgut; p<0 . 013 ) , as did ‘null-scGal’ SD , with the exception of two strongly-infected outliers ( 38% flies infected , 6660±18600 parasites/midgut; p<0 . 005 ) . The poor survival of ‘null-scGal’ SD was expected , as un-galactosylated LPG cannot bind to midgut PpGalec receptors [19] , resulting in unattached parasites being excreted with the digested blood meal remnants [21] , [24] . However , the poor PpapJ survival of ‘poly-scGal’ LV39c5 ( Fig . 1A , B; [24] ) suggested that a specific scGal-LPG PAMP , rather than simply the presence of galactosylated LPG , controls L . major promastigote survival in PpapJ midguts following blood meal expulsion . Since the three L . major strains studied here show an average nucleotide sequence divergence of 0 . 15% [43] , comparable to that amongst many L . major strains , molecular differences other than scGal-LPG PAMPs were potentially responsible for the survival differences we observed in selective PpapJ sand fly infections . To generate different scGal-LPG PAMPs in an isogenic scGal-deficient LPG background , we introduced into the SD line a series of constructs expressing members of the previously characterized SCG family of telomeric phosphoglycan-side chain- ( β1 , 3 ) galactosyltransferases ( PG-scβGalTs ) [30] , [31] Critical to these studies is the fact that SCG-encoded PG-scβGalTs have different enzymatic properties mediating the addition of different numbers of scGal residues , ranging from 0 to 12 ( [30] , [31]; Dobson et al . , in preparation ) . Thus , SD promastigotes were transfected with different SCG constructs , using either the episomal pXG-type vector which expresses passenger ORFs at moderate levels [37] , episomal cosmids identified previously bearing SCG genes [30] , or the integrating pIR1SAT vector which expresses passenger ORFs at high levels following integration into the ribosomal RNA small subunit ( SSU ) locus [36] . LPGs were purified from SD transfectants and LPG galactosylation patterns determined as described above ( Methods ) ; from these studies we chose a key set of SD-SCG lines exhibiting a range of scGal-LPG PAMPs ( Tables 1 , S1 ) briefly summarized here . SD transfectants bearing an integrated catalytically inactive SCG5 ORF ( SD-SSU:SCG5 ) synthesized scGal-deficient LPG indistinguishable from the parental WT SD line ( ‘null-scGal’ LPG PAMP; 0 . 02 avg . scGal chain length ) . Two SD transfectants expressed ‘mono-scGal’ LPG PAMPs: SD-cSCG3 ( 0 . 9 avg . scGal chain length ) , containing the episomal SCG3 cosmid B3979; and SD-SSU:SCG3 ( 1 . 3 avg . scGal chain length ) , containing an integrated SCG3 ORF ( SSU::IR1SAT-SCG3 ) . SD-SSU:SCG4 transfectants bearing an integrated SCG4 ORF ( SSU::IR1SAT-SCG4 ) synthesized a ‘poly-scGal’ LPG PAMP ( 3 . 1 avg . scGal chain length ) . A novel ‘oligo-scGal’ LPG PAMP ( 1 . 9 avg . scGal chain length ) was synthesized by SD-SSU:SCG1 , which bears an integrated SCG1 ORF ( SSU::IR1sat-SCG1 ) . Together these SD-transfectant scGal-LPG PAMPs spanned the natural range of L . major LPG side chain variation as well as providing new LPG galactosylation patterns for study . To confirm that SD transfectants had not experienced a general non-specific loss of “sand fly virulence” during their generation and propagation in the laboratory , we examined their survival in two independent infections involving Phlebotomus duboscqi PdubM sand flies originating from Mali ( Fig . S1 A , B ) . P . duboscqi is a sibling species of P . papatasi , and PdubM flies are able to support the full development of WT SD in the laboratory [23] . Although L . major survival in P . duboscqi is LPG-dependent , it is not strongly affected by scGal-LPG PAMPs since various ‘null-scGal’ , ‘mono-scGal’ , or ‘poly-scGal’ L . major strains have been shown to survive expulsion of the digested blood meal [8] , [17] , [22] , [44] . Female PdubM sand flies were allowed to feed on the indicated L . major-infective mouse blood containing ‘null-scGal’ ( WT SD , SD-SSU:SCG5 ) , ‘mono-scGal’ ( WT FV1 , SD-SSU:SCG3 ) , ‘oligo-scGal’ ( SD-SSU:SCG1 ) , or ‘poly-scGal’ ( SD-SSU:SCG4 ) promastigotes . As expected , all PdubM flies were successfully infected with high numbers of parasites at early time points ( Fig . S1 , “+ blood” panels ) . Following expulsion of the digested blood meal , PdubM flies infected with all L . major lines retained high numbers of midgut parasites ( Fig . S1 “no blood” panels ) and each line went on to establish fully mature infections in the PdubM anterior midgut by day 12 post-feeding ( data not shown ) . These data argue against a general non-specific loss in the ability of SD-SCG transfectants to survive in the phlebotomine sand fly midgut environment . PpapJ flies were allowed to feed on the indicated L . major-infective mouse blood containing SD transfectants expressing different scGal-LPG PAMPs . The results from two independent experiments are shown ( Fig . 2A , B ) . At early times post-infection when the midgut blood meal was retained , all SD transfectants behaved similarly: 100% of PpapJ flies were infected with high numbers of promastigotes , similar to control ‘mono-scGal’ WT FV1 infections ( Fig . 2A , B , “+ blood , d2” ) . However , we observed clear differences in PpapJ midgut survival amongst SD lines expressing different scGal-LPG PAMPs after the digested blood meal had passed out of the midgut ( Fig . 2A , B , “no blood , d5”; Tables 1 , S2 ) . First , and as expected , SD-SSU:SCG5 transgenic parasites expressing a ‘null-scGal’ LPG PAMP survived poorly following blood meal excretion , with a 81–92% decrease in mean parasite numbers relative to control FV1-infected flies ( p<0 . 0005 ) , and 38–40% of PpapJ flies having lost their infections ( Fig . 2A , B , “no blood , d5” ) . Second , SD transfectants expressing ‘mono-scGal’ LPG PAMPs ( SD-cSCG3 , SD-SSU:SCG3 ) generally survived well post-blood meal expulsion ( Fig . 2A , B , “no blood , d5” ) . Most flies remained infected , and mean SD-cSCG3 and SD-SSU:SCG3 parasite numbers ( 18600±4732 , 18900±4900 and 18800±5100 , 13200±4800 parasites/midgut , respectively ) were not significantly different from control WT FV1 infections ( 13500±3500 , 30000±4400 parasites/midgut ) . By contrast , ‘mono-scGal’ SD-cSCG3 survival was significantly enhanced relative to ‘null-scGal’ SD-SSU:SCG5 ( 2560±1720 , p<0 . 0004 and 2420±600 parasites/midgut , p<0 . 003; Fig . 2A , B ) . Although ‘mono-scGal’ SD-SSU:SCG3 survival was also enhanced relative to ‘null-scGal’ SD-SSU:SCG5 , this difference reached significance in only one experiment ( p<0 . 0002 and p<0 . 264; Fig . 2A , B ) . Third , SD-SSU:SCG1 transfectants expressing a novel ‘oligo-scGal’ LPG PAMP survived poorly . Only 37–55% of SD-SSU:SCG1 flies remained infected post-blood meal expulsion and parasite levels ( 4120±1940 , 12500±6300 parasites/midgut ) were significantly reduced relative to control WT FV1-infected flies ( p<0 . 018 and p<0 . 06; Fig . 2A , B ) . In fact , ‘oligo-scGal’ SD-SSU:SCG1 survival was not significantly better than observed for ‘null-scGal’ SD-SSU:SCG5 ( p<0 . 968 and p<0 . 954; Fig . 2A , B ) . Lastly , and consistent with the results from natural isolates , SD-SSU:SCG4 transfectants expressing a ‘poly-scGal’ LPG PAMP survived poorly . Only 42–56% of SD-SSU:SCG4 flies remained infected and parasite levels ( 683±215 , 2267±831 parasites/midgut ) were significantly decreased , 92–95% relative to control WT FV1 infections ( p<0 . 0007 and p<0 . 0001 , Fig . 2A , B ) . Thus ‘poly-scGal’ SD-SSU:SCG4 PpapJ survival was not significantly better than ‘null-scGal’ SD-SSU:SCG5 parasites . These findings are summarized in Fig . 3 , showing the relationship between relative PpapJ survival post-blood meal expulsion and the average scGal chain length in purified procyclic promastigote LPG . Isogenic SD-SCG transfectants whose LPG closely approximates the ‘mono-scGal’ LPG PAMP of the WT FV1 line ( i . e . SD-cSCG3 , SD-SSU:SCG3 ) clearly survived well in PpapJ sand flies . In contrast , isogenic SD transfectants expressing either scGal-deficient LPG ( ‘null-scGal’ SD-SSU:SCG5 ) or LPG with longer side chain polymers ( ‘oligo-scGal’ SD-SSU:SCG1 , ‘poly-scGal’ SD-SSU:SCG4 ) survived poorly in PpapJ flies , mirroring infection outcomes with naturally-occurring L . major strains SD or LV39c5 ( ‘null-scGal’ or ‘poly-scGal’ LPG PAMPs , respectively ) . Together , these data firmly implicate the scGal-LPG PAMP causally in controlling the ability of PpapJ flies to support L . major midgut survival post-blood meal expulsion . The studies above established that a ‘mono-scGal’ LPG PAMP was necessary for L . major survival in selective PpapJ sand flies , following blood meal expulsion . We next asked whether this scGal-LPG PAMP would be sufficient , by examining its effect on the PpapJ survival of a different Leishmania species . We chose L . donovani Sudanese strain 1S-2D ( Ld ) since these parasites possess unmodified LPG ( ‘null-scGal’ LPG PAMP; [11] , [45] ) and have been shown to survive poorly in PpapJ sand flies [12] , [24] . We used three Ld-transfectant lines developed previously [30]: a ‘null-scGal’ line devoid of any side chain sugars ( Ld-vector , 0 avg . scGal length ) and two different lines exhibiting ‘mono-scGal’ LPG PAMPs , Ld-cSCG3 and Ld-pSCG2 ( 0 . 7 and 1 . 1 avg . scGal chain length , respectively; Tables 1 , S1 ) . When PpapJ sand flies were fed on L . donovani-infective mouse blood containing ‘null-scGal’ Ld-vector or ‘mono-scGal’ Ld-cSCG3 promastigotes , all flies were successfully infected with comparably high numbers of parasites when examined at a time when the midgut blood meal was present ( Fig . 4A , PpapJ “+ blood , d3” panel ) . Thus , these parasites were able to survive well in the initial steps of sand fly infection . However , following expulsion of the blood meal at day 5 post-feeding , parasites from both of these lines were completely lost in >90% of PpapJ flies , and those flies retaining infections had very low levels of parasites ( 180 and 125 parasites/midgut respectively; Fig . 4A , PpapJ “no blood , d5”; Table S3 ) . Thus , despite generation of the optimal highly substituted ‘mono-scGal’ LPG PAMP in the Ld-cSCG3 line , survival in PpapJ was not enhanced ( Table 1 ) . As a control , these Ld transfectants were fed to P . argentipes PargIN , a natural vector of Ld transmission originating from India [12] , [24] . Previous studies have shown that midgut survival of both L . donovani and L . major in this “permissive” sand fly species is not strongly affected by LPG galactosylation patterns [8] , [12] , [24] . Due to the limited number of PargIN flies available for analysis , a single infection time point was analyzed comparing flies without blood meal remnants in the midgut on day 5 post-feeding . In contrast to the loss of midgut infections in PpapJ flies , both Ld-vector and Ld-cSCG3 promastigotes persisted and were maintained a moderate infection intensity in most PargIN flies after the digested blood meal was expelled ( 88% or 78% infected flies; 11263 or 6822 parasites/midgut; Fig . 4A PargIN “no blood , d5” , Table S3 ) . These data argue against a general non-specific loss in the ability of these Ld transfectants to survive in the phlebotomine sand fly midgut environment . In separate experiments PpapJ flies were fed on Leishmania-infective mouse blood containing ‘null-scGal’ Ld-vector , ‘mono-scGal’ Ld-pSCG2 , or control ‘mono-scGal’ WT L . major FV1 promastigotes ( Fig . 4B , C ) . As expected , most PpapJ flies were infected with high numbers of parasites prior to expulsion of the blood meal , although the numbers of Ld-vector and Ld-pSCG2 were significantly less than control WT L . major FV1 ( Fig . 4B “+ blood , d3” panel ) . However , in PpapJ flies that had expelled their blood meal , neither Ld-vector nor Ld-pSCG2 survived ( 0% infected flies ) , whereas good survival was seen with the WT FV1 control ( 100% infected , 13200 parasites/midgut; Fig . 4B “no blood , d4” panel; Table S2 ) . When PpapJ sand flies were infected with a 4-fold higher concentration of parasites to compensate for the diminished early growth of Ld transfectants compared to WT FV1 , we again observed poor survival of both ‘mono-scGal’ Ld-pSCG2 and ‘null-scGal’ Ld-vector parasites after the midgut blood meal had been expelled ( Fig . 4C “no blood , d5” panel ) , despite massive parasite loads in midguts that retained their blood meals at day 3 post-feeding ( Fig . 4C “+ blood , d3” panel ) . Ld-vector and Ld-pSCG2 numbers were each significantly decreased relative to control WT FV1 ( >90% , p<0 . 001 ) , although a higher percentage of PpapJ flies remained infected ( 38% of Ld-vector , 52% of Ld-pSCG2 , 100% of WT FV1; Fig . 4C “no blood , d5” panel ) . These results are consistent with early observations regarding the ability of high concentration of promastigotes in the artificial blood meal to overcome the natural resistance of P . papatasi to infection with L . donovani [12] , [24] , [46] . Together , these data suggest that while necessary for survival and transmission of L . major in “selective” PpapJ sand flies , the ‘mono-scGal’ LPG PAMP alone is not sufficient to rescue L . donovani-SCG promastigotes in PpapJ sand flies during the critical time of blood meal expulsion . Unlike L . major , L . donovani and most other Leishmania species secrete high levels of acid phosphatases ( SAPs ) covalently modified by PG repeats [47] , [48] , [49] . Since PG repeats attached to SAP bear the same covalent side chain modifications as LPG PG repeats [29] , [41] , [50] , ‘mono-scGal’ SAP could potentially compete for Ld-cSCG3 and Ld-pSCG2 promastigote binding to PpapJ midgut PpGalec receptors , thereby accounting for their failure to survive expulsion of the digested blood meal . To test this hypothesis , we engineered L . major FV1 to express high levels of SAP ( Methods ) . High levels of active SAP were detected in the culture medium of all FV1-SAP transfectant lines , more than 1100 times higher than SAP levels in WT FV1 or control FV1-vector transfectant culture media and thus comparable to SAP levels secreted by L . donovani and other Leishmania species promastigotes ( Table S4 ) . However , in two independent experiments involving infections of PpapJ flies with L . major-infective mouse blood , FV1-SAP1 over-expressors survived as well as control FV1-vector promastigotes , both prior to and after expulsion of the digested blood meal ( Fig . 5A , B “+ blood , d3” and “no blood , d4” panels , respectively ) . These data argue that competition with ‘mono-scGal’ SAP is unlikely to account for the poor PpapJ survival of Ld-SCG transfectants expressing the ‘mono-scGal’ LPG PAMP preferred by this sand fly .
As noted earlier , many workers have grouped sand fly species according to their ability to support in experimental infections the survival ( and , in some cases , experimental transmission ) of a wide versus limited range of Leishmania species [7] , [8] , [12] , [16] , with the former group termed “permissive” sand flies and the latter termed “selective” or “restricted” . The availability of Leishmania mutants specifically defective in LPG ( through the deletion of the gene encoding the LPG-specific galactofuranosyltransferase LPG1 ) has shown that in general , “selective” sand fly species show a strong role for LPG in midgut survival and binding , while the “permissive” sand fly species show little LPG dependency [7] , [8] , [12] , [16] , [17] , [24] . Our panel of engineered and natural L . major , varying greatly in scGal-LPG modification , allowed us to compare the effects seen in a “selective” sand fly , P . papatasi PpapJ from the Jordan Valley , which showed a strong preference for ‘mono-scGal’ LPG PAMPs ( Figs . 1–3 ) . Recently , we have completed studies of more than 15 L . major isolates that reveal a range in the extent of procyclic promastigote scGal-LPG modification , with a general cline proceeding from scGal-deficient ‘null-scGal’ LPG modification in West Africa to short chain ‘mono-scGal’ modification in the Middle East to long chain ‘poly-scGal’ modification in Central Asia ( Cardoso et al . , in preparation ) . Together with the findings presented here , the stage is now set for further explorations of the role of scGal-LPG PAMPs in L . major transmission in other natural settings . Since one natural P . papatasi sand fly vector in this geographic range showed differing abilities to support Leishmania growth which were dependent on scGal-LPG PAMPs ( Figs . 1–3 ) , it seems likely these may play an important role and perhaps even a driving force in the evolution of parasite/vector selectivity . For example , all Israeli L . major lines whose LPG has been characterized show ‘mono-scGal’ LPG PAMPs ( V121 strain , avg . 1 . 1 scGal length; L580 strain , avg . 0 . 7 scGal length; calculated from data in [25] , [27] ) and correspondingly , the ability of a P . papatasi PpapJ colony established from wild caught flies from the Jordan Valley to support L . major midgut survival is strongly dependent on this scGal-LPG PAMP . In this respect it will be interesting to examine the properties of P . papatasi sand flies from Central Asia , including potential structural diversity in their PpGalec midgut LPG receptor , as L . major from this region typically elaborate a ‘poly-scGal’ LPG PAMP similar to that of LV39c5 ( Cardoso et al . , in preparation ) . Our work demonstrating a geographical origin-based specificity between PpapJ sand fly vector and L . major strains also complements the work of Elfari et al . [51] who demonstrated evidence for genetic and biological diversity in L . major strains that correlated with geographical origin and their ability to infect only sympatric animal reservoir hosts . While expression of appropriate scGal-LPG PAMPs is necessary for the survival of L . major in the PpapJ sand fly midgut , is it sufficient ? We tested this by engineering the ‘mono-scGal’ LPG PAMP into a Sudanese strain of L . donovani which normally expresses a completely unmodified LPG coat [45] . We showed by biochemical analyses and agglutination tests ( Table S1 , [30] , [31] ) that the engineered scGal-LPG PAMPs in L . donovani-SCG transfectants were faithful replicas of L . major ‘mono-scGal’ LPG PAMPs synthesized by natural WT L . major FV1 and engineered SD-SCG3 transfectants , all of which exhibited robust long-term survival in PpapJ laboratory infections ( Tables 1 , S2 ) . However , L . donovani-SCG lines bearing a ‘mono-scGal’ LPG surface coat remained unable to survive following expulsion of the blood meal in infected PpapJ flies ( Fig . 4; Tables 1 , S2 , S3 ) . We then explored several possible mechanisms that could account for the failure of L . donovani bearing an L . major FV1 LPG “surface” to survive . First was the possibility that secretion of scGal-modified acid phosphatases ( SAPs , [47] , [48] , [50] ) competed for LPG-dependent midgut binding and parasite survival . While SAP-deficient L . donovani are not available , reconstruction experiments in L . major FV1 promastigotes expressing high levels of PG-modified SAPs ( Fig . 5 , Table S4 ) failed to reveal any alterations in PpapJ survival . Thus , competition by L . donovani scGal-SAP is unlikely to account for the failure of Ld-SCG promastigotes to survive in PpapJ midguts . A second reason was that the engineered ‘mono-scGal’ L . donovani were unable to withstand PpapJ midgut conditions , since early killing of L . donovani promastigotes in the P . papatasi midgut has been reported [52] . In fact , in comparison to the sympatric L . major FV1 strain , the L . donovani lines showed reduced growth in the early blood fed midgut ( Fig . 4B ) , due either to their slower generation times , and/or their greater sensitivity to midgut digestive enzymes . Nevertheless , when the differences in the concentration of parasites present prior to blood meal excretion were overcome by initiating infection with a high dose inoculum , the L . donovani lines were still largely absent in flies that had passed their blood meals ( Fig . 4C ) . Furthermore , L . donovani transfectants were able to survive within the midgut of P . argentipes PargIN sand flies ( Fig . 4A ) . Importantly , survival in this sand fly species cannot be attributed simply to a more permissive midgut environment , as P . argentipes restricts survival of lpg2- Ld lines which lack LPG and other PGs , evidence of a strongly hydrolytic midgut environment [8] , [24] . These data argue that the inability of WT or engineered L . donovani lines to survive in PpapJ sand flies is not due to an inability to withstand the midgut environment , and the timing of the loss of infection is consistent with their failure to attach to the midgut . Thus , while specific scGal-LPG PAMPs are necessary for L . major persistence and midgut binding during expulsion of the blood meal in PpapJ flies , the inability of L . donovani expressing the appropriate L . major scGal-LPG PAMP to survive in the same fly strain suggests most simply that this interaction , while necessary , is not sufficient for midgut attachment . This in turn would argue that an additional parasite ligand ( s ) must be required , one shared in the closely related L . major strains but lacking in L . donovani , which diverged from L . major >80 million years ago [53] . In this model , generation of proper scGal-LPG PAMPs in L . major SD would be sufficient to promote survival , since L . major strains would retain this second L . major-specific interaction; but insufficient in L . donovani , where the second interaction was absent due to evolutionary divergence or loss . In contrast to Ld-SCG transfectants , which “inherited by transfection” only the scGal-LPG-dependent ligand , the enhanced P . papatasi survival of L . infantum - L . major hybrids observed by Volf et al . ( relative to L . infantum; [54] ) is thus predicted to result from the inheritance of both L . major-specific scGal-LPG-dependent and -independent ligands . Whether this postulated second interaction is mediated through a second species-specific receptor for LPG , or an LPG-independent ligand such as the one proposed by Myskova et al . to control midgut binding of certain Leishmania species in “permissive” sand fly vectors [8] , [16] , is unknown . Perhaps the PpapJ ‘mono-scGal’ LPG midgut receptor PpGalec collaborates with a co-receptor , similar to the interactions of certain other pattern recognition receptors such as Toll-like receptors ( TLR1/2/6 ) with each other or with other receptors ( Dectin-1 , CD14 , TLR4; reviewed in [55] , [56] . This putative species-specific co-receptor may be especially relevant to the interaction of L . major strains with P . duboscqi sand flies . This vector , while unable to support the survival of L . major lines completely deficient in LPG biosynthesis [8] , [22] , [44] , is not sensitive to differences in L . major LPG galactosylation patterns ( Fig . S1 ) and naturally transmits L . major strains in West Africa bearing effectively ‘null-scGal’ LPG . Nonetheless , P . duboscqi is a “selective” vector , permitting only the development of L . major in experimental infections ( [8] , Sacks et al . , unpublished ) . These data suggest that the few interactions between predicted P . duboscqi PpGalec midgut LPG receptors [19] and the low number of mono-galactosylated PG repeats in WT SD LPG ( 2% , Table S1 ) is sufficient to mediate parasite attachment to the PdubM midgut epithelium , in concert with a second L . major-specific midgut binding interaction that is especially strong in this particular sand fly species . It is also possible that a scGal-independent ligand present on L . major LPG binds to the alternative receptor and is a sufficient interaction to maintain infection in the PdubM vector . When the factor ( s ) controlling parasite LPG-independent binding and survival of Leishmania in “selective” and “permissive” sand fly species becomes known , it should be possible to test these hypotheses .
|
Phlebotomine sand flies are tiny blood-feeding insects that transmit Leishmania protozoan parasites , which cause diseases afflicting millions of people . The world-wide distribution of Leishmania is determined by the availability of transmission-competent vectors . In the laboratory , some vectors support many different Leishmania , while others are highly restricted . This is best exemplified by P . papatasi , which transmit only L . major despite a wide distribution in regions endemic for many Leishmania species . P . papatasi “selectivity” can be reproduced experimentally , and has been attributed to β1 , 3-linked galactose side chains decorating the abundant L . major surface lipophosphoglycan ( LPG ) adhesin , which mediate parasite attachment to the P . papatasi midgut to prevent elimination when the digested blood meal is excreted . As geographically diverse L . major display very different LPG galactosylation patterns ( n = 0 - 8 βGals/side chain ) , we explored the consequences of this pattern diversity to survival in P . papatasi . Using natural isolates and L . major lines engineered to express a wide range of LPG galactosylation patterns , we showed L . major survival in P . papatasi PpapJ flies was optimized by expression of highly modified ‘mono-galactosylated’ LPG and extremely sensitive to LPG side chain length . Surprisingly , L . donovani lines engineered to express a “PpapJ-optimal” LPG mono-galactosylation pattern did not survive in PpapJ flies , suggesting that additional interactions are required . These studies reveal the fine specificity of Leishmania - sand fly interactions , and the nature of species- and strain-specific parasite molecules that have co-evolved to take advantage of midgut receptors specific to available sand fly vectors .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"evolutionary",
"biology/microbial",
"evolution",
"and",
"genomics",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"microbiology/parasitology",
"infectious",
"diseases/protozoal",
"infections",
"cell",
"biology/cell",
"adhesion"
] |
2010
|
Leishmania major Survival in Selective Phlebotomus papatasi Sand Fly Vector Requires a Specific SCG-Encoded Lipophosphoglycan Galactosylation Pattern
|
Ethnic diversity has been long considered as one of the factors explaining why the severe forms of dengue are more prevalent in Southeast Asia than anywhere else . Here we take advantage of the admixed profile of Southeast Asians to perform coupled association-admixture analyses in Thai cohorts . For dengue shock syndrome ( DSS ) , the significant haplotypes are located in genes coding for phospholipase C members ( PLCB4 added to previously reported PLCE1 ) , related to inflammation of blood vessels . For dengue fever ( DF ) , we found evidence of significant association with CHST10 , AHRR , PPP2R5E and GRIP1 genes , which participate in the xenobiotic metabolism signaling pathway . We conducted functional analyses for PPP2R5E , revealing by immunofluorescence imaging that the coded protein co-localizes with both DENV1 and DENV2 NS5 proteins . Interestingly , only DENV2-NS5 migrated to the nucleus , and a deletion of the predicted top-linking motif in NS5 abolished the nuclear transfer . These observations support the existence of differences between serotypes in their cellular dynamics , which may contribute to differential infection outcome risk . The contribution of the identified genes to the genetic risk render Southeast and Northeast Asian populations more susceptible to both phenotypes , while African populations are best protected against DSS and intermediately protected against DF , and Europeans the best protected against DF but the most susceptible against DSS .
Dengue virus ( DENV ) is the most common mosquito-borne viral infection , infecting approximately 390 million people per year worldwide with one quarter developing dengue disease ( MIM: 614371 ) [1] . Symptoms range from undifferentiated fever , classical dengue fever ( DF ) to shock syndrome ( DSS; hemorrhage , plasma leakage and vital organ impairment ) [2] . Recent -omic approaches provide unbiased genomic insights into mechanisms associated with dengue disease . There has been only one publication on classical genome wide association study ( GWAS ) of dengue [3] compared to a considerable number of transcriptomic studies [4–7] . The reason for this discrepancy is that cohorts of thousands of individuals are required for GWAS to reach genome wide significance . The GWAS work conducted on a cohort of Vietnamese children [3] included 2 , 008 DSS samples versus 2 , 018 controls , replicated in 1 , 737 versus 2 , 934 , and found SNPs in genes MICB and PLCE1 associated with DSS phenotype . Lately , analytical improvements based on admixture mapping have reduced the sample size requirement from thousands to hundreds of individuals or even fewer [8] . Most human populations have some degree of ancestry admixture , which brings together haplotypes that occur at different frequencies in parental populations . Admixture mapping analyses these blocks across the mosaic descendant chromosomes and allows to compare their distribution between case and control cohorts . The lower number of blocks compared with individual SNPs reduces considerably the statistical burden . We have successfully conducted such an admixture study in dengue cohorts from Cuba [9] , and identified two genes involved in lipid metabolism which showed to be protective against the risk of dengue hemorrhagic fever , a protection conferred by the African inherited ancestry . Whereas for RXRA gene there was already functional evidence of its involvement in infection [10] , we also demonstrated functionally by shRNA that the knockdown of OSBPL10 gene had a significant negative impact in DENV replication rate [9] . Epidemiologic reports have shown the existence of ethnic differences in susceptibility to dengue fever not only in Cuba [11] but also in Malaysia [12] where the incidence rate by ethnic group was 3 . 7:1:1 . 3 for Chinese , Malays and Indians , respectively , in the years 1970’s and 1980’s , although no cross-evaluation was performed with other socio-demographic factors . In the present study , we take advantage of the admixed profile of Southeast Asians ( in the nexus between South , Northeast and Southeast Asia ) to perform coupled association-admixture analyses ( BMIX; [13] ) of case/control cohorts of dengue patients: Thai dengue patients who developed DF ( n = 252 ) or DSS ( n = 159 ) , and a control blood donor group ( n = 290 ) ; and the published Vietnamese dataset ( 2018 controls and 2008 DSS patients;[3] ) . Although the admixture in the region has been taking place along time , since the first arrival of modern human after the out-of-Africa migration , a considerable migration from south China began in the 15th century and increased in the 19th and 20th centuries , mainly towards Thailand where about 40% of the population has some Chinese admixture and 14% are identifiable Thai Chinese [14] . This is a similar scenario to the admixture that took place in the Americas , where these local admixture inference tools have been successfully applied [9 , 15 , 16] . We were able to identify distinct candidate genes conferring susceptibility/resistance to the risk of DF and DSS , arguing in favor of independent pathogenic mechanisms for the establishment of the two phenotypes . We further confirmed that one DF candidate gene codes for a human protein that co-localizes with the DENV1 and DENV2-NS5 proteins , and , in the latter case , transiently relocated from the cytoplasm to the nucleus . We also inferred the relative worldwide genetic risks contributed by the detected candidate genes based on their frequencies for the susceptible/resistant haplotypes .
All analyzed individuals have some degree of admixture ( Fig 1; S1 Fig ) . The Northeast Asian background is dominant in Vietnam ( 77 . 3% ) and decreases in Thailand ( 56 . 4% ) , in contrast to the Southeast Asian component , which increases from 20 . 7% in Vietnam to 35 . 1% in Thailand . The South Asian influence is 8 . 5% in Thailand and 2 . 0% in Vietnam . Within the dengue cohorts , we observed a statistically significant increase in the Southeast Asian background in Thailand for both DF ( 4 . 1% increase; p-value = 1 . 25 x 10−7 ) and DSS ( 4 . 8% increase; p-value = 5 . 90 x 10−8 ) compared to Thai control . We began by checking if the BMIX results on the published Vietnamese cohort [3] are in accordance with the results from the classical association mapping , a test of the robustness of the algorithm . BMIX indicates also the association of DSS with MICB and PLCE1 genes ( Table 1 , Fig 2A , S1 and S4 Tables ) . The identified region surrounding MICB encompasses seven significant SNPs , placed along 165 , 080 bp , from the downstream MICA to the upstream LTB gene , a region highly rich in genes . Three linked ( S2 Fig ) SNPs in MICB have the most significant p-values , forming the protective haplotype GTT ( OR = 0 . 77; p-value<0 . 0001 ) , which is the most frequent haplotype in worldwide populations ( Fig 3C ) . The susceptible MICB haplotype ACC ( OR = 1 . 39; p-value<0 . 0001 ) is more frequent in Europeans and South Asians ( 0 . 18 to 0 . 34 ) . The two SNPs found for PLCE1 reached significant p-values and are almost in complete linkage ( S5 Fig ) . The DSS protective PLCE1 haplotype ( CG; OR = 0 . 75; p-value<0 . 0001 ) is more frequent ( Fig 3B ) in Northeast Asia ( 0 . 12–0 . 28 ) and Southeast Asia ( 0 . 19 ) , followed by Europe ( 0 . 04–0 . 14 ) and absent in Africa . We further analyzed the Thai DSS vs . control cohort ( Table 1 , Fig 2B , S2 and S5 Tables ) , and obtained a reliable signal of six linked ( S3 Fig ) significant SNPs for PLCB4 ( phospholipase C , beta 4; S4A Fig ) , a gene in the same family as PLCE1 , and participating in many common pathways , such as dendritic cell maturation , PI3K signaling in B lymphocytes and PPARA/RXRA activation . The DSS protective PLCB4 haplotype ( GAGAGG; OR = 0 . 58; p-value = 0 . 013 ) is rare in most worldwide populations ( Fig 3A ) , reaching the highest frequencies in Africa ( 0 . 21–0 . 28 ) . Only one PLCE1 SNP ( rs2274223 ) was present in the chip used in the Thai dengue cohort and it did not reach significance . Individually , the conventional association study with PCA correction for population stratification in Thai DSS vs . control could not identify any candidate gene when correcting for multiple test ( S6A Fig–the two singled out significant SNPs are spurious signals as linked SNPs do not display significant p-values ) . We also tested 10 runs of pseudo datasets , permutating case and control labels ( S1 Text ) . No SNP is significant in the association tests , and the BMIX algorithm identified spurious significant SNPs ( mostly isolated in different chromosomes ) that do not replicate between runs and that are different from the case-control comparison . Overall posterior p-values were also lower in the pseudo datasets . The higher spurious detections in BMIX than in the association test agree with the fact that the statistical burden of the local ancestry test is considerably lower than the one for the association test , which raises the possibility of detecting a positive signal . The randomness between runs reflects the high variability between individuals in admixture percentages and in distribution of ancestry blocks along the genomes . This argues for a double-careful interpretation of BMIX results in the context of the disease . For the Thai DSS vs . control , the fact that the PLCB4 gene belongs to the same family of the previously independently identified PLCE1 gene is an important additional evidence for considering that gene a strong candidate in DSS phenotype . Calculating the genetic risk of DSS according to the worldwide population frequency of the phospholipase C and MICB protective and susceptible haplotypes ( Fig 4A ) , it can be observed that African and descendent Caribbean populations are best protected , while European , Asian and Latin American populations are more susceptible to DSS . We genotyped two of the BMIX-identified significantly associated PLCB4 SNPs in further Thai control ( n = 244 ) and case ( n = 20 ) samples ( S7 Table ) , and the rs1997696 SNP presents a p-value over the significance threshold of a traditional GWAS ( p = 4 . 7x10-8 ) . When comparing Thai DF vs . control , a distinctive genetic signature was obtained . Three genes located on different chromosomes had at least two SNPs above the BMIX significant posterior probability threshold of 0 . 5 ( Fig 2C ) , forming haplotypes ( S7 , S8 and S9 Figs ) . CHST10 codes for carbohydrate sulfotransferase 10 ( S4C Fig ) , has nine significant SNPs ( Table 1 , S3 and S6 Tables ) , forming the protective haplotype CTCTACGGG ( OR = 0 . 59; p-value = 0 . 0005 ) , whereas the haplotype TGTCGTAAT increased risk of DF ( OR = 1 . 78; p-value<0 . 0001 ) . The protective haplotype is frequent in South Asian populations ( 0 . 38–0 . 57 ) , whereas the susceptible haplotype is frequent in Northeast Asia ( 0 . 61–0 . 74 ) and very rare in the African populations ( Fig 3D ) . AHRR ( S4D Fig ) codes for aryl-hydrocarbon receptor ( AHR ) repressor , has two significant SNPs , and similarly to CHST10 , the protective AHRR haplotype ( TC—OR = 0 . 54; p-value = 0 . 0001 ) is more frequent in South Asian and African populations ( between 0 . 40–0 . 60 ) and the opposite haplotype ( CT—OR = 1 . 89; p-value<0 . 0001 ) is more frequent in Northeast Asian populations ( 0 . 57–0 . 78 ) ( Fig 3E ) . PPP2R5E ( S4F Fig ) codes for protein phosphatase 2 ( PP2A ) , regulatory subunit B' , epsilon isoform ( also known as PP2A-B56 ) , has four significant SNPs , whose protective haplotype ( ACGG—OR = 0 . 62; p-value = 0 . 0002 ) showed high frequency in South Asian populations ( 0 . 76–0 . 86 ) , while African populations have the lowest frequency of this haplotype ( 0 . 09–0 . 20 ) ( Fig 3G ) . Interestingly , the proteins coded by these three genes , and by another gene , GRIP1 ( S4E Fig ) that codes for glutamate receptor interacting protein 1 , with one significant SNP ( T; OR = 0 . 56; p-value = 0 . 0016 ) , are involved in the xenobiotic metabolism signaling pathway ( Fig 5D ) . The GRIP1 protective allele is more frequent in South Asian populations ( 0 . 19–0 . 32 ) and absent in Africa ( Fig 3F ) . Again , individually , the conventional association study with PCA correction for population stratification could not identify any candidate gene when correcting for multiple test ( S6B Fig ) . The contribution to the genetic risk to DF , inferred from an additive model combining the protective and susceptible haplotypes of the four xenobiotic-related genes ( Fig 4B ) , indicates highest protection in European and South Asian populations and highest risk in Northeast and Southeast Asians . African and Latin American/Caribbean populations have an intermediate risk conferred by these genes to DF . The genotyping of six SNPs in these four genes in additional Thai controls ( n = 245 ) and cases ( n = 55 ) improves p-values of a traditional association test in the total cohort to levels of 10−5 in four SNPs and 10−4 in two SNPs ( S7 Table ) . These values are significant after Bonferroni correction for the set-test of six SNPs . We analyzed the expression of these genes in the xenobiotic pathway in a transcriptome dataset including patients sampled during acute phase of DF , DHF and convalescence compared with controls [17] . CHST10 and AHRR expressions did not significantly change during dengue infection ( S10 Fig ) , however , there was a significant increase in PPP2R5E expression and a significant decrease in AHR ( negatively regulated by AHRR ) and GRIP1 expressions during acute dengue infection ( Fig 5A , 5B and 5C ) . These findings are further evidence that PPP2R5E , GRIP1 and AHR can be involved in dengue infection and development of dengue disease . We further checked in the GTEx database if the DF candidate SNPs act as eQTLs . All candidate protective alleles in PPP2R5E and AHRR genes significantly reduce the expression of the respective proteins ( S11 Fig ) . The candidate SNP in GRIP1 gene is not an eQTL in the GTEx cohort , and the two eQTLs ( rs11176317 and rs12322014 ) close to the candidate rs1480010 are not in LD with it . As the GTEx cohort is mainly of European ancestry , we cannot ascertain if this GRIP1 SNP or other linked SNPs can be eQTLs in Asian populations . The recent identification of conserved motifs that provide binding specificity to the PP2A-B56 phosphatase [18] led us to further test the hypothesis of the potential binding of this regulatory region of PP2A protein to DENV proteins . We began by performing an in silico search [19] for the high-affinity LxxIxE motif as well as the intermediate- and low-affinity motifs in the protein reference sequences of the four DENV serotypes ( S8 Table and Fig 6A ) . NS5 presents between three and six motifs in all four DENV serotypes , and at least two of these motifs ( LxxIxE and LxxVxE ) are highly conserved . Other viral proteins also bear motifs , but are more heterogeneous between DENV serotypes . We then tested the hypothesis that PP2A-B56 can interact with NS5 by conducting confocal immunofluorescence co-localization tests . We transfected Huh7 cells with a mammalian expression plasmid containing DENV2-NS5 tagged with an orange fluorescence protein . We fixed and stained with antibody against PPP2R5E at 24h , 48h and 72h after transfection . In non-transfected cells , PPP2R5E is localized in the cytoplasm ( Fig 6B ) . At 24h of post-transfection , both PPP2R5E and NS5 are localized in the cytoplasm , but by 48h they both co-localize in the nucleus , and at 72h PPP2R5E returns to the cytoplasm while NS5 remains in the nucleus ( Fig 6C ) . We then deleted the xLxxIxE motif in our DENV2-NS5 vector ( Fig 6D ) and transfected cells in the same way . The deletion of this motif prevented the translocation of the viral NS5 protein to the nucleus ( Fig 6D ) . Testifying to the existence of differences between serotypes , the immunofluorescence co-localization test between PPP2R5E and DENV1-NS5 ( Fig 6E ) showed that the two proteins co-localize in the cytoplasm , but the entrance in the nucleus is almost negligible , and little accumulation of NS5 can be detected in the nucleus at 72h .
Our successful association-admixture analyses in Thai population have provided evidence that different genes/pathways contribute to the genetic susceptibility or resistance to different outcome of dengue infection . We suggest that xenobiotics and lipid metabolism , as well as interaction of viral proteins to these molecules and to its phosphatases , are critical in the development of classical DF , whereas more severe forms of dengue are caused by over reactive immunity leading to cytokine storm and/or defect in endothelial cell dysfunction and coagulation system . We reinforced the association of phospholipase C gene family with the DSS phenotype in Thai patients , as had been found in Vietnamese DSS patients [3] , but this time the gene detected was PLCB4 instead of PLCE1 . These enzymes have been implicated in a high number of signal transduction pathways [19] and on immune regulation [20] . More recently , Lin et al . [21] found PLCB4/PLCB1 susceptibility loci for coronary artery aneurysm in Kawasaki disease when analyzing a Han Chinese cohort . During the acute stage of this disease , inflammation occurs by the infiltration of T cells and macrophages and the activation of vascular endothelium cells ( ECs ) with increased serum proinflammatory cytokines and predominant damage of small- , medium-sized vessels and the coronary artery . The injured vascular tissues show subendothelial edema , vascular damage , gap formation , and fenestration of ECs . It thus seems that phospholipase C genes are involved in several diseases presenting the phenotype of inflammation of the blood vessels , as it is the case in dengue shock syndrome . The association of the four genes ( CHST10 , AHRR , PPP2R5E and GRIP1 ) from the xenobiotic metabolism signaling pathway with Thai DF patients is the first genetic evidence for its implication in dengue pathogenesis , although functional studies have already indicated this association before . Xenobiotics are toxic non-endogenous compounds that together with other toxic endogenous compounds must be eliminated from the body by drug/xenobiotic metabolizing enzymes ( DME/XME ) and transporters . The DMEs are induced by their own substrates , through signaling cascades involving three specific receptors: the constitutive active receptor ( CAR ) , the pregnane X receptor ( PXR ) and the aryl hydrocarbon receptor ( AHR ) ( Fig 5D ) [22] . In particular PP2A regulates the CAR:HSP90 complex , allowing CAR release and its eventual translocation to the nucleus [23] , an event that could be similar/parallel to PP2A involvement in the nuclear translocation of viral proteins . Several lines of evidence show that some of PP2A regulatory ( such as PP2A-B56 ) and scaffold subunits can bind viral proteins [24–26] . In addition , the capsid of West Nile virus , a flavivirus related to DENV , binds to the inhibitor of PP2A proteins ( I2PP2A or SET ) in the precise site where I2PP2A binds to PP2A , causing an increase of PP2A activity in several cell types [27] . These evidences led us to perform immunofluorescence assays that revealed that PPP2R5E co-localizes with the NS5 protein of DENV2 , first in the cytoplasm and then in the nucleus , and that this nuclear translocation does not take place when the specific link motif is deleted from the NS5 protein . NS5 is a crucial viral protein responsible for the virus replication at the endoplasmic reticulum . NS5 accumulation in the nucleus seems to occur late in DENV infection as a hyperphosphorylated form unable to bind NS3 . Adding a layer of complexity to these events , the test of the NS5 nuclear translocation in the four DENV serotypes showed that it only occurs in DENV2 and DENV3 [28] , leading the authors to hypothesize that the NS5 nuclear localization is not strictly required for virus replication but that it is more likely to have an auxiliary function in the life cycle of specific DENV serotypes . Our in silico analysis indicates that all NS5 proteins from the four DENV strains contain the specific link-motifs , but our immunofluorescence assay with DENV1-NS5 confirmed the observation of the authors , that the protein of this serotype does not enter the nucleus . Extra factors , such as importin proteins , must contribute to this differential import into the nucleus . Thus , the dynamics of NS5 seems to play a major role in dengue infection , potentially impacting differential strain virulence , and the PPP2R5E-NS5 interaction must be taken into consideration in future studies . Increased PP2A activity could favor viral infection not only through binding of viral proteins but also through regulation of regulatory T cells ( Treg ) [29] . Suppression and impairment of anti-viral activity of interferon α ( IFNα ) has been shown in hepatitis C infection through inhibition of Jak1/Tyk2/STAT1 phosphorylation [30] and upregulation of PP2A dependent upon NS5A protein [24] . Both Treg and Jak1/Tyk2/STAT1 pathway have been shown to be important in DENV infection [31–34] . The significant association detected for AHRR and lower expression of AhR during acute dengue infection suggests another possible key factor in dengue infection outcome . AhR has been shown to be involved in mediating the biotransformation and carcinogenic/teratogenic effects of environmental toxins . Recently , its role in innate and adaptive immunity has been demonstrated , through its involvement in regulation of CD4 , CD8 and Treg after viral infection [35] . We hypothesize that the outcome of dengue infection depends on a fine-tuning of the xenobiotic metabolism signaling pathway between the AhR and PP2A/CAR pathway . While increased PP2A/CAR activity promotes viral infection , decreased AhR activity results in uncontrolled immune homeostasis leading to dengue disease . The confirmation that our candidate protective alleles in PPP2R5E and AHRR genes are eQTLs leading to lower expression of the respective proteins supports this hypothesis . When cells are invaded by DENV , there will be an increase in the expression of PPP2R5E , possibly due to the interaction with NS5 . So Asian individuals that are genetically PPP2R5E-low-expressing are protected against this link . For the AHRR gene , the protective phenotype also has a lower expression profile , and as this protein will inhibit AHR protein , these individuals will have a higher expression of AHR , which is opposite to the expression pattern observed in the dengue patients transcriptome . For both DF and DSS phenotypes , Northeast and Southeast Asian populations have a higher ancestral prone risk when compared with other geographical regions , considering the particular genes identified in this work . Specifically , Southeast Asian ancestry has a slightly higher risk for DF than Northeast Asian ancestry ( Fig 4B ) . These genetic predictions agree with observations that almost 75% of the global population exposed to dengue live in Asia-Pacific , with rates of severe dengue being 18 times higher in this region compared with the Americas [36] . African and its descendant populations are the most protected ones against DSS , and displaying an intermediate protection against DF , adding genetic evidence to previous claims that this ancestry is protected against worse dengue phenotypes [37 , 38] . Our inferred genetic risk for DF in Africa , slightly higher that the risk in America , agrees quite well with the risk predictions inferred by Bhatt et al . [1] of 16% and 14% , respectively , of the global burden . Climatic change and globalization are enlarging the spread of dengue vector and virus to northern latitudes , putting Europe and North America at risk of autochthonous infections [39] . The considerable number of autochthonous infections that occurred in Madeira Island , Portugal , in 2012/2013 [40] is the first example of a reality that can take place in a near future in continental Europe . The genetic risk calculated here , for the newly and confirmed susceptible/resistant haplotypes , shows that European populations ( as well as South Asian and USA ) present an even higher risk than Southeast Asian populations to DSS , while they are the best protected ones against DF .
We enrolled 411 patients ( age range , 1–31 years; male to female ratio , 0 . 984 ) with symptomatic DENV infection during 2000–2003 from two hospitals in Bangkok ( Ramathibodi , and Siriraj ) and one in Khonkaen , Thailand ( S1 Fig ) . These patients were first admitted to the hospitals with suspicions of dengue infection based on clinical features , and DENV infection was later confirmed by either detection of viral genome or a comparable immunoglobulin G ( IgG ) and immunoglobulin M ( IgM ) titers , measured by an enzyme-linked immunosorbent assay , in late acute and/or convalescent sera . Dengue severity was defined according to 1997 World Health Organization ( WHO ) criteria [2] , and we ended up with the following case cohorts: 252 patients with DF with no evidence of plasma leakage but incapacitating dengue fever; 159 with severe plasma leakage and/or bleeding , leading to shock or profound shock ( grades 3 and 4 ) . Information about the DENV serotype and primary/secondary infection are provided in S9 Table . These case samples were genotyped with the Illumina Human 660W Quad BeadChip ( Illumina , San Diego , CA , USA ) . The control cohort was collected in the same hospitals , geographically matching the cases , consisting in 290 healthy individuals with no fever and being treated for minor injuries . Control individuals were genotyped with the Illumina HumanOmniExpress BeadChip . Quality control was performed in PLINK [41] , and SNPs with more than 5% missing genotypes , minor allele frequency ( MAF ) below 5% , and Hardy-Weinberg equilibrium ( HWE ) deviation p-values of less than 0 . 001 were filtered out from downstream analyses . We also checked visually for outliers in principal-component analysis ( PCA; S12 Fig ) , and excluded samples that had evidence of being a second-degree relative or closer to another sample in the study ( identity by descent >30%; or identity by state >90% ) . All studied samples passed these criteria . SNPs located in X and Y chromosomes and in mitochondrial DNA were removed from the analyses , leading to a final account of common 261 , 660 autosomal SNPs . The Vietnamese cohort consisted in 2018 controls and 2008 DSS patients typed with the Illumina Human 660W Quad BeadChip , amounting in 479 , 905 SNPs after QC . Eight SNPs detected in the BMIX analyses were genotyped in additional Thai cohort samples ( 61 DF; 20 DSS; and 250 controls ) , through TaqMan assays ( Life Technologies , Carlsbad , CA , USA ) with commercial probes for the SNPs . The screen was conducted on a QuantStudio 12K Flex ( Life Technologies , Carlsbad , CA , USA ) and the results were analyzed in TaqManGenotyper software ( Life Technologies , Carlsbad , CA , USA ) . Written informed consent was obtained from all subjects or , in case of individuals under 18 years of age , from their parents or tutors . The protocol was approved by the ethics committees from the Faculty of Medicine , Ramathibodi Hospital , Mahidol University; the Faculty of Medicine , Siriraj Hospital , Mahidol University; the Khon Kaen Hospital; and the Thailand Ministry of Public Health . The tests were conducted in the following comparisons: Thai DF versus controls ( DF test ) ; Thai DSS versus controls ( DSS test ) ; and Vietnamese DSS versus controls ( DSS test ) . Samples were phased in SHAPEIT v . 2 [42] using HapMap reference panel and fine-scale genetic map . For admixture mapping , we applied the RFMix algorithm [43] and used three ancestral data sets: the phased data from the 1 , 000 Genomes Database [44] for the Chinese Dai in Xishuangbanna ( CDX ) ; and the Indian Telugu from the UK ( ITU ) representing the Northeast and South Asian ancestries , respectively; and the Malaysian complete genomes from Singapore [45] representing the Southeast Asian ancestry . We checked if the global admixture profile inferred by RFMix would be reproduced when using ADMIXTURE analysis [46] for K = 3 running together the parental populations and Thai and Vietnamese cohorts . As can be verified in S13 Fig , the three-background admixture profiles for the Thai and Vietnamese individuals are identical between RFMix and ADMIXTURE analysis . ITU population is quite homogeneous , while CDX and Malaysians have themselves around 20–30% admixture , but this does not affect the overall proportion inference of the three components . Information on the three ancestry backgrounds was obtained for each locus along chromosomes for every individual , and these values were averaged in each cohort . Two-tailed Mann-Whitney test ( non-parametric test , not requiring normal distribution ) was applied to assess the significance between the global ancestry proportions inferred for the three ancestral backgrounds within the Thai and Vietnamese cohorts . BMIX [13] was implemented on the Thai and Vietnamese groups , by coupling the admixture mapping inferred through the RFMix algorithm with the association data . SNPs with a posterior probability of a joint ancestry and association effect equal or higher to 0 . 5 were considered significant . The annotation of the significant SNPs was inferred by using the Variant Effect Predictor ( VEP ) tool from Ensembl . The involvement of the significant genes in pathways was checked in the Ingenuity Pathway Database ( https://targetexplorer . ingenuity . com/index . htm ) . Odds ratios ( ORs ) , 95% confidence intervals , beta parameters and Yates p-values ( corrected for continuity ) for the significant haplotypes/SNPs in DF and DSS phenotypes were calculated in http://vassarstats . net/odds2x2 . html . The dengue phenotype ( DF or DSS ) genetic risk score was calculated by multiplying each individual’s significant haplotype/allele ( either protective or causative ) count for each locus by the respective beta coefficient and summing the product for all loci ( as in [47] ) . For DF , the four genes ( CHST10 , AHRR , PPP2R5E and GRIP1 ) participating in the xenobiotic metabolism signaling pathway were considered . For DSS , both phospholipase C ( PLCB4 and PLCE1 ) and MICB genes were included in the calculation . Expression of candidate genes was checked in a Thai whole blood transcriptome cohort [17] measured with the Human U133 Plus 2 . 0 Arrays ( Affymetrix , Santa Clara , CA , USA ) , including: nine healthy controls samples; 28 samples collected between days 2 and 9 after onset of symptoms ( acute illness ) from secondarily infected patients ( 18 DF and 10 DHF ) ; and 19 samples collected at convalescence , four weeks or later after discharge . Huh7 liver cell line was cultured in DMEM ( Life Technologies , Carlsbad , CA , USA ) with 10% bovine serum and penicillin-streptomycin and maintained at 37°C in 5% CO2 and used for three transfection assays: wild type DENV2-NS5; LxxIxE-deleted DENV2-NS5; and wild type DENV1-NS5 . The DENV2-NS5 was fused with an orange fluorescent protein and cloned into pCMV3-C-His vector ( DENV-2 ( strain New Guinea C–GenBank: AF038403 ) NS5 open reading frame ( ORF ) ( 2700bp ) mammalian expression plasmid; SinoBiological , Beijing , China ) . The wild-type DENV2-NS5 clones were mutagenized with the Q5 Site-Directed Mutagenesis Kit ( New England Biolabs , Ipswich , MA , USA ) , following the manufacturers’ protocol . DENV1-NS5 ( isolate KDH0026A , Kamphaeng Phet Provincial Hospital , Thailand–GenBank: HG316481 ) was also fused with OFP and cloned into the same vector using overlapping primers and the HiFi DNA Assembly Protocol ( New England Biolabs , Ipswich , MA , USA ) , according to the manufacturers’ recommendations . The designed primers are described on S10 Table , and the LxxIxE-deletion and DENV1-NS5 assembly were confirmed by Sanger sequencing performed in a 3130xl Genetic Analyzer ( Applied Biosystems , Foster City , CA , USA ) . Transient transfections were performed using Lipofectamine 3000 reagent ( Invitrogen ) according to the manufacturer’s protocol . Wildtype and mutated DENV2-NS5 proteins expressions were confirmed in ZOE Fluorescent Cell Imager ( Bio-Rad , Hercules , CA , USA ) . Transfected cells were harvested 24h , 48h and 72h after transfection and fixed with 4% paraformaldehyde . PPP2R5E was tagged with a primary rabbit anti-PPP2R5E antibody ( Atlas Antibodies , Bromma , Sweden ) and revealed with a secondary goat anti-rabbit Alexa Fluor 488 antibody ( Thermo Fisher Scientific , Waltham , MA , USA ) . Imaging was obtained on a TCS SP5 II ( Leica , Wetzlar , Germany ) Laser Scanning Confocal microscope . Images ( green and red channel ) were aligned with Huygens Software ( Chromatic Aberration Corrector module ) , by using 0 . 2 μm TetraSpeck Microspheres ( Thermo Fisher Scientific , Waltham , MA , USA ) embedded in the same conditions as the samples . The images were merged and adjusted for brightness and contrast with Fiji software [48] .
|
Dengue fever is endemic in tropical and subtropical areas of East Asia and America , but globalization and climate changes are introducing vector and virus to the naïve regions of Europe and North America . In this work we conducted a statistically robust , coupled association-admixture test in two dengue cohorts from Thailand ( classical dengue fever , DF , and dengue shock syndrome , DSS ) and a published Vietnamese ( DSS only ) cohort . We identified new candidate genes associated with DF risk and confirmed known gene family association with DSS risk . In DF , phosphatase control is crucial , including through binding to viral proteins , as we showed for PPP2R5E protein co-localization with DENV1 and DENV2-NS5 proteins within liver cells and differential cellular localizations along time . In DSS , cytokine dynamics , inflammation and activation of vascular endothelium cells are dominant features . The particular genetic risk conferred by these genes indicates that Southeast and Northeast Asians are highly susceptible to both phenotypes , while Africans are best protected against DSS , and Europeans best protected against DF but the most susceptible against DSS .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"xenobiotic",
"metabolism",
"tropical",
"diseases",
"geographical",
"locations",
"genetic",
"mapping",
"ethnicities",
"genetic",
"predisposition",
"genome",
"analysis",
"neglected",
"tropical",
"diseases",
"sequence",
"motif",
"analysis",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"infectious",
"diseases",
"genomics",
"bioinformatics",
"dengue",
"fever",
"metabolic",
"pathways",
"biological",
"databases",
"thai",
"people",
"metabolism",
"people",
"and",
"places",
"biochemistry",
"haplotypes",
"heredity",
"database",
"and",
"informatics",
"methods",
"genetics",
"population",
"groupings",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"computational",
"biology",
"europe",
"genetics",
"of",
"disease",
"genomic",
"databases"
] |
2018
|
Joint ancestry and association test indicate two distinct pathogenic pathways involved in classical dengue fever and dengue shock syndrome
|
Bacteria regulate many phenotypes via quorum sensing systems . Quorum sensing is typically thought to evolve because the regulated cooperative phenotypes are only beneficial at certain cell densities . However , quorum sensing systems are also threatened by non-cooperative “cheaters” that may exploit quorum-sensing regulated cooperation , which begs the question of how quorum sensing systems are maintained in nature . Here we study the evolution of quorum sensing using an individual-based model that captures the natural ecology and population structuring of microbial communities . We first recapitulate the two existing observations on quorum sensing evolution: density-dependent benefits favor quorum sensing but competition and cheating will destabilize it . We then model quorum sensing in a dense community like a biofilm , which reveals a novel benefit to quorum sensing that is intrinsically evolutionarily stable . In these communities , competing microbial genotypes gradually segregate over time leading to positive correlation between density and genetic similarity between neighboring cells ( relatedness ) . This enables quorum sensing to track genetic relatedness and ensures that costly cooperative traits are only activated once a cell is safely surrounded by clonemates . We hypothesize that under similar natural conditions , the benefits of quorum sensing will not result from an assessment of density but from the ability to infer kinship .
Microbes use quorum sensing to regulate a large number of phenotypes . During growth , cells secrete autoinducers , which are small , diffusible compounds that accumulate in the environment . High autoinducer concentration around cells induces expression of many metabolically costly traits , which includes secretions that promote the growth of surrounding cells [1–6] . The canonical explanation for the function of quorum sensing is that autoinducer concentrations can be used as a proxy for cell density . Estimating the density of cells in a given environment allows microbes to cooperate in a coordinated manner . Cells can tune the expression of density-dependent phenotypes , like virulence factors or secreted enzymes , so that they are only expressed when there are enough cells to make them useful [1 , 4 , 7 , 8] . The benefits of quorum sensing can also be affected by diffusion conditions [9–11] , which may favor the evolution of multiple quorum sensing signals [12] . Quorum sensing strains then may outcompete strains that constitutively express cooperative phenotypes . However , this does not guarantee quorum sensing evolution because there is also competition from strains that do not cooperate at all , often known as “cheater” genotypes [4 , 13–16] . For example , Pseudomonas aeruginosa cells with a defective lasR gene preventing them from responding to quorum sensing signals , can outcompete cooperating wild-type cells [4 , 14] . This is because the mutants have a higher growth rate since they do not produce the costly cooperative secretions , but can benefit from the secretions of wild-type cells . Quorum sensing then has the potential to be evolutionarily unstable in mixed genotype cultures [4 , 7 , 8 , 13 , 16 , 17] . Such mixed genotype cultures are expected to be common in microbes , which often live in highly diverse and dense communities where many different genotypes meet and compete [5 , 18–21] . What then maintains cooperative phenotypes in microbes ? One explanation is the spontaneous formation of clonal patches within microbial communities by cell division ( also see [22] ) . As cells divide and grow in dense and nutrient-limited conditions , bottlenecks occur that cause genotypes to segregate into large clonal patches . The formation of these patches in initially diverse microbial groups is an empirically well-established phenomenon , having been observed in numerous microbial species including bacteria , yeast and amoeba , and different experimental conditions including agar plates and flow cells [23–32] . Theoretical work suggests that this patch formation in microbial colonies can stabilize the use of beneficial secretions because benefitting cells are now highly related [33] , which has recently been verified experimentally [34 , 35] . Here we investigate the costs and benefits of quorum sensing controlling the secretion of a public good—a key evolutionary dilemma [49]—in diverse and dense microbial communities . We use a realistic individual-based model of microbes that captures key features of the natural ecology of microbial groups [29 , 33 , 36–39] . We first recapitulate the most cited function of quorum sensing in single genotype groups: it allows cells to respond to cell density and diffusive conditions . We then consider what happens when quorum sensing cells are surrounded by competing genotypes . These competitors can either be cells that constitutively produce beneficial secretions ( cooperators ) or non-producers that have the potential to act as cheater genotypes . Our analysis shows how quorum sensing can evolve in dense and diverse microbial communities by enabling quorum sensing genotypes to outcompete both types of competitors . The benefit of quorum sensing we observe is not directly associated with the inference of cell density . In dense communities containing many different genotypes , quorum sensing additionally , and critically benefits cells by inferring when they are surrounded by clonemates .
According to the canonical view , quorum sensing allows cells to initiate shared beneficial traits only once cells are dense enough for the group to benefit . Shared beneficial traits can take a variety of forms including detoxification , slow growth rate and virulence factors . However , the archetypal shared beneficial trait is a secretion that helps cells acquire nutrients or minerals , like enzymes or chelators [14 , 40] . We therefore phrase our analysis in terms of secretions here but the basic conclusions should apply for any trait that carries an energetic cost to the cell that expresses it and benefits surrounding cells . We first test the prediction that density-dependent benefits will allow quorum sensing cells ( Q ) to have an advantage over both constitutive secretors ( C ) and non-secretors ( N ) . Density is defined as mass per unit volume: in our case , we take this volume to be constant ( the whole simulation space ) and simply quantify the total biomass . Starting from a single cell in a well-mixed , “liquid” environment ( Materials and Methods ) , N cells grow and divide , such that the biomass of the population increases exponentially over the course of the simulation ( Fig 1 , green line ) . C cells , on the other hand , initially grow slower than N because a fraction of their growth is redirected into the secretion of a factor that helps all cells around them to grow . For example , this could be a protease that breaks down proteins into amino acids or peptides for import into the cell [14 , 40] . After an initial phase of slow growth , C cells then experience a burst in growth once the concentration of the secreted factor is high enough to have an effect ( Fig 1A , blue line , and Fig 1C ) . This density-dependent benefit to secretions is consistent with experimental evidence [41–43] and is in accordance with previous models [7 , 33 , 38] . Q cells perform better than both N and C . In our model , quorum sensing cells secrete a non-costly diffusible autoinducer , sense its concentration , and only begin to produce the costly public good once a threshold concentration of the autoinducer is perceived . Quorum sensing cells grow identically to the N cells initially , and only start to produce public goods once they reach quorum , at which point the public good quickly accumulates to the threshold concentration such that all cells benefit ( Fig 1A and 1C , solid red lines ) . These results confirm that our simulations capture the canonical paradigm of quorum sensing , in which Q cells can prevent wasteful secretion of public goods and maximise the efficiency of public good secretion . We assume that the cells constantly secrete the autoinducer and do not consider positive feedback in its production , which can sometimes occur [44–46] . Introducing positive feedback would allow an increased potential to tune and optimise the timing of the quorum sensing response . Our predictions on the evolution of quorum sensing , therefore , are conservative in the sense that adding in additional complexity should only improve the scope for quorum sensing to evolve . Our first model assumes that the different genotypes live alone in clonal communities with no direct interactions between genotypes . This assumption favors strains with the maximum yield of biomass: the more cells a colony can generate , the higher the probability that its offspring will colonize new patches . While distinct clonal patches may well capture the biology of some microbial species , the extreme levels of diversity found in nature [18–20 , 47 , 48] suggest that strains are often surrounded by other genotypes , whether they are of the same species generated through genetic diversification or of distantly-related species . We next implement such competition in the model and recapitulate previous experimental [4 , 22] and theoretical work [7 , 10] , which shows how quorum sensing can be evolutionarily unstable in genetically diverse and well-mixed environments . Specifically , we next initialize the system with two cells that produce secretions , one cell of either the C or Q genotype together with a non-secretor cell ( N ) , in a well-mixed environment , equivalent to growth in liquid . While both C and Q outcompete non-secretors in the clonal groups ( Fig 1A ) , direct competition results in cells from all genotypes benefiting from the public good while non-secretors do not incur the costs of its secretion . They therefore outcompete both secretor genotypes ( Fig 1B ) . As predicted by previous theoretical and experimental models [4 , 7 , 8 , 10 , 22 , 36] , then , our simulations demonstrate how secretors can be exploited by non-secretors in well-mixed groups and our model is an example of a public goods dilemma [13 , 49] . While direct competition in liquid favors non-secretors , theoretical and experimental work shows that secretor genotypes can outcompete non-secretor genotypes in direct competition in a spatially structured environment [24 , 33 , 38 , 50] . Therefore , we next map out the effects of spatial structure on direct competitions between secretors and non-secretors [33] , before then considering how quorum sensing genotypes will fare under the same conditions ( Fig 2 ) . Previous work has shown that when nutrients are limiting , secretors can sometimes outcompete non-secretors in direct competition [33] . The reason is that low nutrient levels lead to population bottlenecks and the emergence of large patches of a single genotype , which prevents N genotypes from using the secretions of C genotypes ( Fig 2B , [33–35] ) . These bottlenecks leading from well-mixed to spatially structured populations represent a well-studied phenomenon that has been widely observed in microbial communities [29 , 30 , 51 , 52] . In social evolution terminology , the population bottlenecks drive an increase over time in genetic relatedness—the probability that two individuals are more genetically similar than the population average , or in this context , the probability that a focal cell is surrounded by genetically identical cells compared to other genotypes ( Materials and Methods ) —which is one of the major predictors of the evolution of cooperation [33 , 53] . However , secretors are often outcompeted by non-secretors , despite bottlenecking effects that increase relatedness . In particular , whenever constitutive secretors start to grow surrounded by cells of competing genotypes ( low genetic relatedness , [33 , 38] ) , it is more difficult for them to compete successfully ( Fig 2A , 1:4 competitions ) . Why does a constitutive secretor fare badly when surrounded by non-secretors ? The reason is that constitutive secretors are providing secretions that benefit non-secretors as much as secretors but only they are paying the cost to make them . The result is that faster growing non-secretors can physically overgrow and smother them , preventing secretors from capitalizing on the benefits of their cooperation . Constitutive secretors lose then under conditions of nutrient limitation and high diversity , both of which are thought to often occur in natural microbial communities [5 , 19 , 20 , 38 , 54] . Indeed , these conditions are expected whenever a focal genotype lands in an environment that has already been populated by other genotypes . Our initial analysis shows that when the social environment is highly competitive—nutrients are limiting and there are many genotypes—constitutive secretion can be a disadvantageous strategy . We next explore how a quorum sensing strain performs under such competitive conditions . We competed a Q strain , or a C strain , in pairwise competitions against the non-secretor N strain . Again , our scenario is that the population comprises a number of non-secretor ( N ) strains and we ask: what is the fitness of a single focal secretor of the Q or C strain , as a function of its initial frequency ? We consider biofilms comprised of 1 secretor strain mixed with 1 , 2 or 4 non-secretor strains ( Q or C: N—1:1 , 1:2 , and 1:4 ) to study the effect of increasing evolutionary competition ( i . e . increasing local diversity ) . We also consider a range of quorum sensing strains ( Q1 to Q4 ) that all produce autoinducers at the same , constant rate , but are induced to produce public goods at different , increasing , threshold concentrations of the quorum sensing autoinducer . When secretors ( C or Q1–4 ) were seeded at a 1:1 proportion ( low evolutionary competition ) , all five strains outcompete the N strain in each of the 100 simulations , with C achieving the highest mean relative fitness , although secretors do not differ much in their relative fitness ( Fig 3A ) . This changes when we consider strong evolutionary competition with a low initial proportion of secretors . Importantly , Q strains succeed against competing non-secretors where constitutive secretors fail . Quorum sensing then enables cooperation in highly competitive environments where non-secretors would otherwise dominate . What is the cause of the advantage to Q strains over C strains under these conditions ? Regulating secretion means that Q strains invest less in secretion overall than constitutive secretors . But in the supplement we reduce the investment of constitutive secretors to remove this difference and show that they still lose where Q strains win ( S1 Fig ) . The key to success of Q then is the timing of secretion ( in S2 Fig an artificial time-delay strategy demonstrates this ) . Quorum sensing allows a newly colonizing strain to compete against non-secretors and establish a clonal patch prior to activating energetically costly secretions , whereas C strains get overgrown and smothered while inefficiently investing into cooperative secretions . In these simulations of competitive , surface-attached communities then , the critical benefit to quorum sensing comes from the fact that expressing a costly secretion early means that such a genotype will be rapidly overgrown by faster growing genotypes . The benefit to quorum sensing that we observe is not directly linked to the assessment of cell density or total biomass . Rather , the relevant variable is the extent to which cells of a focal genotype are surrounded by clonemates relative to other cell types , which we compute using the “segregation index” . This segregation index is proportional to measures of kinship or relatedness within the range of a social trait [16 , 33 , 38] ( Materials and Methods ) . A strain is predicted to be more successful then , if it can match its cooperative secretions not to the cell density or total biomass , but to the appropriate level of local relative cell density or genetic relatedness . Furthermore , as cells in mixed microbial colonies divide and increase in number , the likelihood of being surrounded by clonemates compared to different genotypes increases and average relatedness between neighbouring cells increases [30–32 , 51 , 52]; relatedness , then , correlates positively with time . The importance of relatedness over cell density in defining the quorum sensing threshold is demonstrated by Fig 4 . We first consider a case where the seeding population is artificially sorted by genotype ( Q1 vs N ) and population bottleneck effects are removed . This manipulation artificially increases relatedness from the start . Here a Q1 genotype , which has a low quorum threshold and activates secretions immediately ( or even a C genotype , S3 Fig ) , can outcompete non-secretors because they are already surrounded by many clonemates ( Fig 4A ) . However , Q1 ( or C , S3 Fig ) cannot outcompete non-secretors when we seed the same density of cells and genotypes but without forced sorting of genotypes ( Fig 4B ) . Here , only a genotype with a higher quorum threshold ( Q4 ) can outcompete non-secretors ( Fig 4C ) . Q4 turns on public good production at a 4 times higher autoinducer concentration than Q1 , which means that Q4 delays costly secretion further in time ( see S1 Table ) . In competition with the N genotype , Q1 now gets buried under conditions where Q4 can escape burial and form clonal towers . The reason is that Q4 only initiates public good secretion when cells are surrounded by clonemates rather than others ( Fig 4E and 4F ) thereby amplifying the fitness gain from forming clonal clusters and outperforming non-secretors N on average ( Fig 3 ) . While still correlated with cell density , autoinducer concentration is tracking the shift from competitive , genetically-mixed populations to conditions where cells are surrounded by clonemates . Quantitatively , this can be seen from the fact that Q4 starts secretion at a higher segregation index ( Fig 4G–4I , Materials and Methods ) . And while other quorum sensing thresholds may initiate public good secretion at a more optimal moment compared to Q4 , this example strategy illustrates that using a quorum sensing threshold to delay cooperation in dense and diverse communities can be advantageous . The optimal threshold will depend not only on cell density but , critically , on the ability to keep track of time and the concomitant change in local genetic relatedness , as the cell group transitions from low relatedness to sufficiently high relatedness under which cooperation becomes favourable . Our model shows how a rare quorum sensing strain can succeed in a population of non-secretors , and so quorum sensing is predicted to evolve under these conditions . However , once common , how robust is quorum sensing-controlled public good secretion to potential exploitation by rare non-secretors ? In the supplement we ask how a rare non-secretor genotype fares in local competition with Q ( S4 Fig ) , which shows that quorum sensing strains also outcompete non-secretors when the latter are rare . Overall , our results show how the regulation of public good secretion by quorum sensing can be a more robust strategy than constitutive secretion in competitive environments . Specifically , in our system , quorum sensing strains can outcompete non-secretors where constitutive secretors will not ( Fig 5 ) . This is because quorum sensing strains can infer the local relative density of their own kin , and start cooperating once they have reached the appropriate level of genetic relatedness . But what about direct competition between a quorum sensing and a constitutive-secretor genotype ? In general , these two genotypes perform similarly in direct competitions although quorum sensing genotypes do maintain a slight advantage ( S5 Fig ) . This is again because Q cells grow faster initially and compete well for bottlenecks and , after establishing a clonal group , Q can benefit from its public good secretion . This initial edge leads to the success of Q over C , which is robust to reciprocal competition where a rare genotype C competes with frequent cells of genotype Q . The dominance of quorum sensing can also be seen in a competition involving all three main genotypes C , Q and N ( Fig 3B and 3C ) , and in the supplement we show that the benefits of quorum sensing are robust to a more complex model where cells in different biofilms disperse and compete globally ( S6 and S7 Figs ) . In our model , only quorum sensing cells respond to autoinducers . Importantly , we also assumed that only the quorum sensing strains produce the autoinducer . This scenario will occur whenever quorum sensing strains differ from competing strains at both the loci for induction and response . The use of a genotype-specific signal , as observed in strains of Bacillus[47] , raises the possibility that a focal genotype can detect the number of clonemate cells in the face of high variability in genetic mixing ( a form of kin discrimination [55] ) . However , such kin discrimination is not needed for the evolutionary benefits to quorum sensing that we observe . In some situations , all genotypes may produce autoinducers even if they do not all respond to them , such as when autoinducers are linked to common metabolic waste products [54 , 56] . We therefore also consider a system where all competing genotypes make the autoinducer . Under these conditions , Q cells maintain their advantage against non-secretor strains , albeit at different ( higher ) quorum sensing thresholds ( S8 Fig ) . While a genotype-specific signal is likely to be more robust to variability in strain mixing , therefore , it is not required for the evolutionary benefits to quorum sensing that we observe . Even if the change in autoinducer concentration reflects total cell density , a Q genotype can still use autoinducer concentration as a timer to infer genetic similarity . If the physical and social environment is sufficiently predictable , then , the indirect inference of kinship via autoinducers is sufficient for the effects we see , as opposed to strict kin detection and discrimination . This inference is possible whenever cell density correlates with the genetic similarity of neighbouring cells due to the changing spatial structure of the colony over time ( Fig 4 ) .
Two key issues have been identified as central to the evolution of quorum sensing to regulate the secretion of public goods . The first is a benefit over constitutive secretion . This is typically thought to come from the ability to infer cell density and perhaps diffusion rates . The second issue is that quorum sensing cells must resist competition from non-secreting cheater mutants . Here we show that , under conditions of strong competition , these two issues combine to provide a critical benefit to quorum sensing that is not simply due to the assessment of cell density . With nutrient limitation and high numbers of competing genotypes , the key benefit to quorum sensing in our system comes from the ability to delay public good secretion and grow quickly when first in a new environment . This allows quorum sensing cells to outgrow constitutive secretors while keeping up with fast growing non-secretors until clonal patches have formed in a microbial group . In particular , the delay in cooperative secretion in quorum sensing genotypes increases their chances of forming clonal clusters compared with constitutive secretors , and outcompeting them . Positive feedback from cooperation then amplifies the gain from forming a clonal cluster , which allows quorum sensing genotypes to outcompete non-secretors as well ( Fig 5 ) . Our model makes some simplifying assumptions . Firstly , auto-inducer production carries no cost in our simulations . While this may not always be the case in reality , many quorum sensing molecules , such as autoinducer-2 ( AI-2 ) carry little to no cost [57 , 58] . Secondly , some modeling choices and parameter values , such as the quorum sensing thresholds , were determined based on previous work or by parameter sweeps rather than experimentally measured values . This approach then demonstrates that wide and realistic parameter ranges exist in which quorum sensing is adaptive . Finally , the importance of the predicted effect , whereby quorum sensing can delay cooperation until relatedness has increased will , of course , depend on how often microbes find themselves in highly competitive environments , which transition from diverse mixtures of strains to clonal clusters as they grow . In cases where cells grow in isolation as microcolonies , the benefits of quorum sensing are likely to come from the inference of cell density and diffusion as is typically assumed [10 , 12 , 13 , 16] . However , there is a growing recognition of the importance of competition for understanding microbial phenotypes [5 , 6 , 54] . A focal microbial genotype may often land in a dense community where other genotypes are already present; in fact , most of microbial life is assumed to take place under these conditions [20] . And although our model here assumes that competing genotypes differ only in a few loci with all else being equal , higher background diversity will not alter our conclusions . Our model suggests then that quorum sensing can be particularly advantageous when competition between many different genotypes is fierce . This leads to a testable hypothesis: when surrounded by foreign genotypes , a strain that uses quorum sensing to control public good secretion can grow as fast as non-secretors and only activate secretions once it is safely surrounded by its own genotype .
We define fitness of each genotype ( for example , x and y ) as the mean number of rounds of cell division per unit time that cells of a focal genotype achieve during the interval between initial seeding at t0 and tend when a maximum amount of nutrients were consumed . Fitness w , therefore , is calculated as w x = 1 t e n d l o g 2 N x , t e n d N x , 0 , ( 1 ) where Nx , t is the number of cells of genotype x present within the cell group at time t . The relative fitness of a genotype x in local competition with another genotype y is defined as: log 10 ( w x w y ) and , therefore , competition is successful when w > 0 . The mean was taken over 100 such simulations and we show convergence of the results in the supplement ( S7 Fig ) . A simple meta-population analysis was conducted following the same approach as in [33] . This determines whether a rare mutant strain would succeed in a metapopulation of cell groups with reoccurring dispersal and colonisation events . We assume a very large number of cell groups where the great majority of groups are of a single dominant strain and only a small minority will contain the mutant . Under these conditions , a genotype x ( rare mutant ) can invade a meta- population of genotype y ( majority resident ) if the fitness of x in local competition with y is greater than the average fitness of the whole metapopulation , denoted 〈wy〉 . For each invasion analysis , wx was computed in 100 replicates of the simulations ( with varying inoculation frequencies of the two genotypes and a total of 800 cells initially , see relevant figures ) . Because the great majority of cell groups in the meta population consist purely of the majority genotype y , 〈wy〉 is approximately the mean fitness of the majority genotype , y , when growing on its own . To calculate 〈wy〉 , the mean of wy over 100 simulations is computed , where the cells of genotype x inoculated initially are replaced with y cells and a mono-culture of the majority genotype is simulated . The invasion index Ix → y of a rare mutant x into a metapopulation with majority genotype y was calculated for each of the 100 replicates as I x → y = w x ⟨ w y ⟩ . Under the assumptions of our model , we conclude that x can invade a metapopulation of y when Ix → y > 1 . Simulation results are from 100 independent replicates . Fitness data is non-normal and often bimodal distributed where the bimodality differs between simulations with different initial frequencies and/or initial cell densities meaning it is difficult to apply standard statistics . In some figures we show box plots and test the median fitness value with non-parametric sign tests . This is only an indicator as from an evolutionary perspective , the mean relative fitness is the determining parameter of evolutionary success . Therefore , we further conducted convergence analyses that show how the mean fitness converges after about 100 simulations ( S7 Fig ) . The segregation index used here is identical to that used in previous work [38] . To measure segregation in a population of M cells , we consider each cell ci , i = 1…M in the population and identify all other individuals within a distance of 10μm . The N cells in this neighbourhood are indexed by cj , with j = 1…N . We define a genotypic identity function , g ( ci ) : g ( c i ) = 0 , c j is not the same genotype as c i 1 , c j is the same genotype as c i ( 2 ) Segregation with respect to a focal cell , s ( ci ) , was calculated as the mean product of the g and m functions for every cell in its neighbourhood: s ( c i ) = 1 N ∑ j = 1 N g ( c j ) ( 3 ) Finally , we define the segregation index σ for the entire cell group as the mean value of s ( ci ) across the population of cells: σ = 1 M ∑ i = 1 M s ( c i ) ( 4 ) The segregation index measures the degree to which co-localised cells are clonally related to each other . Relatedness in social evolution is defined as the probability that two individuals are more genetically similar than the population average . The segregation index is equal to a form of the relatedness coefficient from social evolution theory under the following assumptions: ( i ) A cell expressing the cooperative phenotype equally benefits all other individuals within a 10 cell-length radius; and ( ii ) Cell groups are seeded randomly from a large population pool [33] . In this meta-population , the frequency of a given focal strain is constant and small . To seed our simulations , then , we sample a number of strains ( between 2 and 5 , one of which is our focal strain ) randomly from this population . The likelihood of a cell of the focal strain interacting with its clone in the meta-population is negligibly small . The segregation index then computes the likelihood of the focal cell interacting ( presence over a 10 cell-length radius—this radius is somewhat arbitrary , and was kept identical to previous studies ) with its clone relative to the null expectation in the metapopulation ( close to 0 ) .
|
Bacteria secrete signal molecules into their environment and use these to regulate many of their key phenotypes . This is called quorum sensing and it is thought to evolve because it allows cells to sense their density . Here we propose a new function for quorum sensing that sheds light on its evolution . We develop a realistic model of a bacterial community and show that quorum sensing can function as a way to outcompete neighbors in patches occupied by many different genotypes . Growing aggressively at first makes quorm sensing genotypes a match for competitors . This strategy allows them to surround themselves with clonemates before reallocating resources to costly traits like cooperative secretions . This works because quorum sensing can act as a timer , which cells can use to infer how related they are to their neighbours and tune their investment into costly and exploitable cooperation based on the threat of competition from unrelated genotypes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biofilms",
"organismal",
"evolution",
"medicine",
"and",
"health",
"sciences",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"microbiology",
"cloning",
"simulation",
"and",
"modeling",
"physiological",
"processes",
"microbial",
"evolution",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"microbial",
"physiology",
"molecular",
"biology",
"evolutionary",
"genetics",
"cell",
"biology",
"physiology",
"secretion",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"quorum",
"sensing"
] |
2016
|
The Evolution of Quorum Sensing as a Mechanism to Infer Kinship
|
Global climate change , increasingly erratic weather and a burgeoning global population are significant threats to the sustainability of future crop production . There is an urgent need for the development of robust measures that enable crops to withstand the uncertainty of climate change whilst still producing maximum yields . Resurrection plants possess the unique ability to withstand desiccation for prolonged periods , can be restored upon watering and represent great potential for the development of stress tolerant crops . Here , we describe the remarkable stress characteristics of Tripogon loliiformis , an uncharacterised resurrection grass and close relative of the economically important cereals , rice , sorghum , and maize . We show that T . loliiformis survives extreme environmental stress by implementing autophagy to prevent Programmed Cell Death . Notably , we identified a novel role for trehalose in the regulation of autophagy in T . loliiformis . Transcriptome , Gas Chromatography Mass Spectrometry , immunoblotting and confocal microscopy analyses directly linked the accumulation of trehalose with the onset of autophagy in dehydrating and desiccated T . loliiformis shoots . These results were supported in vitro with the observation of autophagosomes in trehalose treated T . loliiformis leaves; autophagosomes were not detected in untreated samples . Presumably , once induced , autophagy promotes desiccation tolerance in T . loliiformis , by removal of cellular toxins to suppress programmed cell death and the recycling of nutrients to delay the onset of senescence . These findings illustrate how resurrection plants manipulate sugar metabolism to promote desiccation tolerance and may provide candidate genes that are potentially useful for the development of stress tolerant crops .
The desiccation tolerant grass , Tripogon loliiformis , is a small , tufted diploid grass and member of the poaceae family of cereals that is native to Australia and New Guinea and grows in rocky outcrops and nutrient poor soils with low water retention[1] . In these microenvironments , T . loliiformis is constantly subjected to environmental extremes and as such has evolved remarkable mechanisms for survival; plants live even after snap-freezing with liquid nitrogen or heating for short periods at temperatures > 60°C[2] . Accordingly , resurrection plants have been investigated for the identification of novel stress tolerance strategies . The advent of “omics” technologies and systems biology approaches provide the experimental power to address the mechanistic details and identify the key mediators of how resurrection plants display the robustness to withstand environmental extremes . Transcriptome , proteome and metabolome studies have been performed on several resurrection plants and have revealed numerous mechanisms that account for the remarkable resilience observed ( for a review refer to Dinakar and Bartels , 2013 ) . Fundamental discoveries of the tolerance strategies utilised by resurrection plants include the early detection of dehydration and shut-down of photosynthesis , the presence of extensive ROS scavenging systems , even in the hydrated state , the accumulation of sugars , as well as the enrichment of transcripts associated with cell wall plasticity[3–12] . Importantly , transcripts and metabolites typically associated with gene profiles observed in seeds are often detected within vegetative tissues , leading to the hypothesis that resurrection plants conform to a dormant “seed-like” state upon drying [3 , 4 , 13 , 14] . The regulation of carbohydrate and nitrogen metabolism also appears to be an integral component of stress tolerance strategies in resurrection plants . In addition to sucrose metabolism , several resurrection plants accumulate substantial levels of the dissarcharide trehalose during drying [9 , 15 , 16] . Trehalose is a non-reducing sugar present within a wide-range of organisms including , insects , fungi , bacteria , yeast and several plants and is thought to play a protective role against various environmental stresses[17] . Many of these roles were originally identified in yeast where trehalose plays a protective role by functioning as a chemical chaperone , which prevents protein denaturation , aggregation and influences protein folding through trehalose-protein interactions[17] . In resurrection plants a precise role for trehalose within desiccation tolerance remains elusive as studies have indicated that the trehalose contents accumulated are insufficient to act as either a chaperone or energy source [18] . Treatment with trehalose significantly prolongs the vase life of cut Gladiolus flowers and treated flowers display higher water content and membrane integrity as well as decreased protein degradation . Importantly , a unique role for trehalose from that of other sugars was distinguished by H-NMR spectroscopy which demonstrated that at least in cut flowers trehalose does not play a role in osmotic adjustment but functions to protect vacuolar water [22] . Recently , a new role for trehalose in the induction of mammalian m-TOR independent autophagy pathways was elucidated [19] . Treatment of podocytes with trehalose triggered autophagy and alleviated the effects of mutant proteins associated with Hungtinton and Parkinson disease [19] . Importantly , the effects of trehalose were suppressed by autophagy inhibitors thus linking trehalose with autophagy and not proteasome-mediated pathways [23] . Moreover , trehalose triggered m-TOR independent autophagy did not involve reactive oxygen species and correlated with reduced levels of apoptotic cells[20] . Accordingly , the accumulation of trehalose may play a similar role within the activation of autophagy pathways for the maintenance of tissue vitality in desiccated resurrection plants . Autophagy is a catabolic cellular process that eukaryotic cells instil as a last ditch effort to restore homeostasis during severe stress and involves the sequestering of cellular material for degradation and recycling or “self-eating”[21] . A double-edged sword , autophagy promotes both survival and death outcomes depending on the context and thus requires intricate regulation to help the cell maintain homeostasis and prevent programmed cell death ( PCD ) [22 , 23] . Prolonged stress and excessive autophagy results in “over-eating” of cellular components , insufficient molecular machinery to function and cell death . On the other hand , autophagy is a crucial component of the cellular response system that removes damaged and toxic cellular components , following over stimulation , the accumulation of these components can trigger the induction of PCD pathways[24] . Given the definition of “resurrection” includes “rising from the dead” , the fate of these plants can be viewed in one of two ways , i ) desiccated tissues and cells are dead and new growth occurs upon watering or ii ) pre-existing tissues arise from a dormant state . The study of cell and tissue fate in resurrection plants provides new insights on how this unique group of plants tolerate extreme environmental challenges; to date , however , little to no progress has been made [25] . Here we show that Tripogon loliiformis utilises a myriad of molecular pathways to facilitate the survival of vegetative tissues even in the desiccated state . Key observations include i ) “resurrection of pre-existing tissues upon watering , ii ) absence of cell death in dehydrated and desiccated tissue , iii ) increased detection of autophagy in dehydrating shoots , iv ) accumulation of trehalose throughout dehydration , v ) trehalose-triggerred induction of autophagy pathways in hydrated leaves and vi ) absence of senescence during dehydration and desiccation . These results highlight potential mechanisms that underlie the ability of resurrection plants to survive desiccation and provide functional validation of why some resurrection plants accumulate trehalose at levels that are too low to serve as either osmoprotectants or energy sources . Once activated , autophagy presumably directs the recycling of nutrients and removal of damaged and potentially harmful cellular material to offset senescence and PCD . Effective regulation of autophagy may enable resurrection plants to retain viability even under extreme stress . Due to its broad-sweeping roles from stress-mitigation , anti-aging and prevention of PCD , the study of autophagy in naturally resilient species may provide fundamental information that can be used for the generation of stress-tolerant crops .
Despite significant research on resurrection plants , little is known of the fate of individual cells throughout desiccation and rehydration . To determine whether pre-existing tissues resurrect or new growth occurs upon watering , time-lapse photography was performed over the rehydration process ( S1 Video ) . The video clearly shows pre-existing tissues resurrecting . The revitalisation of pre-existing growth rather than new growth is an important demarcation and suggests that T . loliiformis’ vegetative tissues return from a dead-like , cryptobiotic state . To further investigate whether desiccated leaf tissues are alive and “resurrect” upon watering , hydrated , dehydrating ( 60% and 40% Relative Water Content ( RWC ) ) and desiccated leaves ( at <10% RWC ) were harvested and stained with Evans blue , an established vital stain that permeates membranes of dead cells . As a control , hydrated leaves boiled for five minutes were also stained . The boiled controls displayed strong Evans blue staining to indicate membrane damage and cell death . As shown in Fig 1 , no staining was observed for young leaves at all hydration levels ( Fig 1 ) . The absence of Evans blue staining suggests that T . loliiformis tissues remain viable in the desiccated state . The observation that pre-existing tissues are alive and return from the desiccated state suggests that T . loliiformis employs efficient cytoprotective measures that prevent cell death during drying . Previous studies suggest that vascular resurrection plants conform to the latter of the two . To test these hypotheses and potentially identify previously unknown stress tolerance strategies we performed RNA-seq analysis on dehydrating ( 60 and 40% RWC ) , desiccated ( < 10% RWC ) and rehydrating T . loliiformis shoots . As controls , hydrated shoots were also sequenced . These stages of dehydration , desiccation and rehydration were used as they are known to represent vital thresholds that dictate the metabolic activity of resurrection plants . Following Illumina library construction and sequencing , the reads were mapped to a de novo assembled blast annotated reference transcriptome . To focus our analysis on whether cellular repair mechanisms were being induced or PCD was being suppressed we compared the expression profile of established plant PCD genes . Key genes associated with senescence were also investigated . A total of 49 genes associated with apoptototic-like cell death ( 20 ) , senescence ( 16 ) and autophagy ( 13 ) were identified in the T . loliiformis transcriptome and RNA-seq expression analysis revealed the following features , i ) transcripts associated with apoptosis and senescence showed less accumulation , ii ) transcripts associated with the delay of senescence were increased , and iii ) autophagy-related transcripts were more abundant throughout dehydration and desiccation ( details for all genes investigated are provided in S1 Table ) . Subsequent qPCR of select contigs correlated with the fold-changes and the overall trend of the RNA-seq data thus validating the observations ( S2 Table ) . Excessive abiotic and biotic stresses can induce PCD of a few constrained cells presumably for the benefit of the organism as a whole . The transcriptome data shows that during dehydration and desiccation T . loliiformis suppresses transcription of PCD- and senescence-related genes whilst favouring the expression of autophagy genes . To further investigate whether PCD occurs during dehydration and desiccation of T . loliiformis leaves , TUNEL assays which allow for the visualisation of DNA fragmentation , an established marker of apoptotic-like cell death , were performed . Consistent with the absence of Evans blue staining and despite numerous TUNEL positive cells in DNase treated controls , no TUNEL positive cells were observed in hydrated , dehydrating or desiccated T . loliiformis leaves ( Fig 2 ) . The Evans blue and TUNEL assay results suggest that T . loliiformis cells do not succumb to death during desiccation and rather than resurrect , the vegetative tissue is restored from an anhydrobiotic state . The precise mechanisms of how T . loliiformis cells retain viability in conditions that devastate the majority of angiosperms remain to be elucidated , however , it has been suggested that resurrection plants conform to a dormant seed-like state for survival and longevity upon desiccation . Autophagy pathways have been associated with cellular survival by recycling essential nutrients as well as the removal of damaged organelles and other cellular toxins , thus providing the cells with essential resources for the maintenance of cellular homeostasis . Notably , autophagy is essential for drought and stress tolerance as autophagy deficient plants are significantly more sensitive to drought [24] . The transcriptome data showed that autophagy is triggered in T . loliiformis during dehydration and desiccation . Could resurrection plants more effectively regulate autophagy pathways than their angiosperm counterparts thus explaining their innate ability to tolerate extreme environments ? To further investigate the extent that autophagy pathways are triggered physiologically , hydrated , dehydrating and desiccated T . loliiformis leaves were i ) immunoblotted for detection of transient phosphatidylethanolamine ( PE ) conjugation of ATG8 and ii ) stained with the autophagosome dye Monodansylcadaverine ( MDC ) and viewed by confocal fluorescence microscopy[26] . The transient conjugation of Atg8 to the amino group of the membrane lipid phosphatidylethanolamine ( PE ) is an integral component required for autophagosome formation and can be detected by immunoblotting . As shown in Fig 3 and consistent with the transcriptome data , the presence of the ATG8-PE adduct increased during dehydration thus indicating increased accumulation of autophagosomes ( Fig 3 ) . Importantly , lipidation of the ATG8 was confirmed by digestion with phospholipase D which converted the ATG8-PE conjugate into the free form ( Fig 3 ) [27] . The immunoblot was subjected to densitometry analysis by ImageJ software after normalisation for equal loading against ponceau stained total proteins transferred to the PVDF membrane , further confirming the accumulation of ATG8-PE throughout dehydration and desiccation ( Fig 3 and S1 Fig ) . MDC stains non-autophagic cells diffusely while autophagosomes are detected as punctate vesicles . Moreover and in accordance with the transcriptome data , no autophagosomes were detected in hydrated T . loliiformis leaves , multiple autophagosomes however , were observed in dehydrating and desiccated leaves ( S2 Fig ) . Taken together , these results suggest that autophagy is triggered in T . loliiformis during desiccation . The transcriptome , immunoblot and microscopy data suggest that autophagy may play a major role in the preparation of T . loliiformis in tissues for desiccation and may facilitate the observed prolonged viability compared to sensitive plants . In the search for autophagy regulatory partners we noted that the non-reducing disaccharide trehalose is an inducer of M-TOR independent autophagy pathways in mammals [19 , 20] . Trehalose has also been reported to accumulate in several resurrection plants [9 , 28] . It is tempting to speculate that resurrection plants utilise trehalose to trigger autophagy and this represents a unique survival strategy . To further investigate this possibility , GCMS was performed on T . loliiformis shoots throughout dehydration . Additionally , to confirm that trehalose can trigger autophagy , hydrated T . loliiformis leaves were treated with trehalose and viewed by Transmission Electron Microscopy ( TEM ) for the presence of autophagosomes ( Fig 4 ) . As shown in Fig 4 , GCMS analysis showed that sucrose concentration rose steadily during drying and peaked in desiccated tissue . In accordance with sucrose , trehalose accumulated , albeit at much low levels , throughout dehydration and also peaked in desiccated tissue; hydrated tissues contained low levels of trehalose ( Fig 4 ) . These results were consistent with the transcriptome , confocal microscopy and immunoblot data and show that trehalose accumulates in dehydrating and desiccated T . loliiformis shoots at the same stages of desiccation that autophagosomes are detected . To further confirm this relationship and assess whether trehalose triggers autophagy directly in vitro hydrated T . loliiformis leaves were treated with 1 or 5 mM trehalose and 1 μM Concanaymcin A solution for 24 hrs and viewed by TEM . As controls , leaves treated with ½ MS salts and Concanamycin A alone were also assessed . Consistent with the previous MDC staining , no autophagosomes were detected in the control samples ( Fig 4 ) . Conversely , numerous punctate structures resembling autophagosomes were observed in both the 1 and 5 mM trehalose-treated samples when visualised by TEM ( Fig 4 ) . These results provide direct in vitro evidence for trehalose-triggered autophagy in plants and suggest that an association exists between trehalose accumulation and the induction of autophagy in T . loliiformis .
Resurrection plants are unique in their ability to prolong the life of their vegetative tissue in a desiccated state and rejuvenate upon watering . Our results showed that Tripogon loliiformis plants do not “resurrect from the dead” but implement pro-survival autophagy pathways , possibly facilitated by the accumulation of trehalose , to prevent apoptosis and senescence during desiccation . Once regarded as a self-destructive process and a form of PCD , autophagy has emerged as a pro-survival process . The induction of autophagy in T . loliiformis promotes survival by removal of cellular toxins to suppress PCD and the recycling of nutrients to delay onset of senescence . Most importantly , the accumulation of trehalose may maintain the efficiency of autophagy pathways when it is most needed , i . e . 60% RWC and below . The manipulation of autophagy pathways may present great potential for the development of stress tolerant crops .
Tripogon loliiformis plants were collected from Charleville ( Queensland , Australia ) , transferred to 30 cm pots and allowed to revive in a growth chamber ( 27°C , 16 hour light period ) for three weeks . Five plants were transferred to glasshouse conditions for seed setting . Tripogon loliiformis plants were germinated from seeds collected from a single plant and germinated in a growth chamber at 27°C and 16h photoperiod . Fifteen , 65mm pots containing multiple plants were grown for a period of two months . Prior to dehydration all plants were watered to saturation . One day post-watering three replicate samples from the hydrated plants were randomly collected . The remaining plants were dehydrated by withholding water until they were air dry and their relative water content ( RWC ) dropped below 10% . During dehydration , triplicate samples were collected at 60% , 40% and <10% RWC . Desiccated plants were watered and rehydrated samples collected after 48hrs . The percentage RWC was determined on T . loliiformis shoots and was calculated according to Barrs and Weatherley , 1962 using the formula ( RWC ( % ) = ( ( Fresh Weight—Dry Weight ) / ( Turgid Weight—Dry Weight ) ) x 100 ) [48] . All the shoot and root samples were snap frozen in liquid nitrogen and stored at -80°C until RNA extraction . To assess cell vitality throughout dehydration Evans blue staining was performed . At least ten shoots from hydrated ( ≈ 94% RWC ) , dehydrating ( 60 & 40% RWC ) and desiccated ( <10% RWC ) T . loliiformis plants were harvested , placed into 2 mL microfuge tubes and soaked in water for 2 hours to facilitate staining . To serve as a positive control hydrated leaves were boiled for 5 minutes . Following soaking , the water was replaced with 0 . 25% w/v Evans blue dye and the samples were incubated at room temperature for 20 mins . Stained leaves were rinsed with de-ionised water to remove excess Evans blue dye and stained cells were visually assessed by light microscopy . For sequencing analysis , total RNA was isolated from shoot and root tissue of hydrated , dehydrating , dehydrated and rehydrated T . loliiformis plants using a modified Trizol ( Invitrogen ) and spin column ( Qiagen ) method . Details of the extraction protocol can be found in supplemental information . RNA integrity and quality were verified using a Bioanalyser ( Agilent technologies ) . For library preparation , polyadenlyated RNA was enriched , chemically fragmented and cDNA was synthesised using an Illumina RNA-seq kit according to manufacturer’s recommendations . Sequencing of the cDNA libraries was performed at Texas A&M AgriLife Genomics and Bioinformatics service , USA using an Illumina HiSeq 2500 Sequencer ( Illumina Inc . ) . Single-read sequences of length 100 bp were collected . All reads have been deposited in the Sequence Read Archive ( SRA ) at NCBI , Accession number PRJNA288839 . Sequences were assessed for quality control , trimmed to remove the primer and barcode sequences . A readily available de novo assembled , blast annotated T . loliiformis transcriptome assembly served as a reference for RNA-Seq profiling of the independent cDNA libraries . Over 80% of the reads from each sample were mapped . All data sets were paired and an in silico microarray experiment was performed using CLC genomics workbench . Using the Hydrated sample as a reference , each data set was enriched for genes that had a fold change ≥ 2 or ≤ -2 , an experimental difference ≥ 5 or ≤ -5 and an adjusted p-value < 0 . 05 following normalisation by comparison of means . Superscript III Reverse Transcriptase ( Invitrogen ) was used to generate cDNA from 0 . 8μg of Total RNA using an oligo ( dT ) ( 100ρmol ) primer . Quantitative PCR was done using a ViiA7 Real-Time PCR System and the SYBR Green PCR Master Mix kit ( Applied Biosystems ) according to the manufacturer’s instructions using 300 mM primer and 1/100 dilution of cDNA and standard cycling parameters . Gene specific primers for selected genes were designed using Primer3 bioinformatic software ( MIT ) and are listed in S3 Table . The data analysis was completed using ExpressionSuite Software ( Life Technologies ) . The Tripogon loliiformis homologue of Arabidopsis Actin identified from the annotated transcriptome were used for quantitative normalisation . Fold changes were calculated against hydrated tissue . To determine whether T . loliiformis cells were undergoing Apoptotic cell death during desiccation TUNEL assays were performed using the In situ Cell Death Detection Kit , Fluorescein ( Roche ) as described by Hoang et al . , 2014[49] . A hydrated sample was also included as a negative control . TUNEL positive cells were made visible by Nikon A1 Confocal Microscopy . To detect the presence of autophagosomes in dehydrating Tripogon loliiformis plants , shoots from hydrated , dehydrating and desiccated plants ( as mentioned above ) were excised , immersed in 100 uM Monodansylcadaverine ( MDC ) and incubated for 30 min at room temperature in the dark [50] . Four leaves for each hydration point were assessed ( hydrated , 60% , 40% and 10% RWC ) . Each leaf was sliced into 1 cm sections and immersed into a 2 mL microfuge tube containing 100 μM MDC stain diluted in ½ MS solution; samples were incubated in the dark for 30 mins . Following staining , each leaf section was washed twice with ½ MS media to remove excess stain and further sectioned by hand using a razor blade . Leaf sections were mounted onto glass slides and observed for the presence of autophagosomes by confocal microscope ( Nikon air confocal ) using a DAPI filter with excitation and emission at 335nm and 508nm , respectively . Samples were viewed under 40 and 60x oil immersion lenses . To assess whether exogenous application of trehalose can trigger autophagy pathways hydrated leaves were treated in triplicate with trehalose and viewed by Transmission Electron Microscopy ( TEM ) . Fully emerged hydrated leaf samples from three month old glasshouse grown T . loliiformis plants were harvested , divided into 1 cm segments , vacuum infiltrated with a 1 and 5 mM trehalose solution containing 1 μM concamycin A prepared in ½ MS basal salts and incubated for 24 hrs in the light . As a positive control , leaf samples treated with Tunicamycin ( 5 μg/mL ) were also included . For transmission electron microscopy , Trehalose and control treated T . loliiformis leaf sections were fixed in 3% glutaraldehyde in 0 . 1M sodium cacodylate buffer followed by post-fixation in 1% osmium tetroxide in 0 . 1M sodium cacodylate buffer . Samples were subsequently rinsed in UHQ water and dehydrated through a graded series of acetone and embedded in Embed-812 resin . Ultrathin sections were cut on a Leica UC7 ultramicrotome ( Leica Microsystems , Wetzlar , Germany ) and imaged with a JEOL JEM-1400 transmission electron microscope at an accelerating voltage of 80kV . Shoots from dehydrated and desiccated two month old T . loliiformis plants were snap frozen in liquid nitrogen ground and homogenised in extraction buffer ( 50mM HEPES-KOH ( pH7 . 5 ) , 150mM KCl , 1mM EDTA , 0 . 2% Triton-X100 , 1mM DTT ) . Protein samples were quantitated by Bradford assay and approximately 30 μg were separated by Urea-SDS-PAGE ( 12% Acrylamide , 6M Urea ) and transferred onto a polyvinylidene difluoride ( PVDF ) membrane ( Immobilon-PSQ , Millipore membrane ) . To verify lipidation , protein extracts were digested at 37°C for 1 hour with Streptomyces chromofuscus Phospholipase D ( Sigma; 250 units mL−1 final concentration ) as described by Chung et al . , 2009 ) [27] . Immunoblotting analysis was performed using a polyclonal antibody raised against Arabidopsis ATG8 ( 1:1000 dilution ) ( ABcam ) , then visualized using a peroxidase-conjugated goat anti-rabbit IgG ( 1:1000 dilution ) and the Supersignal West Femto Maximum Sensitivity Substrate ( Thermo Scientific ) according to manufacturer’s instructions . Quantitation of the ATG8-PE band was performed by ImageJ analysis of the Immunoblot normalised against the rubisco protein transferred to the membrane as detected by Ponceau staining . Data are represented as densities of the ATG8-PE bands normalised against their respective Ponceau stained counterparts . To analyse changes in T . loliiformis metabolite and potential Trehalose accumulation during dehydration GCMS was performed . Three month old hydrated , dehydrated ( 60 & 40% RWC ) and desiccated <10% RWC plants were harvested , snap-frozen in liquid nitrogen and lyophilised overnight . Following lyphilosation , the dry weight was measured for normalisation and the samples were ground to a powder using a Qiagen tissue lyser ( 2 x 1 min ) . Metabolites were extracted and analysed as described by Fiehn et al and Hu et al [51 , 52] . Electron ionisation of mass spectra were recorded at a scanning range of 30–650 m/z . For trehalose identification , single ion monitoring with ions 191 and 361 m/z corresponding to the most abundant and specific ions of trehalose methoxyamin 8 trimethylsilyl was performed . All experiments were conducted using three biological replicates .
|
Over coming decades , climate change models suggest that droughts and other unpredictable weather patterns will appear more frequently . It is imperative that we develop crops that can survive future climates but continue to yield . Numerous studies have shown that stress tolerance is genetically encoded . Naturally tolerant species therefore represent an ideal starting point for the search for stress tolerance . Resurrection plants belong to a small group of vascular plants that possess unique stress tolerance mechanisms to withstand extreme desiccation with the ability to recover fully upon the availability of water . Here we describe a unique regulatory role for trehalose in the activation of autophagy pathways in T . loliiformis . We show that T . loliiformis leaves are alive in desiccated plants and that pre-existing tissues resurrect upon the addition of water . By using a combination of transcriptomics , confocal microscopy and spectroscopy we show that autophagy is induced during dehydration . Notably , we establish that treatment of leaves with trehalose triggers autophagy in vitro and that trehalose accumulation in dehydrating leaves correlates with the presence of autophagosomes . We postulate that resurrection plants modulates trehalose metabolism to induce and maintain autophagy pathways that preventing senescence and programmed cell death .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Trehalose Accumulation Triggers Autophagy during Plant Desiccation
|
Rats discriminate surface textures using their whiskers ( vibrissae ) , but how whiskers extract texture information , and how this information is encoded by the brain , are not known . In the resonance model , whisker motion across different textures excites mechanical resonance in distinct subsets of whiskers , due to variation across whiskers in resonance frequency , which varies with whisker length . Texture information is therefore encoded by the spatial pattern of activated whiskers . In the competing kinetic signature model , different textures excite resonance equally across whiskers , and instead , texture is encoded by characteristic , nonuniform temporal patterns of whisker motion . We tested these models by measuring whisker motion in awake , behaving rats whisking in air and onto sandpaper surfaces . Resonant motion was prominent during whisking in air , with fundamental frequencies ranging from approximately 35 Hz for the long Delta whisker to approximately 110 Hz for the shorter D3 whisker . Resonant vibrations also occurred while whisking against textures , but the amplitude of resonance within single whiskers was independent of texture , contradicting the resonance model . Rather , whiskers resonated transiently during discrete , high-velocity , and high-acceleration slip-stick events , which occurred prominently during whisking on surfaces . The rate and magnitude of slip-stick events varied systematically with texture . These results suggest that texture is encoded not by differential resonant motion across whiskers , but by the magnitude and temporal pattern of slip-stick motion . These findings predict a temporal code for texture in neural spike trains .
Rodent whiskers , like human fingertips , are tactile detectors that are actively moved through the environment to sense position , shape , and surface features of objects . A particularly salient surface feature is texture , which is more readily distinguishable using touch than vision [1] . Rats discriminate textures using their whiskers with a precision that rivals human fingertips [2–5] . How whiskers read out texture information , and how that information is encoded in the nervous system , are vigorously debated , and have important implications for sensory processing in the whisker system [6 , 7] , which is a major model system for studying cortical function and plasticity [8 , 9] . Rats have an array of approximately 30 large whiskers ( macrovibrissae ) on each side of the face . Whisker length varies systematically across the whisker pad , with caudal whiskers being longer than rostral whiskers . Whiskers are moved rhythmically at 5–15 Hz to explore objects in the environment , including textures [5 , 6 , 10] . Two main hypotheses exist for texture discrimination by the whiskers , based on experiments using detached whiskers and in anesthetized animals . The resonance hypothesis derives from the observation that whiskers are resonant beams , with characteristic resonance frequency inversely related to whisker length [11 , 12] . Whisker-length variation across the whisker pad results in a spatial map of fundamental resonance frequency ( FRF ) . In this hypothesis , whisker-tip motion across surface microfeatures causes tip vibration at a frequency that varies with texture spatial frequency . Only when textures generate tip vibration at the FRF will vibrations most effectively build up and be transmitted to the whisker follicle , where transduction occurs . As a result , each whisker is best activated by a specific range of textures , and each texture preferentially activates a subset of whiskers , leading to a spatial code for texture in the relative amplitude of vibrations across the whisker array [13] . An alternative model is that texture is encoded temporally , by unique temporal patterns of movement ( “kinetic signatures” ) that are induced within single whiskers scanning across surfaces . These patterns have been proposed to include both mean speed ( amplitude × frequency ) of whisker vibration [7 , 14] , spectral composition of whisker vibrations [15] , and the precise , irregular velocity profile of whisker motion [7] . This latter feature provides higher-resolution texture information than vibration speed or frequency alone [7 , 16] . To distinguish these models , it is critical to measure whisker vibrations and neural responses in awake , behaving animals voluntarily palpating surfaces . This is because the dynamics of voluntary whisker movement will critically impact the transformation of surface features into whisker-motion signatures . Whiskers are known to exhibit multiple modes of vibration during voluntary palpation of surfaces , including resonance vibration and irregular , high-velocity motion events [17] . However , which of these features correlate with , and therefore may encode , texture , is not known . Here , we evaluated the resonance and kinetic signature models of texture by precisely measuring whisker vibrations in awake , behaving rats trained to actively whisk onto textured surfaces . Results showed that whisker resonance occurs during free whisking in air and during brief , discrete epochs while whisking onto textures . However , the magnitude of resonance vibrations did not vary across textures , as required for the resonance hypothesis . Instead , whisker resonance on surfaces primarily represented transient ringing during brief ( 5–10 ms ) , high-velocity , high-acceleration slip-stick events . Slip-stick events were a prominent component of whisker motion on surfaces , and the rate and magnitude of these events correlated well with texture . These results indicate that whisker resonance occurs in awake rats and shapes natural whisker vibrations , but that texture is not encoded by differential resonance across whiskers , at least under these behavioral conditions . Instead , slip-stick events may contribute to a kinetic signature for texture in the whisker system .
To measure whisker movement in awake , behaving rats , we trained rats to whisk in air and against textured surfaces . Two behavioral paradigms were used . In Behavior 1 , six rats ( N1–N6 ) positioned their nose in a small aperture ( the nose poke ) and whisked in air and onto surfaces for approximately 0 . 5 s to receive a water reward ( Figure 1A ) . Textured surfaces ( sandpapers of varying roughness , mounted on aluminum backing ) were positioned statically in the whisking path of the right whiskers using a computer-controlled stepper motor . Whisker motion in the protraction–retraction plane ( roughly rostrocaudal , parallel to the face ) was measured optically from whisker shadows cast by a collimated plane of laser light onto a linear charge-coupled device ( CCD ) imaging array below the training cage ( Figure 1C–1E ) . Each trial consisted of whisking either in air or onto one surface , and lasted 491 ± 179 ms . Between trials , rats moved to a separate chamber to receive a water reward , and surfaces were changed using the stepper motor . Rats performed 123 ± 43 ( mean ± standard deviation [s . d . ] ) trials per daily session . In Behavior 2 , four rats ( H1–H4 ) were habituated to being transiently head-fixed , and whisked voluntarily in air and onto surfaces ( Figure 1B ) . During each daily session ( 15–30 min ) , rats performed 69 ± 36 trials , with a trial defined as a 3-s epoch that included a variable duration of whisker motion . See Materials and Methods for training techniques . In both behaviors , motion of one to four identified whiskers was tracked at 4-kHz frame rate and approximately 5-μm spatial resolution , using the linear CCD array . Because whisker shadows did not cross during whisking , up to four whiskers could be identified and tracked simultaneously using automated software ( Figure 1D ) . Nonimaged whiskers were trimmed weekly at the base . Surfaces were presented parallel to the face , less than 5 mm from the whisker tips . Whisker motion was tracked 6–14-mm ( typically 10 mm ) from the face ( for whisking in air ) , and halfway between the surface and the whisker pad ( for whisking onto textures ) . All training and whisker measurements were performed under computer control using custom-written programs in Labview ( National Instruments ) . We first tested the resonance hypothesis by asking whether whiskers resonate , and whether a map of resonance frequency exists , in awake , behaving rats whisking in air . For rats whisking in the nose poke ( Behavior 1 ) , whisker motion typically included periods of regular , 5–15 Hz whisking , periods when the whiskers were held stationary , and periods of erratic motion . Examples are shown in Figure 2A . During all three types of motion , bandpass filtering ( 20–1 , 000 Hz ) revealed prominent high-frequency whisker vibrations ( 20–150 Hz ) , with approximate peak-to-peak amplitude of 0 . 1 to 0 . 5 mm , that were superimposed on the low-frequency motion ( Figure 2B ) . These high-frequency vibrations were not apparent during motion of an isolated whisker attached to an electric motor moving sinusoidally at 8 Hz ( Figure 2C ) , indicating that they were not due to external vibrations in the recording apparatus or to interaction between the moving whisker and air . To test whether vibrations were due to head motion , versus whisker motion relative to the head , we simultaneously measured head and whisker motion in one rat ( N4 ) by attaching a horizontal bar to the top of the skull . The bar cast a shadow on the CCD array that could be tracked independently of the whisker shadows . Head motion and D1 whisker motion showed coherence at low frequencies ( <8 Hz ) , but very little coherence ( mean 13% ) at frequencies greater than 10 Hz ( Figure 2D , green trace ) . The same was true for head motion and D3 whisker motion ( Figure 2E ) . Thus , head motion is not the source of high-frequency whisker vibration at greater than 20 Hz . Consistent with this conclusion , high-frequency whisker vibrations were also prominent in head-fixed rats whisking in air ( unpublished data ) . Despite the lack of coherence between whisker motion and head motion , neighboring whiskers exhibited high coherence in the 20–150 Hz range ( Figure 2D and 2E ) . This suggests a common driving force for high-frequency vibrations across whiskers . Coherence between whiskers decreased slightly with whisker separation on the face , suggesting that neighboring whiskers receive the greatest common drive ( Figure 2E ) . Together , these measurements indicate that high-frequency , coherent vibrations occur during free whisking in air , superimposed on low-frequency whisking motion . We tested for whisker resonance during free whisking in air by examining the relationship between the frequency spectrum of whisker vibrations in air and the intrinsic resonance frequencies of the whiskers . Experiments were performed in rats N1–N4 performing the nose poke task ( Behavior 1 ) . Power spectra during whisking in air were calculated for each whisker across 44–122 trials . Power spectra were not smooth , but rather showed modest peaks and shoulders representing dominant frequencies of whisker vibration . Two such power spectra , from the D1 and D2 whiskers imaged simultaneously in rat N2 , are shown in Figure 3A . Peaks and shoulders were identified precisely as minima in the second derivative of the logarithm of the power spectra , which correspond to points of negative concavity ( filled circles in Figure 3A and 3B ) . After whisking in air , rats were anesthetized; the intrinsic resonance frequency for each whisker was directly measured by manually delivering an impulse to the whisker , and the FRF from the resulting decaying oscillations in air was calculated , as imaged on the CCD array ( see Materials and Methods ) . Examples are shown in Figure 3C for the D1 whisker ( length: 38 . 7 mm , FRF: 44 . 1 Hz ) and D2 whisker ( length: 31 . 6 mm , FRF: 60 . 5 Hz ) from rat N2 ( same whiskers as for the power spectra in Figure 3A ) . Theoretical first harmonics of the FRF were calculated as ( 10 . 6/4 . 4 ) × FRF , as predicted for a conical beam model of the whisker ( see Materials and Methods ) . For this rat , peaks in the power spectra during voluntary whisking in air ( filled circles ) were found to align well with the measured FRFs ( open circles ) and calculated first harmonics ( asterisks ) obtained by the impulse method in the anesthetized animal ( Figure 3A ) . Across eight whiskers in rats N1–N4 ( n = 4 rats ) , the first peak in the power spectra during natural , active whisking aligned well with the measured FRFs ( Figure 3D ) , and the second peak aligned with the predicted first harmonics ( Figure 3E ) . Moreover , the ratio of the frequencies of the first and second peaks in the power spectra was 2 . 35 ± 0 . 14 ( n = 8 ) , close to the theoretical ratio of 10 . 6/4 . 4 = 2 . 41 for f1/FRF for a conical beam . Thus , during natural whisking in air , whiskers preferentially vibrated at the FRF and its first harmonic , though the magnitude of these vibrations was small . To test whether a map of whisker resonance exists across the whisker pad in the awake , behaving rat , we compared first peaks in the power spectra during whisking in air and FRFs measured by the impulse method , to whisker length ( Figure 4 ) . We found a systematic relationship in which longer whiskers exhibited lower FRFs and vibrated preferentially at these lower frequencies during whisking in air . This indicates that a map of resonance frequency exists across the whiskers in awake , whisking animals , as predicted by the resonance hypothesis [11–13] . To confirm the role of whisker resonance in shaping high-frequency vibrations in air , we systematically altered whisker resonance by trimming whiskers in two rats ( N2 and N3 ) . Power spectra were measured daily , for 10–11 d , during whisking in air for whiskers δ , D1 , D2 , and D3 . In rat N2 , the D2 and D1 whiskers were trimmed by approximately 2 mm after each day's measurement , while δ and D3 whiskers were left untrimmed . In rat N3 , δ and D1 were trimmed 2–4 mm shorter each day . Whisker length was measured daily . Results are shown in Figure 5 . The power spectrum for whisking in air on each day is presented as a color plot in each vertical strip . Whisker FRF was measured daily using the impulse method , and first and second harmonics of the FRF were calculated using a model of the whisker as a truncated ( i . e . , trimmed ) conical beam , rather than an intact conical beam ( [18]; see Materials and Methods ) . The FRF and first and second harmonics , calculated from the impulse measurements , are plotted as open circles , asterisks , and diamonds , respectively , on each day's power spectra . Results showed that as trimming decreased whisker length , power spectra for whisking in air shifted systematically towards higher frequencies , as expected if resonance filtering shaped whisker vibrations . Trimming shifted the shoulders of the power spectrum ( black filled circles ) in parallel with the resonance frequencies ( FRF and harmonics ) measured by the impulse method . This was particularly evident for the D1 whisker in rat N2 and the δ whisker in rat N3 , where bands of amplification ( shoulders ) in the power spectra closely followed the resonance frequencies measured by the impulse method . In contrast , power spectra remained stable for untrimmed whiskers , measured simultaneously in the same behavioral trials . In a converse experiment ( n = 1 rat ) , the D2 whisker was trimmed substantially , and then allowed to regrow by 12 mm over 14 d . Power spectra for whisking in air were measured before regrowth ( when the whisker was trimmed ) and afterwards . Results showed that regrowth was accompanied by a pronounced shift in the power spectrum of the D2 whisker towards lower frequencies , without substantial changes in the power spectra for nearby , simultaneously measured , untrimmed whiskers , whose length did not change appreciably during the regrowth period ( unpublished data ) . Together , these results indicate that resonance properties of whiskers shape high-frequency ( >20 Hz ) whisker vibrations during natural free whisking in air . This suggests that whisker resonance may be a relevant mechanism for filtering whisker input during active whisking in awake animals . High-frequency whisker vibration in air is not due to head movement ( Figure 1D and 1E ) , and therefore is likely to reflect high-frequency drive by whisker muscles . High-frequency muscular drive is plausible because high-frequency ( 83 Hz ) electrical stimulation of motor axons in the facial nerve can cause whisker movements at stimulation frequency [19] . We observed high-frequency whisker vibrations in response to facial nerve stimulation in anesthetized rats , and found that evoked vibrations can strongly drive whisker resonance ( Figure S1 ) . To test whether whisker muscles drive high-frequency whisker vibrations in awake , whisking rats , we measured electromyogram ( EMG ) activity from whisker muscles while imaging whisker motion in air ( n = 5 rats ) . EMG was measured from intrinsic muscles and the extrinsic muscle m . nasolabialis , which drive whisker protraction and retraction , respectively , during whisking [20 , 21] . In different rats , EMG was measured from m . nasolabialis , intrinsic muscles , or both simultaneously , together with the movement of one or two different whiskers ( Table 1 ) . In total , four m . nasolabialis EMG recordings were obtained simultaneously with movement of seven whiskers , and three intrinsic EMG recordings were made simultaneously with movement of six whiskers . EMG activity was coherent with whisking ( Figure 6A ) , as previously reported [20 , 21] , with intrinsic muscles generally active during protraction and m . nasolabialis active during retraction ( unpublished data ) . The rectified , differential EMG ( |∇EMG| ) power spectra revealed high-frequency muscle activity up to 50 Hz ( Figure 6B ) . To determine whether high-frequency muscle activity drove high-frequency whisker movement , we measured the spectral coherence between |∇EMG| and whisker position in air , during all types of whisker motion ( whisking , erratic , and flat ) . For both intrinsic and extrinsic muscles , coherence between |∇EMG| and whisker motion was generally statistically significant , with values between 0 . 15 and 0 . 65 , for frequencies less than 50 Hz , and fell below significance by approximately 50 Hz ( Figure 6C and 6D ) . This was true for both arc 1 ( D1 and C1 ) whiskers , and arc 2 ( D2 and C2 ) whiskers . The maximal frequency of significant coherence , termed the cutoff frequency , was defined as the frequency at which coherence magnitude fell below the p = 0 . 05 significance level for nonzero coherence ( see Materials and Methods for confidence interval calculation ) . For arc 1 whiskers , cutoff frequency was less than approximately 50 Hz for six of seven measurements , and approximately 90 Hz in the remaining measurement ( Figure 6E ) . Thus , muscle activity was significantly , but only modestly , coherent with whisker motion at the FRF of arc 1 whiskers ( median measured FRF: 36 . 9 Hz ) . Coherence at the FRF was even weaker for arc 2 whiskers , which also showed cutoff frequency of less than approximately 50 Hz in six of seven cases , and had a median FRF of 57 . 0 Hz ( Figure 6E ) . We conclude that whisker muscles provide some high-frequency energy that could drive whisker vibrations , but because coherence was weak at high frequencies , how muscle contractions drive high-frequency vibrations remains unresolved . The above results indicate that resonant motion is prominent during whisking in air , and that a map of resonance frequency exists across the whiskers . To determine whether this resonance map is used to encode surface texture , we explicitly tested the two central predictions of the resonance model for texture coding: first , that whiskers resonate at distinct , characteristic resonance frequencies as they sweep across surfaces; and second , that the amplitude of resonance frequency vibrations in each whisker depends on surface texture , resulting in one preferred texture that drives the strongest vibrations . Together , these properties have been proposed to result in a spatial map of texture across the whiskers [11 , 22] . We measured whisker motion on sandpaper surfaces in three rats performing the nose poke task ( rats N4–N6 ) and two rats that whisked while head-fixed ( rats H1–H2 ) . We used seven sandpapers: P150 ( roughest ) , P240 , P400 , P600 , P800 , P1200 , and P1500 ( finest ) . These correspond to 100- , 58- , 35- , 26- , 22- , 15- , and 13-μm mean particle size . Rats can readily distinguish two coarse sandpapers [3] , a smooth surface from P100 sandpaper [10] , and can distinguish 60-μm differences in spacing of periodic grooves [5] , suggesting that differences between these sandpapers ( or at least between the roughest and smoothest sandpapers ) should be discriminable using the whiskers . Up to four sandpapers were presented per measurement session , typically in blocks of five to ten trials each . Different subsets of sandpapers were presented on different days . Surfaces were placed parallel to the whisker pad , 5 mm closer to the face than the whisker length . ( Because whisker tips move in an arc , this meant that approximately 5 mm of whisker tip contacted the surface at mid-whisk , and less than 5 mm contacted at maximum protraction and retraction ) . Because whiskers are different lengths , we measured movement of only a single whisker at a time across the surfaces . We verified continuous whisker–surface contact during whisking in each animal , by observing the presence of a consistent whisker shadow on the CCD imaging array when the array was positioned 1 mm from the surface . It was not possible to position surfaces closer to the whisker tip , given the lateral freedom of head position within the nose poke ( approximately 2 mm ) . Whisker motion was measured halfway between the surface and the whisker pad ( ∼10 mm from the follicle ) . An example of whisker motion across a rough ( P150 ) sandpaper is shown in Figure 7A . The rat initially retracted the D3 whisker across the surface ( negative slope in the position trace ) , and then protracted it ( positive slope ) . Whisker velocity and acceleration , calculated from the position trace , revealed approximately three brief , high-acceleration , high-velocity events that occurred during whisker motion . To analyze the time-varying spectral content of whisking on the surface , we calculated the Wigner-Ville time-frequency representation ( TFR ) , qualitatively similar to a spectrogram , for this whisker motion ( Figure 7B ) . The TFR showed prominent , brief epochs of vibration at approximately 150–180 Hz , aligned with the rapid movement events . The integrated TFR across the entire whisking period ( which is equal to the average power spectrum ) revealed a broad peak at approximately 150–180 Hz ( Figure 7B , rightmost trace ) . Similar broad peaks at 50–180 Hz were observed for D1 , D2 , and D3 whiskers moving across a variety of surfaces ( see below ) . We tested whether these broad , high-frequency peaks were consistent with whisker resonance by comparing the peak frequencies across different length whiskers . ( The FRF while the whisker is pinned against a texture will not equal the FRF measured in air , because boundary conditions for vibration are changed and the whisker is effectively shortened [11 , 23 , 24] ) . Figure 7C and 7D show power spectra for vibrations of the D1 and D3 whiskers of rat N4 , measured during palpation on five different sandpapers . Power spectra were calculated as integrated TFRs for all individual protraction and retraction epochs on a given surface , and then averaged across these epochs to obtain the average power spectrum for each surface . Consistent with the resonance model , the D1 whisker showed a high-frequency vibration peak at approximately 80 Hz , while the shorter D3 whisker showed a peak of approximately 150 Hz ( Figure 7C and 7D ) . We repeated this analysis for ten whiskers ( five D1 whiskers , four D2 whiskers , and one D3 whisker ) in five rats ( N4–N6 , H1 , and H2 ) . We calculated the high-frequency peak of the average power spectrum for each whisker moving on each texture ( identified as the first peak in the power spectrum >40 Hz ) . Across all textures , the high-frequency peak for D1 whiskers was found to be between 57 . 7 and 91 . 2 Hz ( mean: 71 . 8 Hz ) ; for D2 whiskers , 68 . 5–114 . 0 Hz ( mean: 86 . 6 Hz ) ; for the single D3 whisker , 142 . 2–153 . 9 Hz ( mean: 147 . 9 Hz ) ( Figure 8A ) . Thus , for both animals performing the nose poke behavior ( open circles , Figure 8A and 8B ) and head-fixed animals ( asterisks , Figure 8A and 8B ) , measured peaks in vibration power spectra were at higher frequencies for the shorter whiskers and lower frequencies for the longer whiskers , consistent with intrinsic resonant properties of the whiskers . These data therefore suggest that whiskers vibrate at characteristic resonance frequencies when moving across surfaces , at least when distance to the surface is kept constant . Subsequent analyses assume that the high-frequency vibration peak represented the whisker's resonance frequency on surfaces . Power at the high-frequency peak during whisking on textures was 6 . 7 ± 3 . 4 ( mean ± standard error ) times greater than power at the resonance frequency during whisking in air ( unpublished data ) . Finally , we tested whether the amplitude of resonance frequency vibrations in each whisker depends on , and encodes , surface texture , as posited by the resonance hypothesis [11 , 13] . In isolated whiskers and anesthetized animals , prolonged , stable application of different texture or vibratory stimuli to the tip of a single whisker generates up to a 10-fold difference in steady-state power at the whisker's resonance frequency , indicating strong tuning for specific textures or vibration frequencies [11 , 22] . In contrast , we found that during natural whisking , the power spectrum for whisker vibrations in a single whisker was remarkably constant across different surfaces ( e . g . , the five sandpapers in Figure 7C and 7D ) . We calculated the power at the presumed resonance frequency as a function of texture for all sandpapers that were presented to each animal . Across the ten whiskers ( rats N4–6 , H1 , and H2 ) , no substantial or systematic relationship between sandpaper grade and vibration power at presumed resonance frequency was observed , either for nose poke or head-fixed rats ( Figure 8B ) . On average , the maximal change in power at the presumed resonance frequency between any two textures for individual whiskers was 49 ± 29% ( mean ± s . d . ) . This is substantially less than the 10-fold variation observed with prolonged , regular stimulation in anesthetized animals and detached whiskers . Two-way ANOVA found no significant differences in power at the presumed resonance frequencies in each behavioral trial for either whisker type ( D1 , D2 , or D3 ) ( F ( 2 , 1624 ) = 0 . 45; p = 0 . 64 ) or sandpaper grade ( P150 through P1500 ) ( F ( 6 , 1624 ) = 1 . 18; p = 0 . 32 ) . Similar analysis of normalized power at the presumed resonance frequency ( normalized to total spectral power , which controls for trial-to-trial variability in total vibration power ) produced identical results ( unpublished data ) . These results indicate that during active whisking under our experimental conditions , whiskers resonate on textures , but resonance magnitude is independent of texture roughness . This is contrary to the expectation of the resonance model for texture coding , which predicts that sustained whisker-tip movement over texture spatial features leads to regular whisker vibrations whose amplitude builds up most effectively when vibration frequency matches whisker resonance frequency [25] . One potential explanation of the present result is that resonance frequency vibrations do not build up in a gradual , sustained manner during natural whisking , but represent transient responses ( ringing ) to discrete high-acceleration , high-velocity events . Such events were a prominent feature of whisker movement across surfaces ( e . g . , Figure 7A and 7B ) , and were commonly associated with transient high-frequency ringing in whisker position , acceleration , and velocity . Representative examples of this behavior measured during protraction of D1 and D3 whiskers on a P150 sandpaper are shown in Figure 9A ( see also Figure 7A and 7B ) . TFRs of these representative events revealed postevent ringing of the D1 whisker at approximately 90 Hz , and postevent ringing of the shorter D3 whisker at approximately 175 Hz ( Figure 9B ) . To determine whether transient ringing induced by these discrete motion events was a significant source of overall resonance vibrations during texture palpation , we compared vibration power spectra in 0 . 4-s epochs centered on high-acceleration events ( defined here as movement events in which acceleration magnitude exceeded mean acceleration by 2 s . d . ) versus equivalent epochs of whisker retraction or protraction when no high-acceleration event occurred . Results showed that power at the high-frequency peak was , on average , 12 . 8 ± 2 . 6 times greater in epochs containing high-acceleration events versus epochs that lacked such events , and 3 . 8 ± 0 . 8 times greater versus all whisking epochs , regardless of whether they contained an acceleration event ( n = 10 whiskers , 5 animals ) . This result is shown for D1 and D3 whiskers of rat N1 in Figure 9C , and for all whiskers in Figure 9D . Together , these results demonstrate that resonance vibrations in whiskers during texture palpation primarily represent transient ringing following discrete high-acceleration movement events , and that the amplitude of resonance vibrations does not vary across the range of sandpapers that were tested . Discrete high-acceleration motion events were prominent on textures , but were generally absent during whisking in air . Representative whisker motion in air and on a rough ( P150 ) sandpaper are shown in Figure 10A . In this example , large-acceleration events ( acceleration > 4 s . d . in air; green dots ) occurred 2 . 5-fold more often on the sandpaper than in air . Acceleration events of all magnitudes occurred more frequently on the texture versus air for this whisker ( Figure 10B , texture: 315 trials , 340 s of whisker-movement data; air: 111 trials , 223 s ) , and the highest acceleration events ( >0 . 3 mm/ms2 ) were detected predominantly during whisking on texture ( inset ) . Although we use acceleration as a convenient marker for these motion events , whisker acceleration and velocity were well correlated in whisker-motion traces ( Figure S2 ) . High-acceleration events occurred during both protraction and retraction , and could be classified into slips ( events in which whisker speed suddenly increased in the direction of whisker motion ) and sticks ( events in which speed suddenly decreased , corresponding to sudden stopping of whisker movement ) . Examples of slips and sticks during protraction and retraction are shown in Figure 10C . The average kinematics of slips and sticks during protraction and retraction are shown in Figure 10D , for the D2 whisker in rat H1 moving across four sandpapers . For each type of event , separate averages were calculated for five ranges of acceleration magnitude . ( High-acceleration events correspond to more abrupt slips and sticks . ) The average position and acceleration traces revealed that whisker slips were followed , on average , by sticks , and sticks were preceded by slips . Thus , sequences of high-acceleration events represented slip-stick motion of whiskers along surfaces . Slips occurred during all phases of protraction and retraction ( Figure 10E ) . To determine the average size and time course of a slip , we compiled histograms of slip magnitudes , durations , and peak speed ( |velocity| ) , for all rats and all whiskers ( n = 10 whiskers , 5 rats ) , including all slip events with acceleration greater than 2 s . d . of the acceleration in air ( Figure 10F ) . Slip duration was defined from the initial acceleration peak to the time when whisker speed returned to the average speed . An example of the calculation of slip magnitude and duration is shown in Figure 10F ( upper left ) . Results showed that during the average slip event , the whisker traveled a mean of 1 . 9 mm , in a mean of 8 . 6 ms , and achieved a peak speed of 0 . 33 ± 0 . 24 mm/ms , before whisker speed returned to average . We tested whether slip-stick events could provide an alternate , nonresonance-based code for surface texture . A slip-stick code is plausible since sharp , high-acceleration , and high-velocity events effectively drive spikes in somatosensory cortex [7 , 26 , 27] , and thus the pattern of slip-stick events is likely to be encoded in the rat's central nervous system ( CNS ) . We again used acceleration to identify these events . We compared acceleration events on four sandpaper textures ( P150 [very rough] , P400 , P800 , and P1200 [very smooth] ) that were interleaved in blocks for each rat within a single day ( five or ten trials per block ) . This measurement was performed for the D1 and D2 whiskers in three rats ( N6: 89–103 trials per texture , H1: 52–56 trials per texture , and H2: 40–43 trials per texture ) . Analysis was restricted to within-day comparisons across textures to avoid complications from day-to-day variability in whisking behavior . For this analysis , an acceleration event was defined as any acceleration peak that crossed a defined threshold , with a minimum of 2 ms between events , and stick versus slip events were not distinguished . Motion of the D2 whisker in rat H2 across a smooth ( P1200 ) and rough ( P150 ) sandpaper is shown in Figure 11A and 11B . ( This is the same whisker whose motion in air and on P150 sandpaper was shown in Figure 10A . ) Low-acceleration events ( red dots , peak acceleration 0 . 062–0 . 248 mm/ms2 , corresponding to 1–4 s . d . above zero on the P1200 surface ) occurred on both textures , as well as in air . In contrast , high-acceleration events ( green dots , >0 . 496 mm/ms2 , corresponding to 8 s . d . above zero on the P1200 surface ) occurred preferentially on the rough P150 sandpaper . This suggested that high-acceleration events may occur systematically more frequently on rougher surfaces . We calculated the average incidence of different magnitude acceleration events on P150 , P400 , P800 , and P1200 textures , as well as during whisking in air , for six whiskers in three rats ( rat N6 performing the nose poke task , and rats H1 and H2 whisking while head-fixed; D1 and D2 whisker motion was measured in each animal ) ( Figure 11C and 11D ) . The number of acceleration events surpassing different absolute acceleration thresholds was calculated per sweep , where a sweep was defined as a single whisker protraction or retraction . Results showed that the total number of acceleration events surpassing low acceleration thresholds ( e . g . , 0 . 1 mm/ms2 ) was not different between whisking in air and whisking on surfaces , but the number of events surpassing high acceleration thresholds ( e . g . , 0 . 4 mm/ms2 ) was higher on surfaces than in air , and was systematically higher on rougher versus smoother surfaces ( Figure 11C and 11D ) . Statistical analysis showed that low-acceleration events ( with peak amplitude in the range 0 . 062–0 . 248 mm/ms2 , corresponding to 1–4 s . d . above zero ) were equally prevalent in air and on smooth P1200 and P800 surfaces , but were significantly less prevalent ( asterisks; Mann-Whitney U-test , p < 0 . 01 ) on the rougher P400 and P150 surfaces , especially for the D2 whisker ( Figure 11E ) . Conversely , high-acceleration events ( >0 . 496 mm/ms2 , corresponding to 8 s . d . above zero ) were systematically more prevalent on rougher versus smoother surfaces , for both D1 and D2 whiskers ( Figure 11F ) . As a result , the ratio of high to low acceleration events per sweep increased systematically and significantly with surface roughness ( Figure 11G; asterisks indicate significant differences in ratio between pairs of textures ) . These relationships between slip-acceleration magnitude/frequency and surface roughness held true for both the nose poke rat ( N6 ) and head-fixed whisking rats ( H1 and H2 ) ( unpublished data ) . These results suggest that either the frequency of high-acceleration events or the relative frequency of high to low acceleration events may contribute to a kinetic signature for surface roughness [7] , independent of whisker resonance .
We found that whiskers exhibited high-frequency ( >20 Hz ) vibrations during active whisking in air , and that the spectral composition of these vibrations varied with whisker length , due to filtering by whisker resonance ( Figures 2–5 ) . Thus , whiskers resonate during natural whisking in air , and a map of whisker resonance exists in awake , whisking rats ( Figure 4 ) . High-frequency vibrations were coherent across neighboring whiskers , were not caused by head motion or interactions between whisker and air ( Figure 2C–2E ) , and could be elicited by high-frequency stimulation of the facial nerve in anesthetized animals ( Figure S1 ) . This suggests that vibrations are due to neurally or mechanically coordinated drive of neighboring whiskers by whisker facial muscles . EMG recordings of extrinsic and intrinsic muscles detected high-frequency components of muscle contraction . However , when we measured spectral coherence between whisker vibrations and EMG activity , we found only modest coherence for frequencies up to approximately 50 Hz ( near the FRF of arc 1 whiskers ) , and nonsignificant coherence for frequencies greater than 50 Hz ( near the FRF of arc 2 and shorter whiskers ) ( Figure 6 ) . This suggests either that ( 1 ) additional coherent , high-frequency muscular drive exists , but was not detected by the EMG recordings , or that ( 2 ) muscles drive resonance vibrations noncoherently , as could occur if sharp , pulsatile muscle contractions induced higher frequency vibrations and excited whisker ringing at the resonance frequency . This latter case is less likely because whisker motion , and muscle drive , are relatively smooth during exploratory whisking . However , sharp contractions may occur during more erratic whisker motion . Resonance vibrations also occurred during active whisking on sandpaper surfaces , as inferred from the presence of spectral peaks in whisker vibration at specific supra-whisking frequencies , with longer whiskers vibrating at low frequencies , and shorter whiskers vibrating at higher frequencies ( Figures 7 and 8 ) . Thus , resonance filters whisker vibrations during whisking onto surfaces . However , resonance vibrations occurred primarily as transient , sporadic ringing events , rather than as sustained oscillation , and neither the amplitude of vibrations at presumed resonance frequencies nor the overall power spectrum varied with texture across a wide range of sandpaper grades ( Figures 7 and 8 ) . Thus , each whisker was not preferentially excited by a specific set of textures . We conclude that differences between sandpaper textures are not encoded by relative vibration amplitude across facial whiskers , at least in the geometrical and behavioral conditions of our study . These data argue against the resonance hypothesis for texture coding . However , they do demonstrate that whisker resonance occurs during surface palpation , and therefore may play a role in amplifying some types of whisker responses [13] . These results confirm a recent study that detected resonance vibrations on textured surfaces , but did not examine whether resonance encoded texture [17] . The critical difference between our results and the resonance hypothesis appears to be in how resonance vibrations are generated during whisker–surface interaction . Linear resonating systems can resonate in two distinct modes: In the transient mode , oscillations are triggered by discrete external impulses , and occur transiently after these impulses , in the absence of additional external vibratory forces . In this case , oscillation dynamics are determined solely by the intrinsic properties of the system , as in the case of transient resonant ringing of a tuning fork after being struck by an object . In the steady-state mode , in contrast , vibrations are produced in an ongoing manner during sustained external vibratory drive . In this case , vibratory responses occur at the same frequency as the external impulses , and vibration amplitude is much larger when external vibrations occur at the intrinsic resonance frequency of the system . The resonance hypothesis assumes that passage of a whisker over a surface generates sustained tip vibrations as the whisker interacts with surface microfeatures , and that this causes steady-state resonance to build up in the whisker . Such steady-state resonance indeed occurs when sustained vibrations are applied to isolated whiskers or to nonmoving whiskers in anesthetized animals [11 , 12 , 22] . However , our results demonstrate that voluntary whisker motion produces discrete , high-acceleration slip-stick events , rather than smooth motion across surfaces ( Figure 10 ) . These slip-stick events drive transient ringing , and this transient ringing is the major source of whisker resonance on surfaces ( Figure 9 ) . The dominance of transient resonance , as opposed to sustained resonance , explains why whisker vibrations vary with intrinsic properties of the whiskers ( Figure 8A ) , but not with surface texture ( Figure 8B ) . These results confirm a previous observation that sustained resonance vibrations do not appear during voluntary whisking on surfaces [12] . Together , these data indicate that whisker resonance occurs in awake animals , both during whisking in air and on surfaces , and may contribute to encoding or amplification of certain aspects of whisker input . However , differential whisker resonance does not encode texture in these behavioral conditions and using these sandpaper surfaces , which are predicted to be discriminable by rats [4 , 5 , 10] . We cannot rule out that , under conditions of behavioral discrimination , rats may adopt a different whisker exploration strategy that may enable resonance-based coding of texture . However , recent studies of texture discrimination have provided no evidence for coding by resonance [10 , 17] , and two arguments suggest that such a coding strategy may be problematic: first , the relationship between whisker resonance frequency and effective whisker length ( Figure 8A ) suggests that any trial-to-trial variation in surface position or angle relative to the face will alter whisker resonance frequency , making it difficult to construct a position-independent resonance code for texture . Second , rats discriminate textures even with substantial trial-to-trial variation in whisking speed [5] . Such variation will alter the relationship between texture spatial frequency and whisker-tip vibration frequency , making it unlikely that a whisker could be “tuned” for a specific texture . A common feature of whisker motion across sandpapers were discrete , high-acceleration slip and stick events ( Figure 10 ) . Slip and stick events occurred during all phases of whisker protraction and retraction ( Figure 10E ) . These events often generated high-amplitude transient ringing at the whisker's resonance frequency ( Figures 7A and 9 ) . Slip-stick events were frequent: for example , 1 . 2 events with acceleration greater than 0 . 4 mm/ms2 occurred per protraction–retraction cycle for the D1 whisker , averaged across all sandpaper surfaces ( Figure 11C ) . This corresponds to approximately 30 events when all 25 large whiskers on each side of the face are considered . These events have also been observed during whisking onto surfaces under very different geometrical and behavioral conditions [17] , and are therefore likely to be basic common elements of the whisker input stream . The average slip was 1 . 9 mm ( measured at the whisker midpoint , ∼10 mm from the follicle ) , and lasted 8 . 6 ms before whisker velocity returned to its mean value ( Figure 10F ) . This corresponds to a mean angular displacement of 10° and a mean velocity of 1 , 100°/s during slips . Peak velocity during slips was 0 . 33 mm/ms . This amplitude and velocity are well within the range of behavioral detectability [30] and spike encoding at primary afferent and cortical levels [26 , 27 , 30] . Thus , slip-stick events are likely to be encoded in the CNS . These slip-stick events are similar to velocity transients observed during artificial whisking onto textures in anesthetized rats [7 , 15] . Because high-acceleration events occur more frequently on textures than in air ( Figure 11 ) , we propose that slip-stick events may encode the presence of a surface , or surface properties , on the whisker array . The kinetic signature hypothesis for texture coding proposes that textures generate unique , identifiable temporal patterns of whisker vibration ( “kinetic signatures” ) in single whiskers , and that these temporal features are encoded in neural spike trains . Candidate components of kinetic signatures for texture include the spectral composition of whisker vibration [15] , the mean speed of whisker vibration [7 , 29] , and the temporal profile of velocity transients [7] . These features vary when whiskers of anesthetized rats are artificially swept across different textures by electrical stimulation of the facial motor nerve , with rough versus perfectly smooth textures generating differences in mean vibration speed [7 , 29] , and finer texture differences ( e . g . , between sandpaper grades ) generating unique temporal profiles of whisker velocity [7] . Our data suggest that slip-stick events may contribute to the kinetic signature for texture . The magnitude and frequency of these events were correlated with texture , with rougher sandpapers eliciting a greater frequency of high-acceleration events ( which tend to also be high-velocity events ) , and a higher proportion of high-acceleration versus low-acceleration events , compared to smoother sandpapers and to air ( Figure 11 ) . This relationship between slip acceleration and texture is expected from a simple model in which rougher surfaces , which have greater friction , require more forward force during whisker protraction ( or retraction ) to overcome static friction and move the whisker tip forward ( or back ) . This increased forward force translates into increased acceleration during forward slips . Thus , more high-acceleration slips , and fewer low-acceleration slips , are predicted on rougher textures . This significantly extends a prior study showing more high-speed slip events on a rough surface versus a completely smooth one [17] . We propose that slip magnitude ( acceleration or velocity ) and frequency are components of the kinetic signature for texture in the whiskers , and that coding of these parameters by S1 neurons provides information about surface texture . In anesthetized animals , whisker deflections evoke phasic , single-spike responses in S1 neurons , with spiking probability positively correlated with whisker velocity and acceleration over the ranges of 0 . 02–1 . 0 mm/ms [31 , 32] and approximately 20–500 m/s2 [33] , respectively . The range of slip speeds and accelerations observed here ( ∼0 . 1–0 . 5 mm/ms and ∼100–1 , 000 m/s2 ) fall within this dynamic range . Thus , the occurrence and magnitude of slips are likely to be encoded by time-locked spikes in S1 ensembles , with texture-related sequences of slip-stick events ( Figure 10A ) encoded by temporal sequences of spikes ( constrained by the intrinsic dynamics of whisker circuits and synapses ) . The occurrence of discrete slip events related to texture , observed here under two behavioral conditions , suggests a potential temporal spike code for texture during awake , active sensation . Such a temporal code has been suggested from S1 recordings in anesthetized rats during electrically evoked whisking on texturally similar surfaces , like the sandpapers used here [7] . In contrast , active whisking onto very distinct textures ( rough vs . smooth glass ) evokes subtly , but significantly different , mean firing rates in S1 [10] . Slip-evoked spikes could drive such texture-specific changes in firing rate , depending on neural sensitivity to slip amplitude and velocity .
Two types of whisker behavior were studied . In Behavior 1 ( whisking in nose poke , six rats ) , rats were trained using operant conditioning techniques to place their nose in a small port ( the nose poke ) and whisk for approximately 1 s in air or on textured surfaces . The behavioral apparatus , modeled after [34] , consisted of an outer reward chamber containing a solenoid-gated drink port , and an inner measurement chamber containing the nose poke , texture stimuli , and whisker-motion recording system ( Figure 1A ) . Rats received water ( 50 μl ) as reward during behavioral training ( 1 h per day ) and during a 1-h ad lib drinking period following each behavioral training session , but not during the remaining 22 h per day , 5 d per week . Water was freely available on weekends . Rats on this regimen were healthy and alert , and gained weight daily . Rats ( age 30 d ) were initially accommodated to handling ( 3–5 d ) and to the behavioral apparatus . Rats were then trained to drink from the drink port in response to a white noise tone ( WNT ) . A phototransistor in the drink port signaled the rat's presence and gated water delivery . Next , rats were trained to nose poke to trigger the WNT and water delivery to the drink port . A phototransistor in the nose poke reported nose poke occupancy . Finally , rats were trained to gradually increase nose poke duration and to actively whisk while in the nose poke . Gross whisking was assessed by four phototransistors that generated voltage pulses when the whiskers passed over them . The number of phototransistor pulses required to trigger the WNT and drink port water delivery was gradually increased until rats were whisking in the nose poke for approximately 0 . 5 s . Each approximately 0 . 5-s bout of whisking in the nose poke was considered a trial , and trials were separated by the rat retreating to the reward chamber to drink . Trained rats performed 80–150 trials per day . Total training time ( after accommodation ) was approximately 23 d . Whisker motion was recorded optically in trained rats whisking in air and whisking onto textures . Textures were 6 × 6-cm sandpapers of grade P150 , P240 , P400 , P600 , P800 , P1200 , and P1500 glued to an aluminum plate and positioned in the whisking path of the right whiskers 5 mm from the whisker tips , parallel to the face . Up to four different textures were mounted on a four-arm Plexiglas holder attached to a stepper motor ( Oriental Motor , PK264B1A-SG10 ) . Textures were rotated into place between trials while the rat was at the drink port . Surface positioning relative to the nose poke was performed as follows: first , using videography , we measured the mean position and orientation of the external edge of the whisker pad while the rat was performing the whisking behavior . Surface orientation was set parallel to the whisker pad . Next , we transiently anesthetized the rat and measured the length of the whisker to be studied ( whisker movement on surfaces was measured for a single whisker at a time ) . We positioned the stepper motor so that the point on the surface closest to the face ( i . e . , the point at the intersection of the surface and of the whisker , when the whisker was normal to the face ) was located 5 mm closer to the whisker pad than the whisker length . Surface positioning was verified by imaging the whisker 0 . 5 mm from the surface , and confirming that the whisker shadow disappeared from the imaging plane when the surface was moved approximately 5 mm from its set position . Whisking in air was measured by rotating the stepper motor into a position with no texture present . Thus , up to four textures ( or three textures plus air ) could be interleaved under computer control during a recording session . Training and recording procedures were controlled by custom routines in Labview ( National Instruments ) . In Behavior 2 ( whisking while head-fixed , four rats ) , rats ( age 30–40 d ) were accommodated to handling ( ∼1 wk ) , and to being placed for 15 min in a loose fabric sack from which the head emerged ( ∼1 wk ) [35] . Rats were habituated to being placed , while in the sack , in a 5-cm–diameter Plexiglas tube ( Figure 1B ) . Rats then underwent surgery to implant electromyogram ( EMG ) recording electrodes ( see below ) , during which a small screw was affixed to the skull with dental acrylic . After 4–6-d recovery from surgery , rats were placed again in the Plexiglas tube , and the head was stabilized via the screw ( Figure 1B ) . Head-fixed rats naturally whisked in response to objects held in front of them . Whisker motion was measured during these whisking epochs . Recording sessions typically lasted 15–30 min . Whisker motion was recorded in 3-s trials with approximately 50–100 trials per recording session . The animal was positioned so that the head and whiskers were in the same spatial relationship to the textures and CCD imaging array as in Behavior 1 . Textures ( or air ) were presented in blocks . For both Behaviors 1 and 2 , behavioral training was performed with all whiskers intact . The day before whisker-motion measurement , rats were transiently anesthetized with isoflurane , and all whiskers whose motion was not being studied were trimmed at the base . For Behavior 1 , all but one to four whiskers ( δ , D1 , D2 , and D3 ) were trimmed . For Behavior 2 , all but two to three whiskers in the C row or D row were trimmed . Whisker motion was measured in one dimension by casting shadows of the whiskers onto a linear CCD imaging array . The light source was a diode laser ( 670 nm ) , positioned above the rat and focused into a collimated line 60-mm long and 1-mm wide , using two cylindrical lenses rotated 90° from one another ( Figure 1C ) . Below the whiskers , a third cylindrical lens focused whisker shadows onto the linear CCD array ( Fairchild imaging , CCD 133AEDC , 1 , 060 elements , 13-μm width per element ) . The output of every other CCD element was sampled at 4-kHz frame rate using custom-built electronics ( UCSD Physics electronics shop ) and a National Instruments data acquisition card ( PCI 6111 ) . Voltage traces from the array were stored and processed offline to determine whisker position . In Behavior 1 , whisker position was recorded for 1 . 5 s starting with nose poke onset ( analysis was restricted to the epoch during which the rat remained in the nose poke ) . In Behavior 2 , whisker motion was recorded in 3-s blocks . The CCD array was positioned parallel to the whisker pad , either approximately 10 mm ( range: 6–14 mm ) from the whisker pad ( whisking in air ) or at the midpoint between the texture and the whisker pad ( whisking on texture ) . Whisker contact with textures was verified for each rat by the consistent presence of whisker shadows when the array was positioned 1 mm from the texture . Each frame of CCD output was subtracted from a baseline CCD image obtained when no whiskers were present ( baseline images were obtained several times during each recording session ) . Whisker shadows appeared as discrete voltage peaks in the baseline-subtracted CCD image , with each shadow covering eight to ten CCD pixels . Position of each whisker shadow was calculated as the weighted mean of all pixels in the whisker shadow , weighted by pixel voltage . Whisker spatial position was calculated from whisker-shadow pixel position via a calibration curve obtained using a 0 . 5-mm spaced wire grid held at whisker position . Repeated measurements showed that whisker position was determined with a spatial resolution of approximately 5 μm . Whisker motion over time was computed algorithmically using custom software in Matlab . Up to four whisker shadows could be tracked simultaneously and identified unambiguously using this method . All but the imaged whiskers were trimmed weekly to the level of the skin , during transient isoflurane anesthesia ( 4% in 2 l/min O2 , delivered via a nose cone ) . If a whisker transiently left the imaging plane , whisker motion was only analyzed up to that point . To measure whisker FRF using the impulse method , rats were anesthetized with isoflurane and the head positioned in the behavioral apparatus at the standard position and angle relative to the CCD array . An impulse was delivered manually to each whisker , and the resulting decaying oscillation was measured with the CCD array . The FRF was calculated as the inverse of the average time between peaks in the oscillations [11] . For untrimmed whiskers , the first and second harmonics of the resonance frequency were calculated as: f1 = ( 10 . 6/4 . 4 ) FRF and f2 = ( 19 . 2/4 . 4 ) FRF , as predicted by theoretical models of tapered beams [23] . For trimmed whiskers , which are truncated tapered beams , the ratios f1/FRF and f2/FRF are functions of the truncated length . We calculated f1 and f2 for trimmed whiskers by interpolating f1/FRF and f2/FRF ratios that were numerically calculated by Conway et al . [18] for four different ratios of truncated to untruncated length of a thin conical beam . Whisker length was measured while rats were anesthetized . Length was measured from the skin surface to the whisker tip , using calipers and 4× magnification under a dissecting microscope . In some experiments , EMG activity was recorded from whisker pad muscles . EMG electrode implantation followed Berg and Kleinfeld [21] . Briefly , surgery was performed using sterile technique , under ketamine/xylazine anesthesia ( 90 and 10 mg/kg , respectively , i . p . ) . Supplemental ketamine ( 20 mg/kg ) was administered approximately every 2 h to maintain anesthetic depth , determined by absence of limb withdrawal reflex and breathing rate of 45–60 breaths per min . EMG electrodes were made from Teflon-coated tungsten microwire ( 0 . 002” diameter; California Fine Wire; 1 mm of insulation stripped at recording tip ) . Microwires were implanted in pairs to record the differential EMG signal . Microwires were implanted via a midline incision at the top of the skull , and a lateral incision caudal to the mystacial pad . One electrode pair was implanted in the extrinsic muscle m . nasolabialis , by exposing this muscle and pressing the recording tips into muscle tissue . Microwires were secured at the muscle entry point using 6–0 Ethicon nylon sutures ( Johnson and Johnson ) . To record EMG in intrinsic muscles , microwire pairs were threaded through a 26-ga targeting needle , which was used to insert wire tips into the whisker pad [21] . Wire tips were bent back at the needle tip to anchor the wires to whisker pad tissue . Wires were sutured in place where they exited the pad . Reference wires ( stripped of 4 mm of insulation ) were implanted in the dermis at the tip of the snout , rostral of m . transversus nasi . Microwire tip position was verified at the end of surgery by passing current to stimulate the muscles and evoke appropriate whisker and pad movements . Microwires were soldered into a ten-pin connector ( Samtec ) attached to the skull . Bupivicaine ( 0 . 1 ml ) was administered for postoperative analgesia . EMG recording commenced 4–6 d after EMG implantation . EMG data were collected in behaving rats in 3-s–long blocks , simultaneous with whisker-motion data . EMG signals were amplified ( 20× gain ) and impedance buffered using an eight-channel head-mounted headstage amplifier ( Plexon Instruments HST/8o50-G20 ) . Headstage output was transmitted via twisted thin-gauge wires to a second amplifier and bandpass filter ( Plexon Instruments PBX2/16sp-G50 ) ( 50× gain , 0 . 3–8 kHz bandpass ) . Amplifier output was digitized at 32 kHz ( National Instruments PCI 6259 ) . Analysis was performed on rectified , low-pass filtered ( 500 Hz cutoff ) difference of neighboring raw EMG signals , denoted |∇EMG| . EMG and whisker data acquisition were performed on separate , synchronized data acquisition cards . Power spectra of whisker motion in air were calculated using the multitaper estimation technique of Thomson ( 1982 ) [36] . Briefly , a whisker motion time series recorded during a single trial was first multiplied by a set of K orthogonal data tapers . The Fourier transform of each tapered time series was calculated using the Fast Fourier Transform algorithm in Matlab , and from each Fourier Transform , the power spectrum was estimated as the modulus squared of the Fourier Transform . The estimated power spectrum of the whisker motion for the nth trial , Z , was then an average over these K power spectra where Yn , k ( f ) is the Fourier Transform of the kth tapered time series of the nth trial . The average power spectrum of whisker motion for a single day's recording session was then an average over all trials performed on that day: where SY ( f ) is the average power spectrum of the whisker motion and N is the total number of trials . Here , N was typically between 50 and 150 trials , and the number of tapers , K , was 5 . With this method , the resulting average power spectrum of a time series of duration T is smoothed over a half-bandwidth of The spectral coherence C ( f ) between the |∇EMG| and whisker motion was calculated similarly to [37] , where Z ( f ) is the Fourier Transform of the |∇EMG| time series and SZ ( f ) is the average |∇EMG| power spectrum . The theoretical confidence intervals for coherence were computed following [38] , where it is estimated that the coherence magnitude will exceed in P × 100% of measurements . Here , we take p = 0 . 05 . Spectral estimation was performed using the Chronux ( http://www . chronux . org ) and signal processing toolboxes in Matlab . The Fourier methods described above are appropriate for describing the average spectral properties of stationary signals [38] . We used the Wigner-Ville TFR to examine the brief transient ringing events during whisking onto textures . The TFR is known to provide good localization in both time and frequency , and is better suited for analyzing time series with time-varying frequencies [39] . This distribution is computed by correlating the entire time series y ( t ) , with a time-translated version of itself and taking the Fourier transform of this locally autocorrelated function , The color plots of the TFR were smoothed over time ( windowsize = 2 . 5 ms ) and frequency ( windowsize = 20 Hz ) for visualization . Power spectra of whisker motion onto textures were calculated by numerically integrating the unsmoothed TFR ( t , ω ) over time [39] . In artificial whisking experiments [40] , an incision was made in the side of the snout posterior to the whisker pad of the anesthetized animal ( urethane 1 . 5 g/kg ) . The buccal motor nerve was separated from the underlying muscle and cut to prevent antidromic activation of the motor nerve [19] . The lower branch of the buccal nerve was also cut , which generated more-natural , horizontal whisks than with this nerve intact . The distal portion of the facial nerve was isolated in a stimulating cuff with electrodes placed around the nerve . Saline was applied to keep the nerve moist . The nerve was stimulated with brief monophasic pulses ( 50-μs duration , 3–6 V ) from a Grass Stimulator ( Model S88K ) . Pulses were generated at 110 Hz in bursts lasting 50 ms followed by 50 ms with no stimulation . Whiskers protracted during the bursts and passively retracted during the 50 ms following the bursts , generating 10-Hz artificial whisking .
|
A fundamental problem in neuroscience is understanding how behaviorally relevant information is collected by a sensory organ and subsequently encoded by the brain . By actively moving their whiskers , rats can discriminate fine differences in textures . Little is known , however , about how whisker dynamics reflect texture properties or how the nervous system encodes this information . In one hypothesis , whisker motion over a texture produces a unique , texture-specific temporal profile of velocity , which is encoded in the temporal pattern of neural activity . In a second , alternative hypothesis , textures excite a specific subset of whiskers due to intrinsic , whisker-specific mechanical resonance frequencies . Information is then encoded by the spatial distribution of neural activity in whisker-related columns in cortex . Here , we assess these hypotheses by measuring whisker motion as animals whisk across sandpapers of varying roughness . We found that whiskers resonate in air and on surfaces , but that these resonance vibrations do not vary with , and therefore do not encode , texture . Instead , whisker motion over a textured surface produces fast , transient slip-stick events whose dynamics are dependent on texture roughness . Texture is likely to be encoded in the temporal pattern of spikes evoked by these slip-stick events .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biophysics",
"neuroscience"
] |
2008
|
Texture Coding in the Rat Whisker System: Slip-Stick Versus Differential Resonance
|
Failure to properly establish the left–right ( L/R ) axis is a major cause of congenital heart defects in humans , but how L/R patterning of the embryo leads to asymmetric cardiac morphogenesis is still unclear . We find that asymmetric Nodal signaling on the left and Bmp signaling act in parallel to establish zebrafish cardiac laterality by modulating cell migration velocities across the L/R axis . Moreover , we demonstrate that Nodal plays the crucial role in generating asymmetry in the heart and that Bmp signaling via Bmp4 is dispensable in the presence of asymmetric Nodal signaling . In addition , we identify a previously unappreciated role for the Nodal-transcription factor FoxH1 in mediating cell responsiveness to Bmp , further linking the control of these two pathways in the heart . The interplay between these TGFβ pathways is complex , with Nodal signaling potentially acting to limit the response to Bmp pathway activation and the dosage of Bmp signals being critical to limit migration rates . These findings have implications for understanding the complex genetic interactions that lead to congenital heart disease in humans .
Establishment of asymmetries along the left–right ( L/R ) axis is critically important for proper placement , morphogenesis and functioning of vertebrate organs [1] , [2] . In the heart , defects in early L/R patterning events are implicated in three of the six most common forms of congenital heart disease ( CHD ) : transposition of the great arteries , chamber septation defects and chamber isomerisms [3] , [4] , [5] . Among the genetic lesions known to associate with cardiac defects are mutations in a number of proteins within the Nodal signaling pathway [6]; the asymmetric activation of which plays a conserved role in specifying the L/R axis in all vertebrates [1] , [2] . Human mutations in Bmp pathway genes have also been implicated in the development of CHD [7] . The zebrafish heart develops from cardiac precursors derived from the lateral plate mesoderm ( LPM ) which migrate to the midline and fuse to form the cardiac cone [8] . Active cell migration within the cardiac cone converts this symmetric , disc-shaped structure into the asymmetric , linear heart tube , with these cellular movements being regulated by Nodal and Bmp signals [9] , [10] , [11] , [12] . However , significant controversies exist concerning the relative requirements for these two pathways . While the laterality of Nodal signaling has been shown to influence the direction of cardiac cell migration [9] , the Bmp pathway has been implicated as providing the dominant laterality cue to the heart [12] . Therefore , a complete understanding of asymmetric cardiac morphogenesis requires clarification of the specific requirements for and interactions between these two TGFβ signaling pathways . Here , we report the identification of separate and parallel functions for the Nodal and Bmp pathways in establishing consistent cardiac asymmetry . We find that the Nodal signaling pathway provides the dominant laterality cue to the heart and directs cardiac cell migration by increasing cell velocities on the left of the wild type ( WT ) cardiac cone . By contrast , we find the Bmp pathway negatively regulates cardiac cell migration rates . The resulting asymmetries in cell velocities across the L/R axis result in the rotation of the cardiac cone to produce the asymmetrically jogged heart . Our findings are particularly important in clarifying the requirements for Bmp signaling in establishing cardiac laterality . Our results are consistent with a role for BMP signaling in limiting cell migration on the right side of the cardiac cone , while the Bmp pathway has historically been thought to positively regulate cell migration and act predominantly on the left of the developing heart [12] , [13] , [14] . We also demonstrate that Bmp4 is dispensable for establishment of cardiac asymmetry in the presence of asymmetric Nodal signaling , suggesting that Nodal plays the more critical role in this process . However , in the absence of Nodal signaling , the heart is extremely sensitive to the dosage of Bmp4 signals , a finding with significant implications for the existence of combinatorial mutations in multiple pathways giving rise to CHD in humans . Finally , we have identified the existence of a novel integration between these two TGFβ pathways . Through genetic analysis , we find that the “Nodal” transcription factor FoxH1 is required for both Nodal-dependent and independent functions within the heart and that , separate from its requirement to transduce Nodal signals , FoxH1 is also necessary to mediate Bmp responsiveness within cardiac cells . Taken together , this work greatly enhances our understanding of the specific requirements for Nodal versus Bmp signaling in establishing cardiac asymmetry and provides new insight into the cross-regulation and integration between these two pathways necessary for the consistent development of proper cardiac laterality .
We have previously shown that the sidedness of Nodal signaling directs the first asymmetry evident in the heart; a left , anterior-directed movement of atrial cells within the zebrafish “cardiac cone” [9] , also see Figure 1A ) . To gain insight into how Nodal signaling establishes the laterality of cell trajectories , we analyzed cell behaviors in time lapse images of cardiac development during conversion of the cardiac cone into the linear heart tube , and find that the left-biased asymmetry of cell movements in WT embryos is established by differences in cell migration velocities along the L/R axis . Cells on the left of the cardiac cone , which in WT embryos are exposed to the zebrafish Nodal southpaw ( spaw ) , migrate with an average rate of 9 . 2 nm/s ( Figure 1A and 1H ) . Cells on the right side of the cone that do not exhibit Nodal target gene expression in WT embryos migrate with significantly slower rates of 5 . 9 nm/s ( p<10−3; Figure 1H; Video S1 ) . Thus , exposure to Nodal signaling appears to induce increases in cell velocities within the cardiac cone . This L/R asymmetry in migration rates consistently leads to a left direction of cardiac jog by 24 hours post-fertilization ( hpf ) in WT embryos ( 3/3 embryos; Figure 2A and 2D ) . To confirm that Nodal signaling is responsible for the increase in migration rates in cells on the left of the WT cardiac cone , we analyzed no tail ( ntl ) morpholino-injected embryos ( morphants ) , which express spaw bilaterally in the lateral plate mesoderm [15] . We find that ntl morphants exhibit a statistically significant increase in cell velocities compared with WT , with the left and right cells migrating an average of 10 . 9 nm/s and 10 . 2 nm/s , respectively ( p<10−3; Figure 1B and 1H; Video S2 ) . This bilateral increase in migration rates leads to a loss of asymmetry in cell trajectories and subsequent loss of asymmetry in heart position at 24 hpf ( 3/3 embryos analyzed ) ( Figure 1B; and data not shown ) . We have further confirmed that this velocity increase is Nodal-dependent by analyzing migration rates in spaw morphants and embryos treated with a Nodal-inhibitor drug , SB-505124 , that blocks activity of the Nodal Type I receptors [16] , [17] . In both conditions , migration rates are significantly reduced compared to WT ( p<10−3 for each ) , but with no statistical difference in average cell velocities between spaw morphant and drug-treated embryos ( p = 0 . 713 ) ( Figure 1C , 1D and 1H; Video S3 ) . These migration rates are also consistent with what has been reported for late zygotic one-eyed pinhead mutants ( LZoep ) , which lack the essential Nodal co-receptor [10] , [18] . Importantly , the loss of biased asymmetry in migration rates in spaw morphants , LZoep mutants , and drug-treated embryos also results in predominant loss of biased asymmetry in cardiac jogging ( see below ) ( Figure 2E , 2F and 2G ) , indicating that differences in cell velocity along the L/R axis of the cardiac cone are required for consistent establishment of asymmetry in cardiac jog . While asymmetric spaw is required to establish differences in cardiac velocities along the L/R axis , the heart can respond to additional laterality cues in the absence of Nodal signaling . Loss of either Spaw or the essential Nodal co-receptor Oep leads to randomized jogging ( left , right or midline ) , not midline jogging , suggesting residual , randomized L/R signals function in the absence of Nodal pathway activation ( Figure 2A–2C , 2E and 2F ) [9] . Evidence from the literature suggests that the Bmp pathway may provide this additional signal , as ubiquitous over-expression of Bmp ligands or global inhibition of the Bmp pathway both lead to alterations in cardiac laterality [12] , [13] , [14] , [19] . If Bmp signals do provide the remaining asymmetric information in the absence of Nodal signaling , we hypothesized that combined inhibition of both the Nodal and Bmp pathways would remove all asymmetric information to the heart and lead to predominantly midline jogging . Bmp4 has been implicated as the Bmp ligand required during cardiac laterality determination [13] , [14] , [19] , so we utilized the bmp4Y180 null mutant [20] to analyze the jogging phenotypes of embryos with diminished Nodal and Bmp pathway activities . Consistent with our hypothesis , loss of both Spaw and Bmp4 produces predominantly midline hearts by 24 hpf , with 89% of embryos lacking L/R asymmetry in cardiac jog ( midline jog; Figure 2H ) . Previous work has suggested that Bmp signals provide the dominant laterality cue to the heart and that Nodal signaling plays a secondary role , required only to ensure a consistent bias in Bmp pathway activity [12] , [14] . However , in contrast to this existing view , we find that maternal-zygotic ( MZ ) MZbmp4Y180 mutants on their own do not exhibit significant jogging defects ( Figure 2I ) . This result suggests that Bmp4 is only required to provide asymmetric cues in the absence of Nodal and is otherwise dispensable for cardiac laterality . Additionally , these data argue that Spaw , not Bmp4 , provides the dominant laterality cue to the developing heart . The differences in our results from those previously published are possibly because those studies utilized overexpression of Bmp2b , producing non-physiological levels of Bmp signaling via a ligand that is not expressed in the cardiac field at this stage of development . Interestingly , we find that in the absence of Spaw , cardiac cells are highly sensitive to even small changes in the dosage of bmp4 , further supporting the idea that the levels of Bmp signaling can influence this process . While embryos containing only one copy of the bmp4Y180 mutation do not exhibit jogging defects ( data not shown ) , when Nodal signaling is inhibited in these embryos , 76% display midline hearts at 24 hpf ( Figure 2J ) . Taken together , our results provide strong evidence that Nodal signals dominantly influence cardiac laterality . Importantly , we find Bmp4 to only be critical in the absence of Spaw to direct jogging , at which point cardiac asymmetry is highly sensitive to the overall level of Bmp signaling present in the embryo . Our data suggests that the hearts that jog directionally in Spaw morphants ( left and right; Figure 2E ) are responding to asymmetric information provided by Bmp4 . We predict this should result in asymmetries in cell migration velocities which produce the resulting directional jog . However , despite their randomized jogging phenotype , asymmetries in cell velocities across the L/R axis were not detected , and both the global population of spaw morphants and individual morphant embryos display bilaterally reduced cell velocities . Due to the profoundly diminished migration rates in spaw morphants , there is a subsequent delay in the conversion of the cardiac cone into the linear heart tube . Thus , it is likely that our time lapse movies were simply not long enough to observe an establishment of L/R asymmetry in cell velocities in the subset of spaw morphants with directional cardiac jog ( 3/7 embryos ) . In addition , even when directional jog is established in these morphants , the heart tubes are displaced from the midline much less significantly than are left-jogged hearts in WT embryos . Therefore , even when asymmetries are established in cell velocities along the L/R axis , we would anticipate those differences to be substantially less than the asymmetry observed in WT cardiac migration rates and , therefore , potentially below the threshold for significance used in our statistical analysis . To determine how Bmp signaling influences cell migration during jogging , we analyzed cardiac cell migration in bmp4Y180−/+ embryos injected with spaw MO . We find that combined loss of Nodal signaling and just one functional copy of bmp4 leads to significantly increased average migration rates ( 6 . 08 nm/s on left and 5 . 82 nm/s on right ) compared with loss of Nodal signaling alone ( 4 . 8 nm/s left and right; p<10−3; Figure 1E and 1H; Video S4 ) . This increase in cardiac migration rate is even more pronounced in embryos injected with bmp4 and spaw MOs , with left- and right-sided cells displaying average velocities of 7 . 8 nm/s and 7 . 4 nm/s , respectively ( Figure 1F and 1H; Video S5 ) . As with all other conditions in which significant L/R asymmetries in migration rate are lost , nearly all embryos with combined inhibition of bmp4 and spaw also exhibit loss of directional cardiac jog ( midline jog −3/5 bmp4Y180−/+ embryos injected with spaw MO; 4/5 bmp4/spaw double morphants ) . The increase in cardiac migration velocities upon inhibition of the Bmp pathway suggests that Bmp signaling is normally required to limit the migratory ability of cardiac cells . The negative influence of Bmp signaling on cell migration rate , along with the significantly slower velocities of right-sided cardiac cells in WT embryos , suggests that the Bmp pathway normally acts on the right of the cone to influence cardiac laterality . However , previous work has established that bmp4 is expressed with a left bias at 20 hpf and that the laterality of this expression is altered in embryos with defects in jogging asymmetry [13] . Given the inconsistencies between earlier reports and our cell migration data , we were interested in determining the specific localization and potential asymmetry of Bmp activity within the heart . To this end , we analyzed Bmp pathway activation in WT embryos by immunofluorescence for the activated Bmp intracellular effectors Smads 1 , 5 and 8 ( phospho-Smad1/5/8 or p-Smad1/5/8 ) . Consistent with previous reports [12] , we find that Bmp pathway activity is asymmetrically increased on the left of the cardiac cone at 20 hpf , as indicated both by increased p-Smad1/5/8 fluorescence intensity in cells on the left compared with right ( p<10−3 ) and by greater numbers of p-Smad1/5/8 positive cells present on the left of the cardiac cone ( p = 0 . 003; Figure 3A–3C , 3M–3O ) . At first glance , this left-biased increase in pSmad1/5/8 in cells with faster velocities appears contradictory to data from our time lapse experiments which strongly support a role for the Bmp pathway in negatively regulating myocardial migration rates . Results from our time-lapse analyses , coupled with our genetic data demonstrating BMP signaling is dispensable for generating asymmetry in the heart in the presence of asymmetric Nodal signaling , suggest that regardless of the increase in p-Smad1/5/8 on the left , the cue from Nodal for cells to increase migration rates is stronger than the influence of Bmp signals on these same cells . By contrast , as cells on the right of the cone do not receive inductive cues from Spaw , Bmp activation on the right significantly diminishes migration rates . In WT embryos , this leads to cell velocities on the right of the cone being reduced compared to those on the left . Thus , when Nodal signaling is absent ( as in spaw morphants and SB-505124-treated embryos ) , all cells in the cone respond to repressive cues from the Bmp pathway and both left and right myocardial cell velocities are substantially reduced . Likewise , when the Nodal pathway is activated on both sides of the cone ( as in ntl morphants ) cell velocities are increased to higher rates than those observed in WT cells exposed to Nodal , presumably because bilateral Nodal signaling diminishes the repressive effects of Bmp on myocardial migration rates . While we observe an increase in p-Smad1/5/8 on the left side of the cardiac cone , we argue that Nodal signaling increases migration on the left and overrides repressive cues from Bmp . Consistent with this hypothesis , analysis of Bmp activity in the hearts of spaw morphants reveals a significant increase in the fluorescence intensity of p-Smad1/5/8 immunostaining in both right ( p<10−3 ) and left ( p = 0 . 002 ) cardiac fields compared with WT ( Figure 3D–3F , 3M–3O ) . These results indicate that Nodal signaling limits the level of Bmp pathway activity , potentially by competing for the common intracellular effector Smad4 , a mechanism of Nodal/Bmp antagonism that is known to occur during earlier stages of L/R patterning in zebrafish and other species [20] , [21] . Interestingly , despite having higher intensities of p-Smad1/5/8 fluorescence , we find that the average number of p-Smad1/5/8 positive cells is significantly diminished in spaw morphants compared to WT ( p = 0 . 02 ) ( Figure 3O ) , which may indicate a role for Nodal signaling in both positive and negative regulation of Bmp pathway activation within the cardiac cone . These results suggest that Nodal signaling ensures the establishment of differential migration rates along the cardiac L/R axis in two ways; first , by directly increasing cell velocities on the left and second , by limiting the level of Bmp activity on the left . Ultimately , robust development of jogging asymmetry appears to require left-restricted activation of the Nodal pathway to increase migration rates and response to the Bmp pathway on the right of the cone , where Bmps serve to diminish migration velocities . In our immunofluorescence experiments , we noticed that the GFP staining in myocardial cells did not significantly colocalize with the p-Smad1/5/8 present in the heart field ( Figure 3P–3P′″ ) . As endocardial cells are also localized to the midline by this stage of development , we hypothesized that Bmps signal more predominantly to the endocardium at 20 hpf . To address this possibility , we performed p-Smad1/5/8 staining in embryos with GFP expressed from the kdrl promoter , which labels endothelial and endocardial cells [22] . We observed significant colocalization of GFP and p-Smad1/5/8 in these embryos , indicating that Bmp activity is primarily upregulated within endocardial cells ( Figure 3Q–3Q′″ ) . By contrast , all direct Nodal targets that have been identified in the heart are expressed within the myocardial population [9] , [12] , [23] , [24] , [25] . Thus , the Nodal and Bmp pathways appear to act in parallel to establish asymmetries in cell migration velocities within the cardiac cone: Nodal pathway activation in the myocardium on the left leads to increases in migration rates while Bmp signaling in the endocardium limits myocardial migration , primarily on the right . Precedence for cross-regulation between the endocardium and myocardium during the earlier migration events leading to cone formation have been described [26] , [27] . Thus , interplay between these two tissues is critical for at least two migrations during cardiac development . While loss of the ligand Spaw or the co-receptor Oep results in randomized jogging ( Figure 2E and 2F ) , embryos with a nonsense mutation in the Nodal transcription factor FoxH1 [28] display 78% midline hearts ( Figure 2K ) . These results suggest that FoxH1 performs both Nodal-dependent and independent functions within the heart . Interestingly , midway jogging defects are strikingly similar to those of embryos lacking both Nodal and Bmp signaling ( Figure 2H , 2J and 2K ) suggesting that FoxH1 may be required for cardiac cell responsiveness to both TGFβ pathways . To address this possibility , we analyzed cardiac cell migration in midway/foxH1 mutants . Cells on the left and right of the cardiac cone in midway mutants migrate with average velocities of 7 . 8 nm/s and 7 . 1 nm/s , respectively ( Figure 1G and 1H; Video S6 ) . These migration rates are significantly faster than those of cardiac cells in embryos lacking Spaw ( p<10−3 ) or treated with the Nodal-inhibitor drug SB-505124 ( p<10−3 ) , confirming a Nodal-independent function for FoxH1 in establishing cardiac laterality ( Figure 1C , 1D and 1H ) . Importantly , midway cardiac velocities are not statistically different than those of bmp4/spaw double morphants ( p = 0 . 580 ) , suggesting that FoxH1 is necessary for cardiac cells to respond to both Nodal and Bmp signals . Consistent with this idea , we observe a significant , bilateral decrease in p-Smad1/5/8 in the hearts of midway mutants compared with WT , both in fluorescence intensity ( p<10−3 ) and in the number of p-Smad1/5/8 positive cells ( p<10−3 ) ( Figure 3G–3I , 3M–3O ) . Interestingly , midway mutants exhibit a more profound decrease in Bmp pathway activation than that observed in bmp4/spaw double morphants in both fluorescence intensity and number of p-Smad1/5/8 positive cells ( p<10−3 ) , indicating that loss of FoxH1 activity profoundly blocks cardiac cell responsiveness to Bmp signals ( Figure 3J–3L , 3M–3O ) . Here , we report the parallel requirements for Nodal and Bmp pathways in establishing cardiac laterality through opposing influences on cardiac cell migration rates ( Figure 4 ) . Additionally , we have uncovered a novel , Nodal-independent role for FoxH1 during this process in mediating cardiac cell responsiveness to Bmp signals ( Figure 4 ) . The Nodal-dependent activity of FoxH1 in regulating cardiac cell migration is likely restricted to myocardial cells , as the Nodal targets lefty1 , lefty2 and has2 all display myocardial expression [9] , [12] , [23] , [24] , [25] , the latter of which has been implicated in previous work [12] to influence myocardial migration . However , it is unclear at present whether the Nodal-independent function of FoxH1 is carried out within the myocardial or endocardial cells of the developing heart . FoxH1 can act as both transcriptional activator and repressor , and an inhibitory requirement for FoxH1 has been reported in the vasculature to limit expression of the VegF receptor , kdrl [29] . Consequently , FoxH1 may be required directly in the endocardium to block expression of Bmp antagonists or activate transcription of required components of the Bmp pathway . Alternatively , the influence of FoxH1 on Bmp activation may be more indirect . Recent work has reported that retinoic acid ( RA ) signaling is necessary for consistent asymmetry in bmp4 expression in the zebrafish cardiac cone , with inhibition of RA leading to bilateral bmp4 expression and an increase in midline jogging [30] . Interestingly , a genome-wide screen for FoxH1 binding sites revealed aldh1a1 , a gene necessary for RA production , as a direct target of FoxH1 in mouse [31] . While zebrafish lack an ortholog of this specific RA processing enzyme , there are a number of other aldh1a family members expressed throughout zebrafish development [32] , making it compelling to speculate that the loss of Bmp responsiveness in midway cardiac cells may be due to defects in RA signaling . Our results support a model in which cardiac laterality is regulated by interactions and cross-regulations both between TGFβ pathways and between the myocardial and endocardial layers of the developing heart that regulate differential motility along the L/R axis . These interactions involve complex integrations between Nodal and Bmp pathways , and we demonstrate that cardiac cells are highly sensitive to the dosage of these TGFβ signals . Bilateral exposure to Spaw increases migration rates beyond what is observed in left cells of the WT cone , and loss of a single copy of bmp4 in addition to Nodal signaling significantly alters both jogging laterality and cardiac cell velocities . Moreover , the signals that can influence laterality in the heart likely involve additional members of the TGFβ family . We note that inhibition of Nodal signaling with the SB-505124 drug decreases cell velocities as expected . However , jogging laterality in these embryos is predominantly midline , which differs from loss of Spaw or Oep ( Figure 2E–2G ) . While this phenotype resembles that of embryos lacking Spaw and Bmp4 ( Figure 2H and 2J ) , the cell migration rates in drug-treated embryos are consistent with loss of Nodal , but not Bmp signaling ( Figure 1H ) and , indeed , we find that the Bmp pathway is still activated within the heart field upon SB-505124 treatment ( data not shown ) . The spaw morpholino completely abolishes expression of spaw in the LPM , strongly suggesting that Spaw is absent in the hearts of these embryos . This , coupled with the similar phenotypes of spaw knockdown and LZoep mutants , suggests that the effect of the drug is not a result of more complete knockdown of Nodal signaling . SB-505124 acts intracellularly on the Alk 4/5/7 Type I receptors , which are utilized by both Nodal and TGFβ ligands . Overall , this suggests that another TGFβ molecule signaling through the Nodal receptors can affect the migration of cardiac cells and may be important for allowing the cardiac cells to respond to fluctuations in Bmp levels when Spaw is absent . Taken together , these results have implications for determining the underlying genetic lesions in CHD , as they suggest that heterozygous mutations in components of different TGFβ signaling pathways may synergize to produce severe phenotypes . Further analysis of integrations of signals within and between cardiac cells will provide insight into the general mechanisms driving asymmetric morphogenesis and will greatly enhance our understanding of the potentially complicated genetic interactions underlying the development of CHD in humans .
All WT and GFP transgenic fish pairs used to generate embryos for this analysis have been determined to exhibit low backgrounds of L/R phenotypes . The WT strains used in these experiments include Tu , Alb/+ , WIK , and AB . The following additional strains were used in this report: Tg ( myl7:egfp ) [33] , bmp4Y180 mutants [20] , Tg ( kdrl:egfp ) [22] , LZoeptz257 mutants [18] . MZbmp4Y180 mutants were generated as previously described [20] . Genotyping to identify embryos carrying the bmp4Y180 and midway mutations was conducted as previously described [20] , [28] . Morpholinos were diluted in phenol red and 500 pL were injected into zebrafish embryos between the 1 and 4 cell stages . Both the antisense start-site spaw morpholino ( 5-GCACGCTATGACCGGCTGCATTGCG-3 ) [34] and the antisense start-site ntl morpholino ( 5-GACTTGAGGCAGGCATATTTCCGAT-3 ) [35] were injected at a 1 ng/500 pL concentration and the antisense splice-site bmp4 morpholino ( 5-GGTGTTTGATTGTCTGACCTTCATG-3 ) [14] was injected at a 2 ng/500 pL concentration . The SB-505124 Nodal inhibitor drug [16] was reconstituted in DMSO and stored at 4°C at a concentration of 10 mmol . A 40 µM dilution of SB-505124 in blue water was added to embryos at tail bud stage . Embryos were fixed in 4% PFA at 4°C overnight followed by gradual rehydration into PBDT ( 1×PBS , 0 . 1% Tween 20 , 1% DMSO ) . After removal of the tails , embryos were incubated in 10 µg/mL proteinase K for 10 minutes . Proteinase K activity was stopped with a 20 minute incubation in 4% PFA . Embryos were then blocked for 2 hours in PBDT containing 10% normal goat serum ( NGS ) and incubated overnight at 4°C in a 1∶100 dilution of monoclonal GFP antibody ( Roche #11814460001 ) and 1∶100 dilution of p-Smad1/5/8 antibody ( Cell Signaling Technology #9511 ) in PBDT . The following day , embryos were washed in PBDT and incubated at 4°C overnight in 1∶100 dilutions of CY3 donkey anti-rabbit and CY2 donkey anti-mouse . ( Southern Biotech #1090-02 ) . After washing , the embryos were imaged on a Leica SP5 spectral confocal microscope . Time-lapse imaging was performed as previous described [9] . Briefly , embryos were screened for GFP expression at the 18–19 somite stage and mounted in 2% low-melt agarose on the converslip bottom of a round dish with the cardiac field positioned directly adjacent to the coverslip . The agarose was covered with a solution of water and tricaine to immobilize the embryos throughout the course of the time lapse . Embryos were then imaged an average of 4 . 5 hours on an inverted Leica SP5 spectral confocal microscope using a heated stage set to 29°C . Multiple embryos ( up to 6 ) were imaged during a single time lapse using the mark and find feature of the Leica software . Once collected , time lapse series were transferred to the Volocity software for analysis where each series was merged to a single plane at each time point . Velocity measurements reported were determined by Volocity software ( Perkin-Elmer , USA ) . Atrial cells were tracked for analysis as these are the cells that we and others report to exhibit the most pronounced asymmetric migrations that drive the process of cardiac jogging [9] , [10] , [11] , [12] . In all embryos analyzed , cells were defined as being “left’ or “right” due to their position along the L/R axis of the cardiac cone . In embryos that lack asymmetry in cell migration within the cone ( all genotypes analyzed other than WT ) , left and right cells migrate along the lateral edges of the cone . However , cells at the posterior of the cone migrated directly towards the anterior with straight trajectories and not along the lateral edges . Given the different migration phenotype of these cells compared to cells from the left or right , we designated them as “center” in our analysis . The “center” cells in these embryos are those that ultimately involute during formation of the linear heart tube . Thus , to keep the “center” designation consistent between WT and other genotypes , we label the cells that normally involute in WT [11] as “center” cells in these embryos . Cells labeled as “left” in wildtype are within the lefty2 expression domain [9] indicating that these cells are exposed to Spaw signals while cells labeled as “right” in wildtype are those lacking expression of Spaw downstream targets and are thus not exposed to Nodal signaling . All images of p-Smad1/5/8 immunofluorescence were taken on an SP5 spectral confocal microscope and all images were taken at the same laser intensity and gain settings . To ensure that no significant variations in fluorescence intensity were detected due to artifact , a minimum of three separate immunostaining experiments were performed and analyzed for embryos of a single genotype , with these embryos being imaged on separate days . A minimum of three embryos were analyzed for p-Smad1/5/8 fluorescence for each genotype . We normalized all images to the right cells in the wildtype controls and each of the three different trials in wildtype embryos produced equivalent results . Image analysis was performed using the IMARIS software ( Bitplane Software , USA ) . The green surface of the myocardium was generated using the surface tool , and the region of interest for identification of p-Smad1/5/8 positive cells was determined to be the GFP positive portion of the image . The p-Smad1/5/8 positive nuclei were selected using the surface tool , allowing spot separation . The field of each image was then split into left and right sides . Student T-tests were used to make all statistical comparisons . In order to test the null hypothesis that the triplet distributions of left , center , and right cell velocities were identical for data collected from two different genotypes , data were normalized as follows . For each of the three spatial designations , the mean and standard deviation were calculated for all data pooled across both genotypes . The mean was subtracted from the pooled data , and the resulting values were divided by the standard deviation . Normalized data were then re-pooled across all three designations separately for both genotypes , and the comparison between genotypes was performed . Comparisons between left cell velocities of one genotype and left cell velocities of another were made directly , without normalization . In each analysis , p values less than 0 . 05 were considered to be statistically significant .
|
Defects in left–right ( L/R ) patterning can lead to severe defects in the formation of the heart . In fact , three of the most common forms of congenital heart disease , transposition of the great arteries , chamber septation defects , and chamber isomerisms , can be caused by earlier defects in L/R asymmetry . The Nodal and Bmp signaling pathways influence the development of cardiac asymmetry , but how these signals function in this process is not well understood . In this report , we have clarified the specific roles for the Nodal versus Bmp pathways in the heart . We find that Nodal signals increase the rate of cardiac cell migration , while Bmp signals decrease cardiac cell velocities . We demonstrate that asymmetric Nodal signaling plays a critical role in directing asymmetry in the heart in contrast to reports suggesting that signaling via Bmp4 is the more critical pathway . In fact , we find that Bmp4 signaling is dispensable for correct asymmetry in the heart in the presence of asymmetric Nodal signals . In addition , we have identified a novel integration between these two pathways at the level of the transcription factor FoxH1 , which is required for cardiac cell responsiveness to both Nodal and Bmp signals . Taken together , this work significantly increases our understanding of how the signals regulating cardiac asymmetry function and integrate to consistently establish cardiac laterality . These results also suggest that human congenital heart defects that have not been found to result from single mutations within individual genes may develop due to combinations of mutations within components of these two separate pathways .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"animal",
"models",
"developmental",
"biology",
"zebrafish",
"model",
"organisms",
"organism",
"development",
"signaling",
"molecular",
"development",
"birth",
"defects",
"cell",
"migration",
"heart",
"development",
"biology",
"organogenesis",
"morphogenesis"
] |
2013
|
Integration of Nodal and BMP Signals in the Heart Requires FoxH1 to Create Left–Right Differences in Cell Migration Rates That Direct Cardiac Asymmetry
|
Influenza A viruses are respiratory pathogens that cause seasonal epidemics with up to 500 , 000 deaths each year . Yet there are currently only two classes of antivirals licensed for treatment and drug-resistant strains are on the rise . A major challenge for the discovery of new anti-influenza agents is the identification of drug targets that efficiently interfere with viral replication . To support this step , we developed a multiscale model of influenza A virus infection which comprises both the intracellular level where the virus synthesizes its proteins , replicates its genome , and assembles new virions and the extracellular level where it spreads to new host cells . This integrated modeling approach recapitulates a wide range of experimental data across both scales including the time course of all three viral RNA species inside an infected cell and the infection dynamics in a cell population . It also allowed us to systematically study how interfering with specific steps of the viral life cycle affects virus production . We find that inhibitors of viral transcription , replication , protein synthesis , nuclear export , and assembly/release are most effective in decreasing virus titers whereas targeting virus entry primarily delays infection . In addition , our results suggest that for some antivirals therapy success strongly depends on the lifespan of infected cells and , thus , on the dynamics of virus-induced apoptosis or the host's immune response . Hence , the proposed model provides a systems-level understanding of influenza A virus infection and therapy as well as an ideal platform to include further levels of complexity toward a comprehensive description of infectious diseases .
Influenza A viruses continue to pose a serious threat to public health causing three to five million cases of severe illness and up to 500 , 000 deaths during the annual epidemics [1] . In addition , novel influenza strains that acquire the potential to infect and transmit efficiently between humans can create pandemics like the 1918 Spanish Flu that killed millions worldwide [2] . Currently , there are only two classes of direct-acting antivirals ( DAAs ) licensed for influenza treatment: fusion inhibitors ( adamantanes ) , which impair virus entry , and neuraminidase blockers ( oseltamivir and zanamivir ) interfering with the release of progeny virus particles [3] . However , resistances against these drugs occur frequently [4] , urging the need for new antiviral agents [6] . In recent years , the discovery of new antiviral targets for influenza treatment has received much attention . In particular , compounds which interfere with host factors promise to be effective antivirals as cellular factors are less susceptible to mutation impairing viral escape strategies . Such compounds can , for example , inhibit virus entry by removing cell surface receptors as was shown for recombinant sialidases , or block viral RNA transcription through PolII inhibition ( for a detailed review of cellular targets and their inhibitors see reference [6] ) . The inhibition of essential cellular signaling cascades like Raf/MEK/ERK signaling , NF-κB signaling , the PI3K/Akt pathway , or the PKC signaling cascade is another promising strategy ( reviewed in [7] ) . Finally , viral proteins themselves are targets for antiviral agents with new inhibitors of the viral neuraminidase , M2 ion-channel , and polymerase on the horizon ( reviewed in [8] ) . With the advent of these DAAs influenza therapy has moved beyond symptomatic treatment toward specifically targeting key steps of viral replication . The development of new and more potent drugs thus requires a deeper understanding of the viral life cycle [6] . In general , the growth of influenza viruses within a host involves at least two distinct scales: ( i ) the intracellular level of infection where the virus synthesizes its proteins , replicates its genome , and assembles new virions and ( ii ) the extracellular level at which it infects new target cells and spreads throughout the tissue . As DAAs can target both scales , understanding how these levels interact and where to interfere to efficiently counteract an infection is vital to the design of new antivirals . In the past , mathematical modeling has provided valuable insights into the kinetics of influenza A virus infection under drug treatment ( [9]–[13] , reviewed in [14] , [15] ) . However , the majority of studies focused exclusively on the extracellular level of infection either neglecting or drastically simplifying intracellular events . While such simplifications allow for the identification of critical infection parameters from sparse data , they can influence model predictions leading to an overly optimistic assessment of the treatment efficiency required to suppress the infection [16] . Other theoretical works examined how drugs affect specific replication steps of different viruses inside an infected cell [17]–[19] . Yet , these approaches only consider a single round of infection and do not account for the spread of the virus to new cells . Recently , Guedj and colleagues showed that combining both levels in a model of hepatitis C virus infection significantly improved its capability to explain clinical observations [20] , [21] . However , as the authors only included viral genome copies at the intracellular level their approach is limited to the analysis of drugs that target genome synthesis , degradation or packaging . Nevertheless , such studies strongly suggest that integrating the intracellular life cycle of a virus into a model for cell-to-cell transmission would facilitate a systematic exploration of new drug targets . The resulting multiscale model can also yield a more realistic description of virus infection [22] , [23] and more accurate estimates of key infection parameters [16] , [20] , [24] . Recently , we developed a model of the complete intracellular life cycle of influenza A virus comprising key steps from virus entry to progeny virion release [25] . Here , we link this description to the transmission of virus between host cells . We first show that this integrated modeling approach successfully captures data on the intracellular level of all three viral RNA species as well as on the extracellular infection dynamics represented by virus titers and the amount of infected cells . We then use the model to investigate potential antiviral targets including the steps of virus entry , nuclear trafficking , viral RNA and protein synthesis , and assembly/release . We provide a ranking of these targets and show that the lifespan of infected cells can be of particular importance for therapy success . Finally , detailed information on the construction of the model is provided in the Materials and Methods section at the end of this manuscript .
Our description of the extracellular level of infection is based on the classical model of viral kinetics within a host or cell population , which accounts for uninfected cells , infected cells , and free virions ( reviewed in [14] and [15] ) . We augmented this framework by explicit consideration of the number of apoptotic cells and by modeling virus entry in more detail ( Figure 1A ) . Once inside a cell , the virus starts producing viral RNA and proteins . To track these intracellular processes our multiscale model accounts for the age of an infected cell , i . e . , the time that has elapsed since its infection ( Figure 1B ) . The amount of each viral component inside an infected cell over its infection age is simulated using a model of the influenza A virus life cycle [25] . This submodel includes the following essential features of viral replication ( Figure 1C ) : the production of viral mRNA and complementary RNA ( cRNA ) from viral ribonucleoproteins ( vRNPs ) , which contain the negative-strand viral genomic RNA ( vRNA ) ; the synthesis of viral proteins; the encapsidation of newly produced cRNA and vRNA into cRNPs and vRNPs , respectively , by the viral RNA-dependent RNA polymerase ( RdRp ) and the nucleoprotein ( NP ) ; the nuclear export of vRNPs regulated by the viral matrix protein 1 ( M1 ) and the nuclear export protein ( NEP ) ; and the assembly and release of progeny virions ( for further details see reference [25] ) . Integration of both levels is achieved by assigning the age-dependent state of a cell to the age-segregated cell population ( Figure 1D ) . In the model , intracellular replication primarily affects the extracellular level via the virus release rate , which depends on the abundance of viral proteins and RNA inside a cell and determines the amount of virions released into the extracellular space . The extracellular level in turn controls the number of infected cells and their lifespan . To ensure an accurate calibration of both levels of the model , we followed a two-way strategy . First , we conducted experiments at a high multiplicity of infection ( MOI ) , i . e . , a high initial number of virions per cell , which results in a single synchronous infection round . This allowed us to measure the intracellular levels of the three viral RNA species together with the number of released virus particles and view them as the response of an average infected cell ( Figure 2A ) . We then performed flow cytometry of low MOI experiments to assess the dynamics of multicycle infections where the virus spreads throughout a cell population in successive waves ( Figure 2B ) . The model was fit simultaneously to both data sets such that the intracellular part , i . e . , the replication inside an average infected cell agrees with the synchronous infection experiments , while its combination with the extracellular model captures the multicycle scenario . Hence , each infected cell behaves according to the time courses shown in Figure 2A and a population of such cells yields the dynamics in Figure 2B when infection occurs at low MOI . Simulation results on both levels are in good agreement with the data showing , for instance , a rapid increase in viral mRNAs upon infection ( Figure 2A ) . In contrast , cRNA and vRNA synthesis does not start until 3–4 h post infection ( hpi ) as the accumulation of viral proteins is required for genome replication [26] . Between 3 and 4 . 5 hpi our model underestimates the cRNA level . However , it does capture the amount of mRNA and vRNA . Since all three viral RNA species are tightly related the model is relatively constrained . Thus , some deviations are to be expected as the model has to balance these time courses as well as the data on the extracellular level . In the late phase of infection progeny vRNPs , which provide the template for cRNA and mRNA , leave the nucleus to be incorporated into new virus particles . This causes a shutdown of RNA synthesis around 5–6 hpi . At the same time , the first progeny virions leave the cell . Hence , the eclipse phase , i . e . , the delay between infection and virus release , is approximately 6 h ( Table 1 ) . After this delay virus production increases as more viral components accumulate before it starts declining toward the end of the productive infection phase when proteins and later genome copies become limiting ( Figure 1D ) . These intracellular dynamics fit well with the progression of infection on the extracellular level ( Figure 2B ) considering that typical errors of adherent cell numbers are in a range of 10–20% due to variations introduced by the measurement technique , handling and trypsinization . Most of the cells become infected between 12 and 19 hpi in the second and third wave of infection . Virus-induced apoptosis then causes a decline in cell numbers within the next two days . At 32 hpi , we observe a large discrepancy between the model and the data . However , this single time point can be regarded as an outlier since the measured total cell concentration increased by 30% between 24 and 32 hpi ( data not shown ) . It is highly unlikely that such an increase occurs this late in infection . The good agreement of the model with the infectious virus titer provides further evidence for this . The titer shows an initial drop due to the attachment of seed virus to cells before it increases reaching its maximum around 30 hpi . Next , we checked the predictive capabilities of the model by comparing it to measurements at different infection conditions that were not used for model construction . These simulations successfully capture the shift of infection dynamics in the presence of higher and lower amounts of virus particles in the inoculum ( Figure 3A ) . In addition , the virus titer prediction for a low seed virus concentration is in good agreement with experiments whereas for higher MOI virus production is overestimated ( Figure 3B ) . We conclude that the model is in good agreement with the intracellular and extracellular dynamics of influenza A virus infection and can be of predictive value especially for low MOI regimes where multiple infection rounds occur , which resembles the in vivo situation more closely than single round experiments that use high MOI . A major advantage of the proposed multiscale model is that it integrates the time course of intracellular virus replication with cell death dynamics . This allows us to assess whether the lifespan of an infected cell constrains virus production . From the measurements in Figure 2B , we obtain the average lifespan of an infected cell as 25 h ( Table 1 ) . Approximately at the same time virus release would stop due to the depletion of viral components ( Figure 1D ) . Nevertheless , many cells may die before the end of this productive phase dependent on how much individual survival times vary around the mean . Models of viral infection usually assume that the probability of cell death is independent of time , i . e . , that survival times follow an exponential distribution ( see references [24] for more details and alternatives ) . Using this assumption , we find that most cells indeed die within 25 h with more than one quarter succumbing to apoptosis before reaching the peak in virus release ( Figure 4 ) . Hence , cell death can affect the number of virus particles an average cell produces . Models of viral infection can be used to simulate the efficiency of antiviral treatment . While previous studies have mostly considered general effects of drugs on cell infection or virus production [9] , [11] , [14] , our multiscale approach can also predict how drug interference affects the intracellular viral life cycle . Figure 5A shows simulation results that illustrate the impact of DAAs on the amount of virus particles an infected cell releases . In our model , inhibitors of viral mRNA synthesis , protein translation , and virus assembly/release are highly effective in reducing virus production even at low drug efficacy . Note that antivirals usually show a maximum efficacy above 90% [9] , [10] , [27] . At these levels , inhibitors of mRNA splicing , cRNA/vRNA synthesis , and nuclear export are also very successful . Intriguingly , inhibition of the two steps of RNA replication can , however , lead to an increase in virus release for low-efficacy drugs . A similar result can be observed when targeting M1 binding and the encapsidation of viral RNAs by NP . Figure 5B shows the simulated time courses of different viral components in response to selected low-efficacy drugs . As expected , the inhibition of viral transcription leads to lower mRNA levels , which impairs protein synthesis and virus release ( Figure 5B upper panel ) . We also observe a minor increase in cRNAs and vRNAs during the early phase of infection due to lower M1 protein levels . Based on experimental evidence [28]–[30] , M1 acts as a negative regulator of cRNA and mRNA synthesis in our model [25] . Together with NEP , it binds to vRNPs controlling their nuclear export . Once outside the nucleus , vRNPs can no longer serve as templates for the two positive-strand RNAs . In addition , M1 proteins fulfill a second role during virus assembly where they form the inner hull of virus particles as their most abundant viral component [31] . Hence , inhibition of particle assembly/release also results in higher M1 levels , a stronger negative regulation of RNA synthesis , and lower RNA levels besides reducing virus release ( Figure 5B middle panel ) . This type of regulation also causes the increase in virus titers seen upon weak inhibition of cRNA synthesis ( Figure 5B lower panel ) . The reduction in RNA levels in the early phase of infection leads to a lower abundance of M1 proteins . The resulting lack of inhibition allows a faster synthesis of RNAs during later stages and consequently a higher rate of virus release . Since the release of virions further drains the pool of M1 proteins , our model predicts a sustained production of virus particles for these drug efficacies . Figure 5B highlights an interesting aspect of viral replication . There may be regimes where a higher overall number of viruses can be produced at the expense of an early virus release . Yet , such an advantage would clearly depend on the lifespan of an infected cell and on whether cell death by virus-induced apoptosis or the immune response shortens it ( Figure 6 ) . While the inhibition of cRNA synthesis may lead to higher titers in our system , drug treatment has hardly any influence on virus production when apoptosis occurs at twofold of its estimated rate ( Figure 6C 1×–2× ) . For an even shorter lifespan of infected cells the antiviral may be deemed effective reducing particle production to half its pre-treatment level ( Figure 6C 4× ) . Hence , the effect of a drug on viral replication has to be judged with respect to cell death dynamics to correctly evaluate treatment potential . Apart from reducing particle production , antivirals can also delay the spread of the virus providing time for the immune system to counteract infection . Figure 7A illustrates the evolution of virus titers in a susceptible host cell population under simulated drug treatment . Again , inhibitors of viral RNA and protein synthesis almost completely suppress viral replication . Therefore , these drugs protect most of the host cells from infection ( Figure 7B ) . Nevertheless , a few cells become infected and produce virus causing the titer to only slowly decline . In contrast , drugs targeting virus entry , i . e . , fusion , endocytosis and binding to the cell surface , are less successful in decreasing peak titers but delay infection by up to 50 h . However , they do not prevent the cell population from becoming infected . Note that , in this scenario , peak titers are primarily constrained by the number of available cells . When cells are depleted virus production ceases and titers decrease with the rate of viral clearance .
Viral infections cover several scales , a form of complexity that computational modeling is poised to address [32] . In this study , we developed a model that integrates the main stages of influenza A virus infection within a host: the intracellular replication of the virus and its extracellular transmission to new host cells . This multiscale approach accurately captures a variety of in vitro measurements , provides insights into virus growth across different scales , and aides the development of DAAs . The limited quantity and diversity of experimental data still represents a significant bottleneck for models of extracellular viral kinetics and their validation [15] . A promising approach to close this gap is to incorporate detailed information on the intracellular viral life cycle . Our model of virus replication was previously validated against a variety of experimental studies including data on virus binding , fusion , RNA synthesis , and regulation by viral proteins [25] . In combination with the quantitative RNA levels presented here , it provides a detailed picture of intracellular events and their impact on virus production . For instance , it yields a delay of 6 h between infection and virus release and suggests that virus production increases for another 7 h as viral components accumulate . While the length of the eclipse phase is in good agreement with estimates from other modeling studies ( 7 h for MDCK cells in bioreactors [33] , 0 . 22–6 h for cultivations in a hollow-fiber system [11] and 6 h for human infection [9] ) , previous models have assumed that virus production proceeds at a constant rate in the productive phase . Since simulations are quite sensitive to such assumptions [16] , [24] , multiscale modeling can lead to more realistic estimates of key infection parameters [20] , which were shown to greatly support the design of antiviral treatment [34] . We estimate the lifespan of an infected cell as 25 h ( including the eclipse phase ) , which is in the range of other studies ( reviewed in [15] ) . Virus production would cease around the same time due to the depletion of viral components in the cell . Nevertheless , most of the cells in our simulations die before the end of the productive phase as survival times vary significantly around the mean . Following the majority of models for viral infection , we assumed an exponential distribution of the survival time . Yet , a recent study suggested that other distributions may be more appropriate to capture viral kinetics [24] . In principle , our model is ideal to accommodate such assumptions since the apoptosis rate can be chosen freely as a function of the infected cell age . However , our simulations ( Figure 2 ) and data on apoptosis induction during single-cycle infections [35] do not justify more complex approaches . Although in good agreement with the data , our simulations underestimate the intracellular level of cRNA between 3 and 4 . 5 hpi . Yet , both the mRNA and vRNA level are captured nicely . This might be due to a differential regulation of viral transcription and replication unaccounted for in the model so far . Such control could , for instance , be mediated by NP ( reviewed in [36] ) , NEP [37] or small viral RNAs [38] . However , the contribution of these mechanisms is still a matter of controversial debate and quantitative data is lacking preventing us from incorporating this type of regulation in the model . We , thus , chose to keep our mathematical framework as simple and constrained as possible . Nevertheless , the model simultaneously captures a rich pool of data indicating that it incorporates the key steps of in vitro influenza virus infection . When testing the model against data for different infection conditions that were not used for construction , we also noticed an overestimation of virus production for high MOI . In our model , peak virus titers primarily depend on the initial cell concentration , which was comparable in the three experiments and the number of virions each cell produces . The cell-specific virus yield follows from the intracellular replication dynamics and the lifespan of an infected cell ( i . e . the yield equals the integral under the dash-dotted virus production curve in Figure 4 ) . Hence , the observed decrease in virus production at high MOI can be explained in two ways . On the one hand , factors present in the inoculum may impair intracellular virus replication reducing the rate of virus production . Defective interfering particles could be such factors [39] . On the other hand , the inoculum can contain substances such as interferons that may reduce the lifespan of an infected cell by increasing apoptosis induction [35] . Experimental work is in progress to discriminate between these two hypotheses . For the construction and calibration of our model , we mostly relied on cell culture experiments due to the limited diversity of available in vivo data . Currently , virus titers are the type of data most frequently used for in vivo models ( reviewed in [14] ) as they are easily attainable from infected individuals and animals . In principle , four parameters are sufficient to describe such titer curves constraining the level of detail one can incorporate into a mathematical model [24] . In contrast , in vitro systems provide access to a variety of information like the number of available cells , their infection status or the intracellular level of viral RNAs . This wealth of data was a prerequisite for the development of our multiscale model . However , now that the model has been established future studies may want to implement modifications to closer resemble the in vivo situation . For instance , the growth and death of uninfected cells is usually neglected in acute infection models as target cell dynamics are assumed to be slow compared to infection [14] . Furthermore , virus loss in the lung is caused by active processes such as phagocytosis and mucociliary clearance as opposed to degradation in cell culture experiments and may , hence , be faster . However , the most prominent feature of in vivo infections our model is currently lacking is the immune system . Although a number of models were proposed that incorporate an immune response none of them agreed completely with the variety of experimental data available [40] . Implementing an adequate description of the immune response , thus , remains one of the major challenges in viral kinetic modeling today [14] , [40]–[42] . Modeling intracellular replication in detail allowed us to simulate the effect of DAAs on the amount of virions an average infected cell produces . Given a drug efficacy above 90% inhibitors of viral transcription , replication , protein synthesis , nuclear export , and assembly/release proved to be most successful in mitigating replication . Indeed , antivirals targeting virus release in the form of neuraminidase inhibition are widely used in influenza treatment today . To exploit this target in the future new compounds are , however , required as the emergence of drug-resistant strains is on the rise [8] . Inhibitors of the viral polymerase are a promising alternative . During viral replication polymerases engage in an autocatalytic reaction where they synthesize cRNA from vRNA and vice versa . In addition , they transcribe the viral genome into mRNAs for new polymerases . Interrupting this positive feedback has detrimental consequences for all major viral components in our model . In agreement with this , compounds which specifically inhibit viral transcription efficiently impair influenza A virus replication in cell culture and mice [43] . Similarly , favipiravir ( T-705 ) , an inhibitor of influenza virus RNA polymerase activity [44] , is potent against influenza viruses in vitro and in vivo [45] , [46] and has entered clinical trials recently [47] demonstrating the potential of viral RdRps as drug targets . Our model could be used to test different dosing regimes and support such clinical trials . In other studies , interference with the assembly of viral polymerase complexes by 25-amino-acid peptides [48] or small molecule inhibitors [49] has been shown to inhibit viral replication . In contrast , RdRp formation has hardly any influence on virus production in our model unless the drug efficacy well exceeds 99% . This discrepancy most likely originates from the kinetics of polymerase assembly in our model . Due to the lack of quantitative data , we assumed that polymerases form from their three subunits according to mass actions kinetics [25] . Rather than the formation itself subunit availability represented the kinetic bottleneck for RdRp assembly in simulations . In light of the above mentioned experimental studies , future models may need to revise this assumption if polymerase assembly is at the focus of investigation . Reconciling model predictions that are initially inconsistent with data provides an ideal opportunity to also refine our understanding of the underlying biology but it requires experiments specifically designed to resolve the discrepancy . Instead of reducing peak virus titers , our model predicts that inhibitors of virus entry mainly delay in vitro infection , which is in agreement with previous studies [11] . This is because they only decrease the infection rate of cells instead of impairing the processes responsible for viral component production resulting in similar cell-specific virus yields . The treatment success of such inhibitors may , hence , depend on mechanisms , which take advantage of the delay and clear infection . Intriguingly , some of our simulations yield regimes where treatment can also lead to an increase in virus production at the expense of early virus release . From an evolutionary perspective this regime might not be beneficial as faster growing strains would out-compete such variants . However , during treatment it may , nevertheless , occur . We show that the lifespan of an infected cell determines whether a slower but more efficient virus production leads to higher titers . An antiviral treatment that was rejected based on the survival times of infected cells in cell culture may thus even be successful when lifespans are shorter . In vivo , the latter is indeed very likely as the immune response increases cell death rates [42] , [50] . Also , virus strain-dependent factors such as the expression of the PB1-F2 protein can lead to faster cell death [51] . Screening approaches for antiviral compounds may , hence , benefit from using conditions that mimic the cell survival times observed in vivo . In summary , we have developed a multiscale model of in vitro influenza A virus infection which integrates the intracellular level of viral replication and the extracellular level of cell-to-cell transmission . We are optimistic that such models will contribute to the development of antiviral drugs , support clinical trials and provide a platform for the establishment of more detailed infection models in the future . To achieve this goal , next-generation models will need to incorporate the immune response , pharmacokinetics and comprehensive information on virus-host interactions . Multiscale modeling provides an ideal framework for such an endeavor as diverse cellular processes can be simulated individually and incorporated as separate modules into a unifying framework .
We used an age-segregated infection model for adherent cells , which follows from the general population balance [23] , to describe the dynamics of uninfected target cells ( ) , infected cells ( ) , and their apoptotic counterparts and , respectively ( 1 ) ( 2 ) ( 3 ) ( 4 ) withwhere uninfected cells grow with specific rate or undergo apoptosis with rate . Growth can occur with a maximum specific rate to a maximum concentration of cells assuming that all non-apoptotic cells occupy a finite surface area . The infection rate is denoted and will be discussed at the end of this section . In Equation ( 2 ) , infected cells have the age and undergo virus-induced apoptosis with an age-dependent rate . Since infection creates cells with age zero , we obtained the boundary condition . Apoptotic target cells in Equation ( 3 ) can either become infected or undergo cell lysis with rate . The same lysis rate is used for apoptotic infected cells . Assuming that there are no infected cells in the beginning ( ) , we can rewrite Equation ( 2 ) in terms of an algebraic equation ( 5 ) where can be interpreted as the infection age density such that gives the number of infected cells with age between and . Equation ( 5 ) illustrates that cells which have age at time were infected at time . The integral term accounts for cell loss due to apoptosis . Using Equation ( 5 ) instead of Equation ( 2 ) , thus , allows us to track the infection front precisely . The equation for infectious virus particles ( ) in the extracellular space follows as ( 6 ) withandwhere denotes the age-dependent virus production rate . We assumed that virions are degraded or cleared with rate . The binding of virus particles to target cells was modeled as described before [25] . In brief , we considered two types of binding sites ( ) : high-affinity ( ) and low-affinity ( ) sites . The virus attaches to or dissociates from these sites with rates and , respectively , whereby the latter rate follows from the equilibrium constant . The concentration of free binding sites was calculated from their total number per cell ( ) , the concentration of target cells , and the concentration of attached virus particles ( ) . In this notation each virion occupies one binding site . Note that we did not consider binding to infected cells as neuraminidase expression on the cell surface limits superinfection [52] . In order to account for drug effects on virus entry , we defined equations for the concentration of attached virions ( ) on the surface of target cells ( considering both and ) as well as for virions in the endosomes of these cells ( ) ( 7 ) ( 8 ) where and denote the endocytosis and fusion rate , respectively . The first two terms in Equation ( 7 ) account for virus binding and dissociation as well as for endocytosis . The last term quantifies the loss of virions with cells that leave the compartment of interest , i . e . , with cells leaving the population of target cells by infection or cell lysis with rate and , respectively . Equation ( 8 ) accounts for the endocytosis of virions attached to both types of binding sites , the fusion of virions with the endosomal membrane , and again the loss of particles due to infection and lysis of target cells . Since we consider a cell ‘infected’ as soon as viral genome copies enter its cytoplasm , the infection rate follows from the fusion rate in Equation ( 8 ) ( 9 ) withwhere corresponds to the number of cells which become productively infected upon the fusion of one virion . This number cannot exceed one but may become lower if several virions are required to cause productive infection . While the first part of Equation ( 9 ) represents the number of cells that become infected per hour , the fraction serves two purposes: substituted in Equations ( 7 ) and ( 8 ) it provides the number of viruses per target cell and in Equations ( 1 ) and ( 3 ) it yields the fraction of non-apoptotic and apoptotic target cells , respectively , to total target cells . Similarly , the lysis rate of apoptotic target cells can be derived as ( 10 ) The intracellular level of infection was essentially modeled as described before [25] . In brief , a set of ordinary differential equations was used to simulate virus entry , viral RNA and protein synthesis , and virus assembly . In contrast to the original description , we modified the equation of the virus release rate ( 11 ) withwhere release depends on the abundance of progeny vRNPs in the cytoplasm ( ) and structural viral proteins ( ) with denoting the number of virus particles for which components must be present in order to reach half the maximum release rate . In its new form , can only increase to a maximum rate of assuming that there is only a limited number of host factors available for virus budding . This change was implemented to avoid unrealistically high virus production rates that occurred in some treatment regimes . For simulations in Figure 1D and Figure 2A , the complete intracellular model was used as described above . However , when coupling the model to the extracellular level , we neglected virus entry and initialized the model with a complete set of eight vRNPs in the cytoplasm . Attachment , endocytosis , and fusion were considered at the extracellular level instead ( Equations ( 7 ) and ( 8 ) ) . In order to ease the computational burden and allow for a more intuitive interpretation of simulation results , we assumed that the extracellular level has little or no influence on intracellular events , i . e . , that each infected cell behaves the same independent of the time of infection and the extracellular environment . As shown by Haseltine and colleagues , this assumption permits the selective decoupling of both levels and reduces the model's complexity significantly [23] . Hence , we could first simulate intracellular virus replication to calculate the virus release rate as a function of the infection age . This rate was then used in Equation ( 6 ) to simulate the extracellular level . The intracellular submodel was solved numerically with the CVODE routine from SUNDIALS [53] on a Linux-based system . Model files and experiments were handled with the Systems Biology Toolbox 2 [54] for MATLAB ( R2010b The MathWorks Inc . ) . We then used Euler's method with a step size of to solve the extracellular model ( Equations ( 1 ) and ( 3 ) – ( 8 ) ) . The integrals in Equations ( 1 ) , ( 4 ) and ( 6 ) were approximated in each step by substituting Equation ( 5 ) for and using the rectangle rule with a step size of . To further reduce computational costs , the integral in Equation ( 5 ) was evaluated prior to simulation following the same approach . The method was checked for numerical accuracy against simulations using smaller step sizes and by comparison to a discrete version of Equation ( 2 ) with a large number of age classes . Table S1 lists the initial conditions of all presented simulations . Parameters of the intracellular model can be found in Table S2 and Table S3 shows parameter values for the extracellular model . Model parameters were estimated by fitting the complete intracellular submodel ( including the equations for virus entry ) to experimental virus titers per cell and the levels of vRNA , cRNA and mRNA measured during high MOI infection ( Figure 2A ) . Simultaneously , the reduced model ( excluding virus entry ) was coupled to the extracellular equations using the same parameters and the complete multiscale model was fit to the time courses of uninfected and infected cells , their apoptotic counterparts , and the virus titer during low MOI infection ( Figure 2B ) . Estimation was performed using the fSSm algorithm for stochastic global optimization [55] . In particular , the algorithm was used to simultaneously minimize the least squares prediction error of all measured state variables , whereby the error of each variable was normalized by its respective maximum measurement value ( e . g . the deviation between measured and simulated vRNA level was weighted by the maximum of the measured vRNA level ) . The summed errors of the intracellular and extracellular part of the model were then divided by the number of measurements , respectively , and added to attain an overall measure of fit quality . Since experiments indicated that real-time RT-qPCR detects free viral RNAs from the seed virus supernatant , which may adhere to cells but cannot enter them , we applied the first measurement value as an offset to all simulation values of viral RNAs . Bootstrap confidence intervals [56] were determined considering the standard deviations in Figure 2A as well as a 20% error for cell counts and 0 . 3 log for virus titers in Figure 2B . In order to simulate drug treatment with efficacy , parameters in the model which correspond to the drug's target ( Table S4 ) were perturbed by . Treatment was assumed to occur at constant efficacy starting from 0 hpi . For results in Figure 5A , the reduced intracellular model was simulated first to determine the virus release rate . The total amount of virus particles produced by an average infected cell over its lifetime ( ) was then calculated by considering cell death according to For single round infections ( Figure 2A ) , adherent MDCK cells ( ECACC No . 84121903 ) were grown in GMEM ( GIBCO ) supplemented with 10% fetal calf serum ( FCS ) ( PAN Biotech ) and 1% peptone ( Lab M ) using T175 flasks and incubated at 37°C under a 5% CO2 atmosphere to maintain pH 7 . 2 . One day before infection , cells were washed twice with phosphate buffered saline ( PBS ) , detached and counted with a Vi-CELL XR ( Beckman Coulter ) . Subsequently , 1 . 75×106 cells were seeded into 35 mm dishes . Infection was performed using influenza A/Puerto Rico/8/34 ( Robert Koch Institute , #3138 ) with a seed virus preparation containing 1 . 23×108 infectious virus particles per mL . Prior to infection , cells were washed twice with PBS and virus was added at a multiplicity of infection ( MOI ) of 6 in 250 µL serum-free virus maintenance medium ( GMEM , GIBCO ) containing 1% peptone ( Lab M ) and 5 units/mL trypsin ( GIBCO ) . Dishes were incubated for 30 minutes at 37°C and 5% CO2 atmosphere before cells were washed once with PBS and 1 mL virus maintenance medium was added . To correctly account for the loss of viral components due to virus release , the total amount of virus particles leaving an average infected cell was determined using the hemagglutination assay as described previously by Kalbfuss et al . [57] . Titer measurements in log10 HA units per test volume ( log HAU/100 µL ) can be converted into hemagglutinating particles per mL by ( 12 ) assuming that at least one virus particle per erythrocyte ( 2×107 cells/mL ) is required to cause agglutination [58] . For detailed information on the multicycle experiment ( Figure 2B ) , the reader is referred to reference [35] from which the measurements were adopted . In brief , adherent MDCK cells were cultivated to confluence in T25-flasks and washed with PBS prior to infection followed by addition of serum-free virus maintenance medium ( GMEM , GIBCO ) containing 1% peptone ( Lab M ) and 5 units/mL trypsin ( GIBCO ) . Subsequently , influenza A/Puerto Rico/8/34 was added at an MOI of 10−4 , 0 . 1 and 3 . For each time point one T-flask was harvested and adherent cells were trypsinized and pooled with the cells from the supernatant . Aliquots of 106 cells were fixated with 1% paraformaldehyde ( Sigma-Aldrich ) and 70% ethanol ( Carl Roth ) and stored at −20°C . Double staining for infection status and apoptosis was performed using a fluorescein isothiocyanate ( FITC ) -labeled anti-NP mAb ( AbD Serotect ) and a terminal deoxynucleotidyl transferase dUTP nick end labeling ( TUNEL ) assay kit ( Roche Diagnostics ) , respectively . Measurements were collected using an Epics XL flow cytometer ( Beckman Coulter ) . In addition to flow cytometry , the infectious virus titer was measured from the supernatant of T-Flasks using a TCID50 assay as described before by Genzel and Reichl [59] . To extract viral RNAs , cells were washed once with PBS , lysed and scraped from the dish . Lysates were stored at −80°C . RNA was extracted using “INSTANT Virus RNA” ( Analytik Jena ) according to the manufacturer's instructions and stored at −80°C . For the real-time RT-qPCR assay , priming strategies for the differentiation of viral RNA species were adapted from Kawakami et al . [60] . In brief , polarity specific and tagged primers ( Table 2 ) were used in reverse transcription as follows . 1 µL of RNA extract was mixed with 1 µL primer ( 1 µM for cRNA and vRNA; 10 µM for mRNA ) , 1 µL dNTPs ( 10 mM each ) and filled up to 14 . 5 µL with nuclease-free water . The mixture was incubated at 65°C for 5 min and subsequently cooled to 40°C for mRNA and 55°C for cRNA and vRNA . Afterward , the reaction mixture ( 4 µL 5×Reaction Buffer , 0 . 5 µL Maxima H-Minus Reverse Transcriptase ( 200 U/µL ) ( Thermo Scientific ) and 1 µL nuclease-free water ) was added . After incubation at 60°C for 30 min the reaction was terminated at 85°C for 5 min . Additionally , a 10-fold dilution series of the corresponding RNA reference standards ( 5⋅10−7 to 5 ng ) each containing 350 ng cellular total RNA was reverse transcribed . Subsequently , the RT reaction was diluted to a final volume of 100 µL . Concentration of viral RNA was determined in molecules per cell using “Rotor-Gene SYBR Green PCR Kit” ( Qiagen ) and Rotor-Gene Q real-time PCR cycler ( Qiagen ) . 4 µL of the diluted cDNA were mixed with 1 µL primer set and 5 µL reaction mixture . The cycle conditions of the real-time PCR were 95°C for 5 min followed by 40 cycles of 95°C for 10 sec and 60°C for 20 sec . Finally , a melting curve from 65°C to 90°C was performed . The concentration of viral RNA was calculated based on the RNA reference standards with linear regression ( Ct-value against log10 of number of molecules ) . The number of viral RNA molecules ( ) was calculated based on the length of the fragment ( ( bp ) ) , where ( ng ) is the mass of the template , denotes the average mass of one base , and ( mol−1 ) corresponds to the Avogadro constant . The number of RNA molecules was then related to the number of cells ( ( cells ) ) to calculate the abundance of viral RNAs per cell ( ( molecules/cell ) ) . The final result was calculated byusing the coefficient for dilution of RT reaction ( ) and the volume of RNA eluate ( ( µL ) ) .
|
Influenza A viruses are contagious pathogens that cause an infection of the respiratory tract in humans , commonly referred to as flu . Each year seasonal epidemics occur with three to five million cases of severe illness and occasionally new strains can create pandemics like the 1918 Spanish Flu with a high mortality among infected individuals . Currently , there are only two classes of antivirals licensed for influenza treatment . Moreover , these compounds start to lose their effectiveness as drug-resistant strains emerge frequently . Here , we use a computational model of infection to reveal the steps of virus replication that are most susceptible to interference by drugs . Our analysis suggests that the enzyme which replicates the viral genetic code , and the processes involved in virus assembly and release are promising targets for new antivirals . We also highlight that some drugs can change the dynamics of virus replication toward a later but more sustained production . Thus , we demonstrate that modeling studies can be a tremendous asset to the development of antiviral drugs and treatment strategies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2013
|
Multiscale Modeling of Influenza A Virus Infection Supports the Development of Direct-Acting Antivirals
|
The primary constituents of plaques ( Aβ42/Aβ40 ) and neurofibrillary tangles ( tau and phosphorylated forms of tau [ptau] ) are the current leading diagnostic and prognostic cerebrospinal fluid ( CSF ) biomarkers for AD . In this study , we performed deep sequencing of APP , PSEN1 , PSEN2 , GRN , APOE and MAPT genes in individuals with extreme CSF Aβ42 , tau , or ptau levels . One known pathogenic mutation ( PSEN1 p . A426P ) , four high-risk variants for AD ( APOE p . L46P , MAPT p . A152T , PSEN2 p . R62H and p . R71W ) and nine novel variants were identified . Surprisingly , a coding variant in PSEN1 , p . E318G ( rs17125721-G ) exhibited a significant association with high CSF tau ( p = 9 . 2×10−4 ) and ptau ( p = 1 . 8×10−3 ) levels . The association of the p . E318G variant with Aβ deposition was observed in APOE-ε4 allele carriers . Furthermore , we found that in a large case-control series ( n = 5 , 161 ) individuals who are APOE-ε4 carriers and carry the p . E318G variant are at a risk of developing AD ( OR = 10 . 7 , 95% CI = 4 . 7–24 . 6 ) that is similar to APOE-ε4 homozygous ( OR = 9 . 9 , 95% CI = 7 . 2 . 9–13 . 6 ) , and double the risk for APOE-ε4 carriers that do not carry p . E318G ( OR = 3 . 9 , 95% CI = 3 . 4–4 . 4 ) . The p . E318G variant is present in 5 . 3% ( n = 30 ) of the families from a large clinical series of LOAD families ( n = 565 ) and exhibited a higher frequency in familial LOAD ( MAF = 2 . 5% ) than in sporadic LOAD ( MAF = 1 . 6% ) ( p = 0 . 02 ) . Additionally , we found that in the presence of at least one APOE-ε4 allele , p . E318G is associated with more Aβ plaques and faster cognitive decline . We demonstrate that the effect of PSEN1 , p . E318G on AD susceptibility is largely dependent on an interaction with APOE-ε4 and mediated by an increased burden of Aβ deposition .
Dementias are complex , polygenic and genetically heterogeneous disorders [1] . The most common form of dementia is Alzheimer's disease ( AD ) , which affects more than 5 . 3 million people in the US [2] . Late-onset AD ( LOAD ) is the most common form of dementia . However , the current model of AD pathogenesis is based on the genetic findings in rare and phenotypically extreme AD cases [3] . LOAD heritability varies from 58% to 79% [4] and , despite the tremendous progress in AD genetics in the last twenty years , the total proportion of phenotypic variance explained by all the combined variants ( including APOE genotype and genome wide association studies [GWAS] signals ) is estimated to be 23% [5] , which suggests a large proportion of the heritability of AD still remains unexplained . Three important factors may account for the missing heritability in AD; first , the clinical heterogeneity of AD remains a significant confounding variable in case-control studies [6] , second , much of the unexplained variance of complex phenotypes may be attributed to low frequency or rare alleles [7] and third , gene by gene or gene by environment interactions [8] . Quantitative intermediate phenotypes have helped to overcome some of these obstacles in complex diseases [9] , [10] . Endophenotype-oriented approaches have greater statistical power , less clinical heterogeneity and offer important insights into the mechanisms by which genetic variants modulate the disease phenotype [6] , [9] , . The primary constituents of plaques ( Aβ42/Aβ40 ) and neurofibrillary tangles ( tau and phosphorylated forms of tau [ptau] ) are the current leading diagnostic and prognostic cerebrospinal fluid ( CSF ) biomarkers for AD [12] . Recently , it was shown that CSF biomarker abnormalities typically precede clinical AD symptoms by decades and reflect the timing and magnitude of pathophysiological changes [13] . These findings suggest that a better understanding of the genetic contribution to the variance in these CSF biomarkers can provide important information about susceptibility to AD . In fact , the two most important known risk factors for AD , APOE genotype and age account for 13% and 14% of the variance in CSF Aβ42 and tau levels , respectively [14] . Likewise , pathogenic mutations in the most important causal genes for familial AD , amyloid-beta precursor protein ( APP ) , and presenilin 1 and 2 ( PSEN1 , PSEN2 ) alter CSF Aβ42 levels [13] , [15] , [16] . Additionally , some genetic variants initially discovered by their association with CSF biomarkers have recently been proven to be modifiers of risk , age at onset ( AAO ) or rate of AD progression [17] , [18] , [19] . Likewise , it was recently described that carriers of PSEN1 mutations exhibit very low CSF Aβ42 , and high tau or ptau levels [13] , [20] , [21] , [22] . Similar CSF biomarker level profiles have been described in sporadic AD cases [23] . However , the genetic variants responsible for CSF changes in sporadic AD have not been found yet . Together , these results suggest that CSF biomarker levels as quantitative traits are useful tools in uncovering genetic variants that are closely related to the physiopathological mechanisms underlying AD . Rare or low frequency coding and non-coding variants have been predicted to be enriched in functional alleles and to exhibit strong effect size [7] , [10] . Recently , a rare ( minor allele frequency [MAF] = 0 . 02 ) coding variant in TREM2 gene p . P47H was found to confer a high risk for AD ( Odd ratios from 2 to 5 ) [24] , [25] , [26] . Two recent studies analyzed the association of genetic variants of APP , PSEN1 , PSEN2 , MAPT , and GRN on risk for AD [27] , [28] . One study was focused on common variants in sporadic AD [27] while the other focused on the identification of very rare coding variants in familial LOAD [28] . However , the impact of low-frequency coding variants of APP , PSEN1 , PSEN2 , GRN and MAPT on sporadic LOAD has not been well studied . Identification of low frequency variants associated with disease remains challenging because standard case-controls design requires very large sample sizes . To overcome this problem we have used quantitative phenotypes . Previously , we identified a pathogenic mutation in a family with LOAD within the PSEN1 gene by selecting the top and bottom 5% from the distributions of Aβ40 , Aβ42 , and Aβ42/40 ratio [29] In the present study , we sequenced individuals with extremes levels of CSF-based biomarkers in order to identify variants in APOE , APP , PSEN1 , PSEN2 , GRN and MAPT genes associated with the CSF biomarker levels . This approach allowed us to identify known pathogenic variants , AD risk factors and identify a low frequency variant that increases risk for AD in a gene-gene interaction mode .
We hypothesized that the coding variants found in individuals at the extremes of the phenotypic distribution of CSF biomarker levels are more likely to have a functional impact on CSF biomarker levels . In order to identify rare or low frequency variants that affect the CSF levels of Aβ42 , tau and ptau levels , we used a two-stage extreme phenotype sequencing design ( Figure S1 ) . A 10-fold difference between the lowest and highest raw values in Aβ42 , tau and ptau CSF levels in each series was found among individuals in these studies . The individuals were selected regardless of their clinical status ( based on the clinical dementia rating [CDR] ) ( Table 1 ) . We combined both series ( WU-ADRC [n = 475] and ADNI [n = 259] ) by normalizing the CSF Aβ42 , tau and ptau levels and adjusting for covariates [17] , [18] . We selected 212 individuals from the top and bottom 15% for each phenotype ( Table 1 ) . The 212 samples were divided in two pools ( Pool 1 and 2 , respectively ) ; targeted and pooled-sample sequencing was performed . All the validated variants were genotyped in the total CSF sample and tested for association with each CSF biomarker . Linear regression ( assuming an additive genetic effect ) was utilized for each variant by adjusting for significant covariates ( age , gender , CDR and site [WU-ADRC or ADNI] ) ( Table S1 in Text S1 ) [17] , [18] . A greater than 30-fold coverage per allele at all positions within the 62 amplicons designed to cover the protein coding regions of the APP , APOE , PSEN1 , PSEN2 , MAPT and GRN were obtained ( Table S2 in Text S1 ) . After adjusting for the sensitivity and specificity parameters of the base-calling algorithm ( SPLINTER ) using negative and positive controls , a total of 396 and 369 variants were called and perfectly annotated in the targeted genomic regions of Pool 1 and 2 , respectively . 73% of these variants were intronic , 8% were missense , 5% were coding-synonymous , 1% were at splicing sites , 12% were located at the untranslated regions ( UTR ) and 2% were called to be near-gene ( Table S2b in Text S1 ) We focused on missense and splicing-affecting variants with a predicted minor allele frequency ( MAF ) below 5% ( by SPLINTER ) in each pool . A total of 27 rare or low frequency non-synonymous variants were validated by direct genotyping in the discovery samples ( both pools ) . 33% of these variants identified ( 9/27 ) were novel . Seven of nine ( 77% ) are located in highly conserved nucleotide ( GERP>4 ) and 88% ( 8/9 ) are predicted to be damaging for the respective protein ( SIFT and polyphen2 algorithms ) [30] . As expected , 48% of these variants are singletons ( 9/27 ) or doubletons ( 4/27 ) ( Table 2 ) . Among the 18 previously reported variants; we found one known pathogenic mutation PSEN1 p . A426P . PSEN1 p . A426P ( rs63751223 ) was reported in a five members of a family with autosomal dominant AD [31] . We also found four high-risk variants for LOAD ( APOE , p . L46P; MAPT , p . A152T; PSEN2 , p . R62H and p . R71W ) [28] , [32] , [33] , six variants that were previously reported in families with AD or frontotemporal dementia ( FTD ) , but classified as non-pathogenic ( GRN , p . R433W , p . P458L , p . R19W; MAPT , p . Q230R; PSEN1 , p . R35Q and p . E318G ) [34] , and seven variants that have been recently reported in public databases with no clear role in human disease to date ( APOE , p . E37K; GRN , p . C231W; MAPT , p . G107S , p . S318L , p . V224G; PSEN2 , p . E317G and p . V300G ) ( A detailed description of each variant can be found in the supporting material in Text S1 ) . These results highlight the relative enrichment of rare and low frequency variants in six genes involved in AD and FTD among individuals at the extremes of the CSF biomarker distribution [29] . Next , we tested whether any of the variants identified by an endophenotype-based approach could improve our understanding of both the genetic architecture and pathophysiology of LOAD [17] , [18] . We ran a linear regression analysis for single SNP using CSF biomarkers as quantitative traits , but we failed to find significant association with CSF tau , ptau or Aβ42 levels for most of the identified variants , even after we collapsed all of the potentially damaging variants in each gene and analyzed the dataset for carriers vs . non-carriers of these variants ( Table 3 ) . Surprisingly , a low frequency coding variant in PSEN1 , p . E318G ( rs17125721 ) ( MAF = 0 . 02 for Europeans Americans , Exome Variant Server EVS: http://evs . gs . washington . edu/EVS/ ) , whose pathogenic role is currently debated [34] exhibited a statistically significant association ( multiple test correction threshold , p = 7 . 0×10−3 ) with CSF tau ( p = 9 . 2×10−4 , Beta = 0 . 14 ) and ptau levels ( P = 1 . 8×10−3 , Beta = 0 . 12 ) , but not with Aβ42 ( p = 0 . 14 , Beta = −0 . 05 ) . Interestingly , it has been reported that the combination of Aβ42 and tau or ptau as a ratio provides the best discriminative value to date for AD cases [35] , [36] and predict the conversion from non-dementia clinical status to dementia [37] . p . E318G exhibited a significant association with the ratio of ptau∶Aβ42 ( p = 9 . 5×10−5 , Beta = 0 . 08 ) and tau∶Aβ42 ( p = 2 . 0×10−4 , Beta = 0 . 06 ) ( Figure 1A–C , 2A ) suggesting that the association of p . E318G with CSF biomarker levels may be an association with clinical AD . In order to confirm this association with CSF biomarkers and to determine whether this or any other SNP in linkage disequilibrium ( LD ) was driving the association , we combined genotype and imputed data from 895 individuals ( WU-ADRC , n = 501 , and ADNI , n = 394 , this dataset constitute the same CSF series that we genotyped ( Table 1 ) plus additional 161 individuals ) to perform a dense fine mapping analysis of PSEN1 genomic region . The number of independent tests ( Meff = 317 ) was calculated based on the number of SNPs after correcting for LD structure ( r2 = 0 . 8 ) within the genomic region ( 250 Kb in each side ) [38] . We performed linear regression assuming an additive genetic model to test the association between each SNP and CSF biomarker levels by adjusting for age , gender and the first three principal components from the population stratification analysis . We confirmed a significant association ( multiple-testing threshold = 1 . 6×10−4 ) between an intronic SNP , rs76342307 ( MAF = 0 . 016 ) and CSF ptau ( p = 8 . 0×10−5 , Beta = 0 . 14 ) , tau ( p = 8 . 4×10−3 , Beta = 0 . 10 ) , and Aβ42 levels ( p = 0 . 02 , Beta = −0 . 06 ) ( Figure 1D–F ) for the PSEN1 genomic region . Rs76342307 is located 0 . 2 Mb 3′ upstream from the PSEN1 gene . We used data from the HapMap and the 1000 Genomes Project to identify all of the SNPs in linkage disequilibrium ( LD , r2>0 . 8 ) with rs76342307 . Six SNPs ( rs76342307 , rs17856583 , rs1110058 , rs117946815 , rs117236337 and rs2091912 ) were found to be in strong LD ( r2 = 0 . 95 , D′ = 1 ) with rs76342307 spanning 0 . 3 Mb ( Figure 1G , H ) . 100% and 97% concordance rates were observed among the directly typed and imputed results for rs76342307 and rs117236337 , respectively . Interestingly , rs117236337 is an intronic SNP in PSEN1 gene , which is also associated with extreme CSF tau ( p = 0 . 02 , Beta = 0 . 08 ) , ptau ( p = 5 . 7×10−4 , Beta = 0 . 09 ) and Aβ42 levels ( p = 0 . 01 , Beta = −0 . 06 ) . Next , we tested whether PSEN1 , p . E318G was in LD with the SNPs identified by the fine mapping analysis . In fact , rs17125721 ( PSEN1 , p . E318G ) is in moderate LD with all of them ( R2 = 0 . 68 , D′ = 1 ) ( Figure 1H ) . To analyze whether the p . E318G and rs76342307 are two independent signals , we ran a conditional analysis including both SNPs ( rs76342307 and rs17125721 ) in the model . When one of the SNPs was included in the model , the association from the other SNP disappeared , suggesting that the association in this locus is driven by a single signal ( Figure 1I ) . We observed that in the subset of individuals with Aβ deposition ( CSF Aβ42 levels lower than 500 pg/ml in WU-ADRC , and 192 pg/ml in ADNI ) [35] , [39] , the frequency of p . E318G carriers ( 4 . 2% , 21/500 ) was higher than in individuals without Aβ deposition ( 2 . 5% , 11/427 ) , although this difference did not achieve statistical significance ( p = 0 . 18 , OR = 1 . 6 , 95%CI = 0 . 78–3 . 4 ) ( Table 3 , 4 ) . In addition , we observed that 93% ( 15/16 ) of the individuals carrying PSEN1 , p . E318G along with APOE ε4 exhibited low CSF Aβ42 levels , while only 45% ( 9/20 ) of the individuals carrying PSEN1 , p . E318G but do not carry the APOE ε4 allele showed low CSF Aβ42 levels , suggesting that APOE ε4 allele is modifying the profile of Aβ deposition in PSEN1 , p . E318G carriers ( Table 4 and Figure 2A ) . APOE ε4 is strongly associated with CSF Aβ42 levels ( Table 4 ) [14] , [18] , and APOE genotype has been reported to modify disease expression in individuals with mutations in PSEN1 [40] and PSEN2 [41] genes . However , previous reports have not found any significant interaction between APOE and PSEN1 p . E318G , most likely due to the low frequency of PSEN1 , p . E318G and small sample sizes [42] , [43] , [44] . To analyze whether there was an APOE-dependent effect on this variant , we tested the association of p . E318G with CSF Aβ42 levels by stratifying it in the presence ( + ) or absence ( − ) of the APOE ε4 allele . We found that the risk of having Aβ deposition is greater for carriers of PSEN1 , p . E318G and APOE ε4 together ( OR = 18 . 3 CI = 2 . 0–166 . 8 , p = 3 . 5×10−3 ) than those carrying APOE ε4 allele alone ( OR = 4 . 5 , CI = 3 . 4–6 . 0 , p<1 . 0×10−5 ) ( Table 4 ) . These individuals are more likely to have a CSF biomarker profile similar consistent with AD ( low CSF Aβ42 , and high tau or ptau levels ) ( Figure 2A ) . p . E318G carriers who also carry APOE ε4+ allele ( n = 20 ) exhibited significantly higher CSF tau ( p = 0 . 04 ) and ptau ( p = 0 . 01 ) levels and significantly lower CSF levels of Aβ42 ( p = 0 . 02 ) compared to those that are p . E318G carriers but do not carry the APOE ε4 allele ( Figure 2 A , B ) . We also found a significant interaction ( p = 0 . 03 ) between APOE ε4+ and p . E318G in individuals with increased burden of Aβ deposition . Taken together , the results of the biomarker analyses suggest that PSEN1 , p . E318G is associated with higher levels of neuronal loss ( reflected by CSF tau and ptau levels ) and with Aβ deposition ( low Aβ42 CSF levels ) in an APOE ε4-dependent fashion . Because the purpose of this endophenotype-based approach is to identify variants implicated in disease , we tested whether the PSEN1 , p . E318G is associated with AD risk , tau/Aβ pathology or rate of cognitive decline in an APOE dependent manner . Analyses of the association between PSEN1 p . E318G and clinical AD status in an independent AD case-control series ( n = 1 , 855 , WU series ) revealed that the risk of AD is significantly higher for p . E318G/APOE ε4 carriers ( OR = 9 . 9 CI = 2 . 6–37 . 5 , p = 1 . 7×10−4 ) compared to individuals carrying APOE ε4 alone ( OR = 5 . 1 , CI = 4 . 1–6 . 3 , p = 3 . 2×10−59 ) ( Table 5 ) . This finding was replicated in an independent sample from the GERAD consortium ( n = 4 , 058 ) . In this dataset , the association of p . E318G with AD case-control status in the presence of at least one APOE ε4 allele ( OR = 10 . 3 , 95% CI = 4 . 1–25 . 5 , p = 4 . 1×10−8 ) was double the risk for AD in the presence of APOE ε4 alone ( OR = 4 . 1 , 95% CI = 3 . 5–4 . 8 , p = 1 . 1×10−79 ) . In the joint-analysis of these two independent series ( 5 , 161 individuals ) , the risk of developing AD in the p . E318G/APOE ε4 carriers ( OR = 10 . 1 , 95% CI = 4 . 8–20 . 9 , p = 9 . 0×10−12 ) is two-fold the AD risk of those that carry APOE ε4 allele alone ( OR = 4 . 4 , 95% CI = 3 . 9–5 . 0 , p = 6 . 8×10−139 ) ( Table 5 ) . In fact , we found that individuals who are APOE ε4 heterozygous and also carry the p . E318G variant are at similar AD risk ( OR = 10 . 7 , 95% CI = 4 . 7–24 . 6 , p = 2 . 5×10−10 ) as APOE ε4 homozygous ( OR = 9 . 9 , 95% CI = 7 . 2 . 9–13 . 6 , p = 5 . 5×10−76 ) and are at double the AD risk compared to APOE ε4 heterozygous that are not carrying p . E318G ( OR = 3 . 9 , 95% CI = 3 . 4–4 . 4 , p = 2 . 8×10−106 ) ( Table 6 , Figure 2C ) . In an independent analysis leveraging two prospective cohorts , the Religious Orders Study and Rush Memory and Aging Project , we confirmed a significant interaction between APOE4 and p . E318G with burden of neuritic plaques at autopsy ( n = 748; P = 0 . 01 ) but we failed to detect any significant association with neurofibrillary tangles ( p = 0 . 47 ) . Interestingly , the effect of APOE ε4 allele alone on neuritic plaques ( n = 748 , p = 4 . 5×10−24 , Beta = 0 . 39 ) was increased by two fold the presence of p . E318G ( n = 204 , p = 0 . 08 , Beta = 0 . 74 ) . p . E318G has previously associated with lower cognitive performance [45] . We tested whether the interaction between APOE4 and p . E318G affect the episodic memory . We found that there is trend between interaction between APOE4 and p . E318G with episodic memory decline ( p = 0 . 08 ) . Furthermore , the significant effect of APOE ε4 allele on episodic memory decline ( p = 1 . 7×10−16 , Beta = −0 . 06 ) was modified by the presence of p . E318G ( p = 0 . 14 , Beta = −0 . 16 ) . However , these interactions showed the predicted direction of effects for these phenotypes based on the results of the biomarker data: In the presence of at least one APOE-ε4 allele , p . E318G is associated with more Aβ plaques , faster cognitive decline and higher risk for AD . The p . E318G variant has been associated with familial AD in different populations [42] , [44] , [46] . However , this association has not been consistently replicated [43] , [47] , [48] , [49] . Our previous analyses indicate that in sporadic AD cases the effect of the p . E318G variant can be detected only in presence of the APOE ε4 allele . We wanted to analyze whether the same effect is found in familial cases . We genotyped probands from 565 total LOAD families and found the presence of PSEN1 p . E318G in 30 families ( MAF = 2 . 5% ) . PSEN1 p . E318G exhibited a higher frequency in individuals with familial LOAD than those with sporadic LOAD ( MAF = 1 . 6% , n = 3 , 989 , p = 0 . 02 ) and a group of age matched control subjects ( MAF = 1 . 5% , n = 830 , p = 0 . 03 ) . Next , we tested whether the association with familial LOAD was due to the interaction of p . E318G with APOE-ε4 allele . The presence of APOE-ε4 allele in p . E318G carriers in familial AD ( 70% , 21/30 ) was higher than that in sporadic AD ( 65% , 84/129 ) but not statistically significant ( p = 0 . 61 ) . On the other hand , APOE-ε4/p . E318G carriers in familial AD were significantly higher ( p = 4 . 0×10−4 ) than those in the control group ( 15% , 10/69 ) . Therefore , the risk conferred by APOE-ε4 and p . E318G carriers in familial AD ( OR = 16 . 4 , 95% CI = 5 . 6–48 . 2 , p = 5 . 8×10−8 ) compared to the control group was higher than the risk associated with sporadic AD ( OR = 10 . 1 , 95% CI = 4 . 8–20 . 9 , p = 9 . 0×10−12 ) . These results suggest that higher risk of the p . E318G variant in familial cases is mostly due to the high frequency of APOE ε4 allele in this population [28] . Interestingly , the p . E318G variant has been reported in multigenerational families with AD [42] , [50] . However , PSEN1 p . E318G is not considered pathogenic in part due to the absence of conclusive evidence for cosegregation with AD [34] , [43] , [47] , [48] . We observed 8 families ( with more than two affected individuals carrying p . E318G ) in which p . E318G segregates with disease ( Figure 2D ) , even in the absence of APOE-ε4 allele ( two families ) ( Figure S2 ) . These families do not carry any other mutations in APP , PSEN1 , PSEN2 , GRN and MAPT genes [28] . In three additional families the cosegregation p . E318G with AD was inconclusive because only a few family members had been sampled and/or because p . E318G carriers were below the mean age of onset for AD in their respective families . Thus , using the largest sample of familial LOAD screened to date for the role of p . E318G in AD , we have demonstrated that minor allele p . E318G increases the risk of familial LOAD . Furthermore , p . E318G cosegregates with AD in 26% of all the familial LOAD carriers . Carriers of PSEN1 , p . E318G have been reported across a wide range of ages ( 45 to 93 yrs . ) [42] , [44] , [46] , [50] . Thus , we tested whether PSEN1 , p . E318G affects AAO regardless of the APOE genotype; we found that PSEN1 , p . E318G carriers have a lower AAO than non-carriers ( 73 . 9 yr . vs . 78 . 2 yr . ; p = 0 . 01 ) ( Figure S3 ) .
Previous data have suggested that mutations in APP , PSEN1 , and PSEN2 genes only cause early-onset familial AD . However , this study and previous studies from our group [28] , [52] indicate that pathogenic mutations in these genes can be also found in late-onset familial and sporadic AD cases . In this study , we observed a known and confirmed pathogenic mutation ( PSEN1 p . A426P , rs63751223 ) in one individual ( 57 years old ) without a clear family history of dementia , out of 258 individuals ( CDR>0 ) , which constitutes 0 . 3% of AD cases . In a previous study , Cruchaga et al , found that 2 . 3% of families with multiple members affected by LOAD carried pathogenic mutations [28] . In this study , we expanded our analyses to sporadic cases , which constitute 95% of the total number of AD cases . Although we found only one case with a pathogenic mutation ( 0 . 3% ) , this could be an underestimate because both of the novel mutations , PSEN2: p . G270S and MAPT p . T263P were found in single cases that met biomarker criteria for AD . A novel variant in GRN , p . C247Y and a known variant in PSEN1 , p . R35Q were found in demented individuals with a non-AD CSF profile suggesting another type of dementia . However , without segregation analyses , additional functional studies are required to determine the potential pathogenicity of these variants . The classification of mutations as not pathogenic , possibly pathogenic , probably pathogenic and definitely pathogenic based on segregation analyses , amino acid conservation , effects on Aβ metabolism in in vitro studies , association studies and presence in healthy individuals has been useful in prioritizing mutations and their likelihood of affecting risk for disease [47] . However , this classification is likely to miss variants with a smaller but real effect ( OR>2 . 0 ) on risk for sporadic AD . The variant GRN , p . P458L is classified as non-pathogenic [34] due to fact that it was reported in an ALS/FTD patient and in 25 out of 492 controls ( MAF = 2 . 5% ) [53] . However , this variant is not reported in the EVS server ( 6 , 515 exomes ) ( EVS-v . 0 . 0 . 18 , ( February 8 , 2013 ) or in our control population of 824 samples ( Table 2 ) . Here , this variant was found in an individual with early onset dementia and with typical biomarker criteria for AD . PSEN2 , p . R71W has been classified as non-pathogenic because it was reported in controls and EOAD cases [34] . However , in a previous study the frequency of the p . R71W variant in AD cases was significantly higher than in controls ( n = 3 , 152 , p = 9 . 0×10−4 OR = 6 . 45; 95%CI = 1 . 95–21 . 39 ) and carriers have a significantly earlier age at onset than affected non-carriers ( p . R71W: 70 . 2 vs . 76 . 7 , p = 5 . 0×10−4 ) , suggesting that this variant could be a modifier of LOAD risk [28] . Here , we found the same trend , PSEN2 p . R71W was also found to be present more frequently in clinical cases than in controls ( p = 0 . 03 , OR = 10 . 3 , 95%CI = 1 . 1–96 . 2 ) . However , it did not reach statistical significance in individuals with Aβ deposition ( p = 0 . 27 , OR = 3 . 4 , 95%CI = 0 . 38–30 . 7 ) . The PSEN1 , p . E318G variant has been considered to be a non-pathogenic variant , because it has been found in non-demented individuals [43] , [48] , [49] and the absence of conclusive evidence for cosegregation with AD [43] . However , it has been suggested that phenocopies , potential presymptomatic individuals , reduced penetrance and gene by gene interactions complicate the interpretation of the p . E318G variant in familial and sporadic LOAD [42] , [44] . This is the first study to systematically screen the presence of PSEN1 p . E318G in a large ( n = 565 ) clinical series of well-characterized families densely affected by LOAD with no mutations in APP , PSEN2 , GRN or MAPT genes . PSEN1 p . E318G was found in 5 . 3% and cosegregated with the disease in 1 . 4% of all families . We also found that PSEN1 p . E318G exhibited a higher frequency in familial LOAD than in sporadic LOAD ( p = 0 . 025 ) , supporting earlier findings that the p . E318G variant has higher frequencies among AD cases with a family history of AD in different populations [42] , [44] , [46] . Additionally , our analyses indicate that PSEN1 p . E318G carriers have an average age at onset that is 4 . 3 years earlier than that in non-carriers ( 73 . 9 yr . vs . 78 . 2 yr ) . Putative pathogenic variants in genes that cause late-onset rather than early-onset dementia could have a less severe effect on protein function due to genetic or environmental modifiers [28] . Our CSF biomarker analyses suggested that PSEN1 p . E318G was associated with higher levels of neuronal loss ( reflected by high CSF tau and ptau levels ) and with Aβ deposition ( low Aβ42 CSF levels ) in an APOE ε4-dependent fashion . Furthermore , in the largest AD case-control series ( n = 5 , 161 ) analyzed for the interaction between PSEN1 p . E318G and APOE ε4 allele to date , we found that the presence of p . E318G and APOE ε4 doubles the risk for AD ( OR = 10 . 3 , 95% CI = 4 . 1–25 . 5 ) compared to the risk with the presence of APOE ε4 alone ( OR = 4 . 1 , 95% CI = 3 . 5–4 . 8 ) . There are several reports of variants that modify the risk of AD in APOE ε4 carriers such as α-1-antichymotrypsin ( ACT ) gene ( APOE ε4/ACT , [OR = 6 . 4 , non 95% CI reported] ) [54] , Cholesteryl ester transfer protein ( CETP ) gene ( APOE ε4/CETP [–629] C allele [OR 7 . 12 , non 95% CI reported] ) [55] , GRB-associated binding protein 2 ( GAB2 ) gene ( APOE ε4/rs2373115 genotype GG [OR = 2 . 36 , 95% CI 1 . 55–3 . 58] ) [56] , CUG triplet repeat , and RNA binding protein 2 ( CUGBP2 ) gene ( APOE ε4/ε4/rs62209 [OR = 1 . 75 , 95% CI 1 . 27–2 . 41] ) [57] . However , all these variants have a modest effect increasing the risk due to APOE ε4 allele . Here , we provided evidence of a low frequency variant in PSEN1 gene with a significant effect on the AD risk in APOE ε4 carriers ( OR = 10 . 7 , 95% CI = 4 . 7–24 . 6 ) comparable only to the effect of a second APOE ε4 allele ( OR = 9 . 9 , 95% CI = 7 . 2 . 9–13 . 6 ) . Moreover , we also found that in the presence of at least one APOE ε4 allele , p . E318G is associated with more Aβ plaques and faster cognitive decline , as recently reported for a low frequency variant in complement receptor 1 ( CR1 ) [58] In addition , p . E318G has previously associated with lower cognitive performance , which support our findings of cognitive decline [45] . The interaction of the p . E318G with APOE ε4 allele was replicated in four different datasets: the CSF dataset ( discovery set ) , WU_ADRC case-control dataset , GERAD1 and the Religious Orders Study and Rush Memory and Aging Project , indicating that this association and interaction is not a type I error , but a real association . All these results together support the role of PSEN1 p . E318G as one of the most important modifiers of the risk of LOAD reported to date . Functional studies , especially concerning the effect on Aβ metabolism in vitro , have further questioned the pathogenicity of the p . E318G variant . One study showed no alteration in the production of Aβ42 induced by p . E318G [43] . However , a recent study using skin fibroblasts from individuals with the p . E318G variation showed an increase in the production of Aβ40 , a decrease in Aβ42 and a subsequent significant reduction in the Aβ42/Aβ40 ratio compare to non-carriers [42] , along with a lack of an inhibitory effect of the exon 9 loop in the presence of the p . E318G variant reported by an independent study [59] . It has been proposed that the activation of γ-secretase results from a cleavage-induced conformational change that relieves the inhibitory effect of the intact exon 9 loop , which is mediated by occupying the substrate-binding site on the immature enzyme before it is cleaved [59] . It was reported that p . E318G abolishes the inhibitory effect of the intact exon 9 loop , which favors the production of Aβ40 [59] . It was also reported that p . E318G affects the processing of PSEN1 by reducing the amount of N-terminal fragment that is generated after cleavage [60] , and augments levels of neuronal cell death after overexpression [61] . We suggest that another approach to test the impact of pathogenic mutations on Aβ metabolism is to examine the effect on the CSF biomarker levels . Most of the published data about CSF biomarkers reveal that PSEN1 gene mutation carriers display a typical AD biomarker signature with low CSF levels of Aβ42 and high CSF tau levels [13] , [20] . There is no published data on the levels of CSF biomarkers for PSEN1 , p . E318G carriers . Here , for the first time we demonstrate that PSEN1 , p . E318G/APOE ε4 carriers have a CSF biomarker profile similar to AD cases . In summary , these results highlight the relative enrichment of low frequency variants in six genes involved in AD and FTD that are at the extremes of the distribution of CSF biomarker levels [29] . We provide evidence that the PSEN1 , p . E318G variant increases the risk for AD in APOE ε4 heterozygous , equivalent to that of APOE ε4 homozygous . We also found that p . E318G increases the risk of familial LOAD and cosegregates with AD in 26% of all the familial LOAD carriers . All these findings have important implications for genetic counseling since PSEN1 , p . E318G is currently considered a non-pathogenic variant [50] . By using CSF biomarker levels as a quantitative trait , we were able to identify a low frequency variant associated with AD risk , PSEN1 , p . E318G . This association is mediated by a SNP-by-SNP interaction , which has not been found using the standard case-control design [43] , [48] , [49] . Together , these results indicate that there are potentially many more low frequency variants associated with complex disease , and that the association results from complex interactions . We were able to identify the association of PSEN1 , p . E318G with risk for AD and its interaction with the APOE ε4 allele because both genes are known to be associated with AD . However , the identification of such an association and interactions in a genome-wide approach remains still challenging and requires novel , powerful approaches . We believe that this endophenotype-based approach is a good alternative to case-control studies and can allow us to gain a better understanding of both the genetic architecture and pathophysiology of LOAD [17] , [18] . In terms of genetics and factors that may explain some of the missing hereditability of complex diseases , these results are important because they are a clear example of low frequency variants that are associated with disease and how such associations are due to epistatic gene by gene interactions .
The Institutional Review Board ( IRB ) at the Washington University School of Medicine in Saint Louis approved the study . Prior to their participation , a written informed consent was reviewed and obtained from family members . The Human Research Protection Office ( HRPO ) approval number for our ADRC Genetics Core family studies is 93-0006 . Two CSF series were used for this study . A total sample of 475 individuals enrolled in longitudinal studies at the Alzheimer's disease Research Center at Washington University School of Medicine ( ADRC ) and 259 participants of the Alzheimer's disease Neuroimaging Initiative ( ADNI ) were used in this study . A subset of 145 participants from ADRC and 67 from ADNI were included in the discovery series ( two DNA pools ) . CSF samples were from individuals of European descent . In the WU-ADRC-CSF series: 60% of sample is female , ranging from 37–91 years of age . 73% of the sample has a clinical dementia rating ( CDR ) of 0 ( cognitively normal ) and 39% of the individuals carry at least one APOE ε4 allele . In the ADNI-CSF series: 44% of sample is female , ranging from 56–91 years of age . 60% of the sample has a CDR higher than 0 ( demented ) and 47% are APOE ε4 allele positive . Table 1 summarizes the demographic data for the CSF series . Covariate-adjusted residuals of CSF Aβ42 , tau and p-tau were used to define the pools ( see statistical analysis , Table S3 in Text S1 ) . 114 individuals in the bottom 15% of CSF Aβ42 levels or individuals in the top 15% of CSF tau or p-tau levels were included in a pool . The second pool consisted of 98 individuals in the top 15% of CSF Aβ42 or individuals in the bottom 15% of tau and p-tau181 levels ( Table 1 ) . The Religious Orders Study ( ROS ) and the Rush Memory and Aging Project ( MAP ) recruit participants without known dementia who agree to annual clinical evaluations and sign an Anatomic Gift Act donating their brains at death . The full cohort with genotype data included 1 , 708 subjects ( 817 ROS and 891 MAP ) . The mean age at enrollment was 78 . 5 years and 69 . 1% were female . At the last evaluation , 24 . 9% met clinical diagnostic criteria for AD and 21 . 8% had mild cognitive impairment . The summary measure of global cognitive performance was based on annual assessments of 17 neuropsychiatric tests . A nested autopsy cohort consisted of 651 deceased subjects ( 376 ROS and 275 MAP ) ; mean age at death was 81 . 5 years and 37 . 6% were male . Proximate to death , 40 . 9% of subjects included in the autopsy cohort met clinical diagnostic criteria for AD . Bielschowsky silver stain was used to visualize neurofibrillary tangles in tissue sections from the midfrontal , middle temporal , inferior parietal , and entorhinal cortices , and the hippocampal CA1 sector . A quantitative composite score for neurofibrillary tangle pathologic burden was created by dividing the raw counts in each region by the standard deviation of the region specific counts , and then averaging the scaled counts over the 5 brain regions to create a single standardized summary measure . Additional details of the ROS and MAP cohorts as well as the cognitive and pathologic phenotypes are described in prior publications [58] , [62] . Follow-up series included 1 , 031 sporadic AD cases , 824 unrelated elderly cognitively normal controls and a single case from NIA-LOAD families ( n = 595 ) [28] . All these samples are independent of the CSF samples . Cases received a diagnosis of dementia of the Alzheimer's type ( DAT ) , using criteria equivalent to the National Institute of Neurological and Communication Disorders and Stroke-Alzheimer's Disease and Related Disorders Association for probable AD [63] , [64] . Controls received the same assessment as the cases but were non-demented . All individuals were of European descent and written consent was obtained from all participants . DNA from ROS and MAP subjects was extracted from whole blood , lymphocytes or frozen post-mortem brain tissue and genotyped on the Affymetrix Genechip 6 . 0 platform , as previously described [58] . Following standard QC procedures , imputation was performed using MACH software ( version 1 . 0 . 16a ) and HapMap release 22 CEU ( build 36 ) as a reference . Association of Aβ42 , tau and p-tau181 with genetic variants was analyzed as previously reported [14] , [17] , [18] . Briefly , Aβ42 , tau and p-tau181 values were log transformed to approximate a normal distribution . Because the CSF biomarker levels were measured using different platforms ( Innotest plate ELISA vs . AlzBia3 bead-based ELISA , respectively ) we were not able to combine the raw data . For the combined analyses we standardized the mean of the log transformed values from each dataset to zero . A stepwise discriminant analysis identified CDR , APOE genotype , gender and age as significant covariates in both series ( Table S1b in Text S1 ) [17] , [18] . No significant differences in the transformed and standardized CSF values for different series were found ( Table S1b in Text S1 ) . CSF biomarker levels were used as a quantitative trait for most analyses . It has been shown that CSF Aβ42 is an accurate predictor of brain amyloid burden regardless of clinical diagnosis [39] . Therefore , the Aβ plaque deposition was assumed using the biomarker levels as a dichotomous variable ( low and high CSF Aβ42 ) . Levels of CSF biomarkers were as follows: for the ADNI-CSF series the cut-off was Aβ42<192 pg/mL [35] . In the WU-ADRC-CSF series , we used CSF Aβ42<500 pg/mL as the cut-off [39] . We used Plink ( http://pngu . mgh . harvard . edu/~purcell/plink/ ) to analyze the association of variants ( individually or collapsed by gene ) with CSF biomarker levels . Age , gender and site were included as covariates . In order to determine whether the association of variants with CSF biomarker levels was driven by case-control status we included clinical dementia rating ( CDR ) or CSF Aβ42 levels as a covariate in the model or stratified the data by case control status . We also performed analyses including APOE genotype as a covariate . Association with AAO was carried out using the Kaplan-Meier method and tested for significant differences , using a log-rank test [17] . Fisher's exact test was used to compare the frequency of each variant and collapse by gene in the case control series defined by CDR or CSF Aβ42 levels ( Table 3 ) . All variants were included in the model independent of their pathogenicity . Analyses of SNP effects on global cognitive decline in ROS and MAP were performed as in prior publications [62] . Briefly , we first fit linear mixed effects models using the global cognitive summary measure in order to characterize individual paths of change , adjusted for age , sex , years of education , and their interactions with time . At least two longitudinal measures of cognition were required for inclusion in these analyses , for which data on 1 , 593 subjects was available . We then used these person-specific , residual cognitive decline slopes as the outcome variable in our linear regression models , with each SNP of interest coded additively relative to the minor allele , and further adjusted for study membership ( ROS vs . MAP ) and the first 3 principal components from population structure analysis . For analyses of neurofibrillary tangle burden , linear regression was used to relate SNPs to the pathologic summary measure , adjusting for age at death , study membership , and 3 principal components . Because the data were skewed , square root of the scaled neurofibrillary tangle burden summary score was used in analyses . Pooled-DNA sequencing was performed , as previously described by Druley TE et al . [28] , [52] , [65] , [66] . Briefly , equimolar amounts of individual DNA samples were pooled together after being measured using Quant-iT PicoGreen reagent . Two different pools with 100 ng of DNA from 114 and 98 individuals were made . The coding exons and flanking regions ( a minimum of 50 bp each side ) were individually PCR amplified using specific primers and Pfu Ultra high-fidelity polymerase ( Stratagene ) . An average of 20 diploid genomes ( approximately 0 . 14 ng DNA ) per individual were used as input into a total of 62 PCR reactions that covered 46 , 319 bases from the 6 genes . PCR products were cleaned using QIAquick PCR purification kits , quantified using Quant-iT PicoGreen reagent and ligated in equimolar amounts using T4 Ligase and T4 Polynucleotide Kinase . After ligation , concatenated PCR products were randomly sheared by sonication and prepared for sequencing on an Illumina Genome Analyzer IIx ( GAIIx ) according to the manufacturer's specifications . pCMV6-XL5 amplicon ( 1908 base pairs ) was included in the reaction as a negative control . As positive controls , ten different constructs ( p53 gene ) with synthetically engineered mutations at a relative frequency of one mutated copy per 250 normal copies was amplified and pooled with the pcr products . Six DNA samples heterozygous for previously known mutants in GRN , PSEN1 , MAPT genes were also included . Single reads ( 36 bp ) were aligned to the human genome reference assembly build 36 . 1 ( hg18 ) using SPLINTER [67] . SPLINTER uses the positive control to estimate sensitivity and specificity for variant calling . The wild type: mutant ratio in the positive control is similar to the relative frequency expected for a single mutation in one pool ( 1 chromosome mutated in 125 samples = 1/250 ) . SPLINTER uses the negative control ( first 900 bp ) to model the errors across the 36-bp Illumina reads and to create an error model from each sequencing run of the machine . Based on the error model SPLINTER calculates a p-value for the probability that a predicted variant is a true positive . A p-value at which all mutants in the positive controls were identified was defined as the cut-off value for the best sensitivity and specificity . All mutants included as part of the amplified positive control vector were found upon achieving >30-fold coverage at mutated sites ( sensitivity = 100% ) and only ∼80 sites in the 1908 bp negative control vector were predicted to be polymorphic ( specificity = ∼95% ) . The variants with a p-value below this cut-off value were considered for follow-up confirmation . All rare missense or splice site variants ( with an estimated allelic frequency less than 5% ) were then validated by Sequenom and KASPar genotyping in each individual included in the pools [28] , [52] , [66] . The validated SNPs were then genotyped in all members of the WU-ADRC-CSF and ADNI-CSF series . Common variants ( >5% ) and synonymous variants were not followed up . An average coverage of 30X-fold per allele per pool is the minimum coverage necessary to get an optimal positive predictive value for the SNP-calling algorithm [67] . The necessary number of lanes to obtain a minimum of 30-fold coverage per base and sample were run ( Table S2 in Text S1 ) . The WU-ADRC samples were genotyped with the Illumina 610 or OmniExpress . The ADNI samples were genotyped with the Illumina 610 chip . Prior to association analysis , all samples and genotypes underwent stringent quality control ( QC ) . Genotype data were cleaned applying a minimum call rate for SNPs and individuals ( 98% ) and minimum minor allele frequencies ( 0 . 02 ) . SNPs not in Hardy-Weinberg equilibrium ( P<1×10−6 ) were excluded . The QC cleaning steps were applied for each genotyping array separately . We tested for unanticipated duplicates and cryptic relatedness using pairwise genome-wide estimates of proportion identity-by-descent . When a pair of identical samples or a pair of samples with cryptic relatedness was identified , the sample from the WU-ADRC or samples with a higher number of SNPs passing QC were prioritized . Eigenstrat was used to calculate principal component factors for each sample and confirm the ethnicity of the samples [68] . The 1000 Genome Project data ( June 2011 release ) and Beagle software were used to impute up to 6 million SNPs . SNPs with a Beagle R2 of 0 . 3 or lower , a minor allele frequency ( MAF ) lower than 0 . 05 , out of Hardy-Weinberg equilibrium ( p<1×10-6 ) , a call rate lower than 95% or a Gprobs score lower than 0 . 90 were removed . A total of 5 , 815 , 690 SNPs passed the QC process . We used PLINK to select the list of SNPs in the gene region ( approximately 250 kb of flanking sequence each side ) from the imputed data . These SNPs were pruned with an r2 cutoff of 0 . 8 . . The simple method [38] was used to calculate the number of informative SNPs within the genomic region for each gene . This measure was then used in a Bonferroni adjustment to estimate the significance threshold . Significant SNPs that were imputed or have a MAF<10% were directly genotyped in all the samples to confirm the association . The AD&FTD mutation database [34] was used to identify sequence variants previously found in other studies of early onset familial dementia and to determine whether or not they are considered to be disease-causative variants . The sequencing data from the 1 , 000 Genome Project and the Exome Variant Server data base ( http://evs . gs . washington . edu/EVS/ ) were used to estimate the frequency of novel and rare ( minor allele frequency less than 5% ) missense , nonsense and splice site variants in samples unselected for studies of AD . Conservation was determined by using the GERP score , which calculates the conservation of each nucleotide in multi-species alignment . A site was called conserved when the GERP score was greater than or equal to 4 [69] , [70] . Data used in the preparation of this article were obtained from the ADNI database ( www . loni . ucla . edu\ADNI ) . The ADNI was launched in 2003 by the National Institute on Aging , the National Institute of Biomedical Imaging and Bioengineering , the Food and Drug Administration , private pharmaceutical companies and non-profit organizations , as a $60 million , 5-year public-private partnership . The Principal Investigator of this initiative is Michael W . Weiner , M . D . ADNI is the result of efforts of many co-investigators from a broad range of academic institutions and private corporations , and subjects have been recruited from over 50 sites across the U . S . and Canada . The initial goal of ADNI was to recruit 800 adults , ages 55 to 90 , to participate in the research -approximately 200 cognitively normal older individuals to be followed for 3 years , 400 people with MCI to be followed for 3 years , and 200 people with early AD to be followed for 2 years . ” For up-to-date information see www . adni-info . org . Data used in the preparation of this article were obtained from the Genetic and Environmental Risk for Alzheimer's disease ( GERAD1 ) Consortium [71] . The GERAD1 sample comprised up to 3941 AD cases and 7848 controls . A subset of this sample has been used in this study , comprising 3333 cases and 1225 elderly screened controls genotyped at the Sanger Institute on the Illumina 610-quad chip . These samples were recruited by the Medical Research Council ( MRC ) Genetic Resource for AD ( Cardiff University; Kings College London; Cambridge University; Trinity College Dublin ) , the Alzheimer's Research Trust ( ART ) Collaboration ( University of Nottingham; University of Manchester; University of Southampton; University of Bristol; Queen's University Belfast; the Oxford Project to Investigate Memory and Ageing ( OPTIMA ) , Oxford University ) ; Washington University , St Louis , United States; MRC PRION Unit , University College London; London and the South East Region AD project ( LASER-AD ) , University College London; Competence Network of Dementia ( CND ) and Department of Psychiatry , University of Bonn , Germany and the National Institute of Mental Health ( NIMH ) AD Genetics Initiative . All AD cases met criteria for either probable ( NINCDS-ADRDA , DSM-IV ) or definite ( CERAD ) AD . All elderly controls were screened for dementia using the MMSE or ADAS-cog , were determined to be free from dementia at neuropathological examination or had a Braak score of 2 . 5 or lower .
|
Alzheimer's disease ( AD ) is the most common neurodegenerative disease affecting more than 5 . 3 million people in the US . AD-causing mutations have been identified in APP , PSEN1 and PSEN2 genes . Heterozygous carriers of APOE-ε4 allele exhibit a 3-fold increased risk for developing AD , while homozygous carriers show a 10-fold greater risk than non-carriers . Here , we sequenced individuals with extreme levels of well-established AD cerebrospinal fluid ( CSF ) biomarkers in order to identify variants in APOE , APP , PSEN1 , PSEN2 , GRN and MAPT genes associated with AD risk . This approach allowed us to identify known pathogenic variants , additional AD risk genetic factors and identify a low frequency variant in PSEN1 , p . E318G ( rs17125721-G ) that increases risk for AD in a gene-gene interaction with APOE . These findings were replicated in three large ( >4 , 000 individuals ) and independent datasets . This finding is particularly important because we demonstrated that a currently considered non-pathogenic variant is associated with higher levels of neuronal degeneration , and with Aβ deposition , more Aβ plaques and faster cognitive decline in an APOE-ε4-dependent fashion . APOE-ε4 heterozygous individuals who carry this variant are at similar AD risk as APOE-ε4 homozygous individuals .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"neurodegenerative",
"diseases",
"population",
"genetics",
"quantitative",
"traits",
"neuroscience",
"epistasis",
"mutation",
"alzheimer",
"disease",
"genetic",
"polymorphism",
"biology",
"dementia",
"genetic",
"association",
"studies",
"heredity",
"genetic",
"screens",
"neurology",
"genetics",
"human",
"genetics",
"genetics",
"of",
"disease",
"complex",
"traits"
] |
2013
|
The PSEN1, p.E318G Variant Increases the Risk of Alzheimer's Disease in APOE-ε4 Carriers
|
Helicobacter pylori infection causes chronic active gastritis that after many years of infection can develop into peptic ulceration or gastric adenocarcinoma . The bacterium is highly adapted to surviving in the gastric environment and a key adaptation is the virulence factor urease . Although widely postulated , the requirement of urease expression for persistent infection has not been elucidated experimentally as conventional urease knockout mutants are incapable of colonization . To overcome this constraint , conditional H . pylori urease mutants were constructed by adapting the tetracycline inducible expression system that enabled changing the urease phenotype of the bacteria during established infection . Through tight regulation we demonstrate that urease expression is not only required for establishing initial colonization but also for maintaining chronic infection . Furthermore , successful isolation of tet-escape mutants from a late infection time point revealed the strong selective pressure on this gastric pathogen to continuously express urease in order to maintain chronic infection . In addition to mutations in the conditional gene expression system , escape mutants were found to harbor changes in other genes including the alternative RNA polymerase sigma factor , fliA , highlighting the genetic plasticity of H . pylori to adapt to a changing niche . The tet-system described here opens up opportunities to studying genes involved in the chronic stage of H . pylori infection to gain insight into bacterial mechanisms promoting immune escape and life-long infection . Furthermore , this genetic tool also allows for a new avenue of inquiry into understanding the importance of various virulence determinants in a changing biological environment when the bacterium is put under duress .
The human gut pathogen Helicobacter pylori has coevolved with humans over thousands of years to dominate the gastric niche [1–3] . The majority of infected individuals ( 80–90% ) carry and transmit H . pylori without any symptoms of disease [4 , 5] . However , H . pylori infection causes chronic active gastritis that may develop into peptic ulceration ( 10–20% ) or gastric adenocarcinoma ( 0 . 5–2% ) [6 , 7] causing a significant burden on public health [8–10] . H . pylori infection is persistent and clinical disease usually develops after many years of chronic inflammation and epithelial damage . Furthermore , due to increasing rates of antibiotic treatment failure [11 , 12] there is a pressing need for further research into the bacterium’s mechanisms for persistence and immune evasion strategies . These are of particular importance to understanding H . pylori pathogenesis and to identifying novel targets for the development of new treatment options . H . pylori is highly adapted to colonizing and surviving in the harsh conditions of the gastric environment . One key adaptation is the virulence factor urease . This multimeric enzyme , consisting of 12 UreA and UreB heterodimers , catalyses the hydrolysis of urea to produce CO2 and NH3 , which acts to buffer the acidity of the local environment around the cell [13 , 14] . Urease is abundantly expressed by H . pylori at levels exceeding that of any other known microbe [15] and is estimated to constitute 10–15% of the bacterium’s total protein content [16] . In addition , urease is essential for establishing colonization as H . pylori urease mutants are unable to infect the host [17–21] . Several lines of evidence suggest that urease plays a significantly greater role in infection than simple acid neutralization . Elevating the gastric pH to 7 . 0 was shown to be insufficient in permitting colonization by a urease negative strain [18] . In several in vitro studies urease and its catalytic products contributed directly to virulence . Ammonia produced by urease activity caused damage to the gastric epithelium by disrupting tight cell junction integrity [22 , 23] and CO2 protected against the bactericidal activity of the nitric oxide metabolite , peroxynitrite , produced by phagocytes to kill engulfed bacteria [24] . Furthermore , several studies suggest that urease may directly interact with host epithelial and immune cells . Urease has been shown to bind to major histocompatibility complex ( MHC ) class II molecules on gastric epithelial cells thereby inducing cell apoptosis [25] and the UreB subunit can stimulate monocytes to release proinflammatory cytokines by binding to cell surface CD74 , a MHC class II associated invariant chain [26 , 27] . In addition , an in vivo study demonstrated that changes to the surface of the urease complex resulted in the eventual clearance of H . pylori infection in mice [28] . Loss of colonization was attributed to the disruption in urease mediated interactions between H . pylori and host cells as urease activity was unaffected by the mutation , ruling out loss of acid resistance or nitrogen assimilation [29] as contributing factors [28] . Clinical isolates maintain high urease activity even after years of chronic infection when the bacterium has established itself in the relatively neutral environment of its gastric niche implicating that ongoing urease expression is required for persistence . However due to the lack of appropriate genetic systems this hypothesis could not be tested experimentally . The necessity of urease activity in establishing colonization hinders the study of its function during persistence when using conventional knockout mutants . The availability of a conditional urease mutant would overcome this constraint by permitting changes to the urease phenotype during an established infection . To conclusively determine if urease is indeed required after colonization is established , we generated conditional H . pylori urease knockout mutants using a tetracycline repressor ( tet ) based system [30] . This system controls gene expression by way of a tetracycline repressor ( TetR ) that binds to specific operator sequences ( tetO ) in the target promoter and silences transcription of the downstream gene . Expression of the target gene can be turned on by the administration of a potent tetracycline inducer , such as anhydrotetracycline ( ATc ) or doxycycline ( Dox ) [31] . This system was recently adapted to H . pylori and gene expression was regulated in vivo during active infection [32] . In the current study , we adapted this system to generate conditional urease mutants . We demonstrate that in an established infection , loss of urease expression is detrimental to the bacterial survival in the host . Strong selective pressure on the bacteria for continuous urease expression is further demonstrated by the emergence of escape mutants that successfully repopulated the mouse stomach six weeks after genetic silencing of urease was initiated .
The urease structural genes , ureA and ureB , encoding for the 27 kDa UreA and 62 kDa UreB urease subunits , are transcribed as a single operon under the control of the ureA promoter , PureA [33] . To regulate the expression of urease in H . pylori , we placed the operon under tet control . Based on previous mutational studies of PureA [32 , 34] , the promoter was mutated to incorporate one or more tetO sequences to generate a series of different PureA derivatives , urePtetO ( I-V ) ( Fig 1A and 1B ) . These tet-promoter constructs were made using PCR based techniques and used to replace the native chromosomal PureA by allelic replacement . This strategy involved first generating a recipient H . pylori strain in which PureA and ureA has been replaced with a rpsL-cat cassette . The urease negative recipient strain was then naturally transformed with the urePtetO PCR constructs to generate strains with tetO modified PureA derivatives and a restored ureA gene . H . pylori strains harbouring these constructs were characterized to identify a tet-promoter construct with regulatory properties that would permit the appropriate level of complementation to ensure colonization yet could be sufficiently silenced to prevent infection . The functionality of these tet-promoters was first assessed in the wild-type background for their ability to drive urease expression by measuring urease enzymatic activity , UreB expression and mouse colonization . The urePtetO constructs were introduced into the wild-type X47 strain , replacing the chromosomal PureA and generating X47 urePtetO strains ( OND2018—OND2022 ) . The urease expression level and the urease enzymatic activity of these strains under standard growing conditions was measured and compared to the parent strain ( Fig 1C and 1D ) . Despite expressing less urease , as determined by immunodetection of UreB in the total cell lysate ( Fig 1C ) , the urease activity measured in strains transformed with urePtetOI , urePtetOII and urePtetOV was found to be comparable to wild-type ( Fig 1D ) . In comparison , the urease activity in strains transformed with urePtetOIII and urePtetOIV were reduced by 60% and 25% respectively , which was also accompanied by substantially reduced amount of UreB in the total cell lysate ( Fig 1C ) . Given that replacement of the wild type PureA with the urePtetO promoters resulted in reduced urease expression which concomitantly may reduce their ability to establish infection , C57BL/6J mice were challenged with strains X47 urePtetO ( I-V ) ( OND2018—OND2022 ) to assess if urease expression in these strains was still sufficient to facilitate colonization . All five X47 urePtetO strains were successfully re-isolated from mouse stomachs two weeks after oral challenge ( Fig 1E ) . The infection rate and bacterial load in mice challenged with strains X47 urePtetOI , X47 urePtetOII and X47 urePtetOV was comparable to the control group challenged with the wild-type strain . However the infection rates were decreased for groups challenged with strains of lower urease activity , with strain X47 urePtetOIII displaying a major defect in colonization . To verify that colonization had not been established due to mutation or reversion of urePtetO , the ureA promoter region of re-isolated strains was sequenced which confirmed that the sequences of the urePtetO constructs remained unaltered after passage through mice . After establishing that the tetO modifications to the ureA promoter did not abrogate colonization per se , we evaluated if these promoters could regulate urease expression in a tetracycline dependent manner . All five ureA promoter derivatives were transformed into a X47 recipient strain that expressed TetR under the control of the strong flaA promoter [32] . The resulting strains , X47 mdaB::ptetR4 , urePtetO ( I-V ) ( OND1954—OND1958 ) , did not express urease when grown on standard CBA plates however urease expression could be induced when grown on CBA plates containing 50 ng/ml anhydrotetracycline ( ATc ) ( S1 Fig ) . These results demonstrated that TetR effectively silenced the tet-promoters in these strains . Regulation of urePtetO promoters by TetR in conditional urease knockout strains X47 mdaB::ptetR4 , urePtetO ( I-V ) was assessed using the urease enzymatic activity assay . Bacteria were cultured in the absence or presence of 50 ng/ml ATc for two successive passages and then collected for analysis . When strains were cultured in the absence of ATc , urease activity was below the detection limits of the assay ( 2 U/ml of Type III urease from Jack bean ) ( Fig 2A ) . For strains grown in the presence of ATc , urease activity in strains X47 ptetR4; urePtetO-I , -II and -V was induced to wild type levels , while the urease activity for strains X47 ptetR4; urePtetO-III and -IV remained below 10% of wild-type activity . These results demonstrated that by using the appropriate tet-promoter urease activity can indeed be regulated by the presence of a small molecule inducer , confirming the generation of conditional H . pylori urease knockout mutants . The tet-responsive promoters urePtetOI and urePtetOV have different genetic architectures ( Fig 1B ) and upon induction also promoted the greatest urease expression levels amongst the tested strains . Based on these results the regulation of these two promoters was further characterized . The kinetics of urePtetO induction and repression was analysed in strain X47 ptetR4; urePtetOI ( OND1954 ) and X47 ptetR4; urePtetOV ( OND1958 ) by immunodetection of the UreB protein ( Fig 2B and 2C ) . After addition of 200 ng/ml ATc to the culture medium , UreB protein expression increased over time and reached maximum levels after 12 h and 8 h for urePtetOI and urePtetOV , respectively ( Fig 2B ) . Withdrawal of ATc from induced cultures led to a significant decrease in UreB protein levels within 3 h , demonstrating that both urePtetOI and urePtetOV were quickly silenced ( Fig 2C ) and that the UreB protein was turned over efficiently , falling to the threshold of detection within 12 h . With the knowledge that tetracycline dependent regulation of urease expression was attainable in vitro we next turned our attention to establishing a mouse model of infection . Based on previous studies involving in vivo tet-systems [35–37] the inducer molecule doxycycline ( Dox ) was first used as a model inducer to identify a maximal dosage of material that could be tolerated by the bacterium in vivo . We found that wild-type X47 could still infect mice when the animals were supplemented with up to 10 mg/l of Dox in their drinking water . Colonization by wild-type X47 was severely attenuated at 100 mg/l of Dox and bacteria could not be reisolated at 1000 mg/l of Dox supplement ( S2A Fig ) . Furthermore , strain X47 ptetR4; urePtetOI ( OND1954 ) , which emerged as the prime conditional urease mutant candidate from the in vitro studies , was tested to verify its ability to establish initial colonisation and then used to optimise the dosage of inducer molecule to regulate urease expression in vivo . OND1954 was only capable of establishing infection in C57BL/6J mice when the animals received Dox supplementation in their drinking water , demonstrating that urease is essential for OND1954 to establish colonization in the mouse infection model ( S2B Fig ) . Addition of Dox supplement at 1 mg/l supported colonization of OND1954 and although attenuated , the conditional mutant was also isolated from animals supplemented with Dox at 5 mg/l and 10 mg/l . Having identified a minimal supplement dose of Dox inducer , we then sought to complete the infection model by investigating two more important factors; the use of the less toxic tetracycline derivative ATc , and attempting to improve the robustness of the conditional urease mutant strain . The original wild-type X47 strain underwent four consecutive transformations to generate the conditional urease mutant OND1954 and consequently the strain may have accumulated secondary mutations that would decrease its fitness in vivo . To address this , the output clones of OND1954 isolated from three individual mice were collected and each clone was verified to be a conditional urease mutant . These clones were then pooled , OND3241 ( A-E ) , and used in subsequent infection studies to test if passage though mice led to improved infection rates . Mice were challenged with either the wild-type strain , the original conditional mutant OND1954 or the mouse passaged urease conditional mutant OND3241 , and supplemented without or with 5 mg/l Dox or ATc . Colonization of OND3241 remained dependent on inducer supplement and using ATc instead of Dox resulted in an improved infection rate and bacterial load in the infection model ( Fig 3 ) . Having established an infection model in which the H . pylori urease phenotype could be regulated in vivo , we proceeded to investigate what effect tet-mediated silencing of urease expression had on established H . pylori infections . Mice were challenged with the conditional strain OND3241 and provided with 5 mg/l ATc supplement to establish infection . After two weeks , the supplement was withdrawn and the animals were sacrificed at indicated time points . The conditional H . pylori urease mutant could still be isolated on days 1 and 3 after supplement withdrawal , however on days 5 and 7 the bacterial load had decreased to below our detection limit ( Fig 4A ) . This data demonstrated for the first time that continuous urease expression is required by H . pylori to maintain colonization even after the bacteria have become established in the gastric niche . To test if H . pylori were under selective pressure to overcome tet-regulation the suppression experiment was repeated and the animals were sacrificed at a much later time point . Mice were challenged with either wild-type X47 or OND3241 and provided with 5 mg/l ATc supplement to establish infection . After two weeks , the supplement was withdrawn from half the groups ( both OND3241 and wild-type ) while the remaining groups were maintained on ATc supplement and the animals were sacrificed at different time points ( Fig 4 ) . No differences in bacterial load or infection rate was observed for animals infected with wild-type X47 over the course of the experiment demonstrating that long-term ATc supplement ( 5 mg/l for 8 wks ) does not interfere with H . pylori infection ( Fig 4B ) . Animals challenged with OND3241 and maintained continuously on ATc supplement had a consistent infection rate of 60% ( Fig 4B ) . In vitro tests confirmed that bacteria re-isolated from these groups remained conditional urease mutants , even after a total infection time of 8 weeks . Withdrawal of ATc supplement from the animal groups challenged with OND3241 resulted in reduced infection load on days 3 and 5 and , although not completely cleared to below our detection limit in all animals , the infection rate had decreased to 20% on day 5 . However , when mice challenged with the OND3241 were left in the absence of ATc for 42 days , the bacterial load and the infection rate had increased resulting in 80% of the animals bearing bacteria in the stomach above our detection limit . Importantly , unlike the bacteria re-isolated at the earlier time points ( day 3 and day 5 ) , H . pylori re-isolated from this last group of mice were all urease positive and they were no longer conditional urease mutants as tested qualitatively in vitro . This result revealed that the strain was under selective pressure to restore urease expression . In an effort to identify possible genetic mutations to overcome tet-regulation of the urease operon , whole genome sequencing of the original conditional urease mutant strain OND1954 , the individual clones of input strain OND3241 ( A-E ) , and 34 output strains recovered at day 42 ( 5 conditional urease-negative strains and 29 urease-positive tet-escape mutants ) was undertaken . Sequence data were mapped against reference strain OND1954 and variants specific to tet-escape mutants were identified ( S1 Table ) . Sequence analysis of the ureAB locus , including the upstream regulatory region , revealed no changes between OND1954 , OND3241 and the tet-escape mutants . Interestingly , the tet-escape mutants harboured non-synonymous substitutions or frameshift mutations in at least one of the following genes , tetR and the flagellar biosynthesis genes fliA , fliE and flgE . These data reveal that tet-regulation was overcome in the tet-escape mutants not by altering the tetO binding sites but through affecting the repressor protein .
A set of conditional urease mutants were generated to demonstrate for the first time that urease expression is essential in the persistence stage of H . pylori infection , which broadens our understanding of the role of this enzyme during chronic infection . Genetic manipulation of PureA to place urease under tet-control led to a decrease in the basal levels of urease expression for all urePtetO constructs tested . However , under the conditions tested , strains transformed with urePtetOI , urePtetOII and urePtetOV were found to have comparable urease activity to that of wild-type . This data can be reconciled as it has been reported that under neutral in vitro growth conditions without added nickel , such as the growth conditions used in this study , a significant amount of urease in wild-type strains is in the inactive apoenzyme form and only a minor fraction of urease is active [38–40] . Urease activity in H . pylori is highly controlled and is modulated through several different mechanisms in response to various environmental cues [39] . Since our goal was to establish a working in vivo model , we decided to directly test if the decrease in urease expression could be tolerated by the bacteria by assessing if the urePtetO strains were capable of colonizing the murine stomach . Interestingly , when analyzing the in vivo colonization data from the X47 urePtetO strains ( OND2018—OND2022 ) a positive correlation between infection rate and in vitro urease expression and activity but not to bacterial load in colonized animals was observed ( Fig 1E ) . This finding suggests that for initial colonization of the murine stomach the amount of urease activity is an important factor likely due to the fact that the bacteria need to withstand the acidity of the gastric lumen until they reach their gastric niche , deep into the gastric mucus near the epithelial surface . However , once the bacteria are established within their environmental niche , although urease is still required for growth , the level of urease expression may be less important for maintaining colonization as mice colonized with strains transformed with urePtetOIII and urePtetOIV had a similar bacterial burden compared to mice infected with strains expressing more urease . Infection of the mouse host by the conditional urease mutant was strictly dependant on supplementation with a tetracycline inducer , confirming that genetic regulation of urease expression was stringent enough to prevent colonization . Furthermore , in the induced state , tet-mediated expression of urease was sufficient to allow and maintain infection by the conditional urease mutant . Withdrawal of the supplement resulted in clearance of the bacterium within 5 days . This is in line with the slow shut-off observed in other mouse models using tet-based regulation systems which has been attributed to the persistence of doxycycline in tissues [41 , 42] . Notably the longer time of clearance of the bacterium in vivo provided the opportunity for the emergence and selection of tet-escape mutants . H . pylori possess several mechanisms , such as an error prone PolA [43] and efficient DNA homologous recombination and transformation systems [44 , 45] , that permit the bacterium to undergo rapid microevolution to adapt to changing environments in its specific host [46 , 47] . The emergence of tet-escape mutants in this study suggests that there is strong selective pressure on the bacterium for continuous urease expression to maintain chronic infection . Whole genome sequence analysis of the tet-escape mutants identified several different mutations that likely explain how tet-regulation of the ureAB operon was overcome . One group of escape mutants had missense or nonsense mutation within the tetR gene . TetR is a finely tuned transcriptional regulator [31] and therefore most amino acid changes are likely to have deleterious effects to the function of TetR by inhibiting repressor dimerization or DNA binding [48–50] . Interestingly , another group of tet-escape mutants harbored either amino acid substitutions within the DNA binding domains of the alternative RNA polymerase sigma factor FliA ( σ28 ) or a truncated FliA due to nonsense mutation and frameshift alteration . FliA controls the transcription of some late flagellar genes ( class 3 ) including the flagellin subunit , flaA [51 , 52] . Previous studies have reported that H . pylori fliA mutants have no detectable flaA transcript and have truncated flagella [51] . In the conditional H . pylori urease mutants generated in this study , the transcription of tetR is driven by a flaA promoter . Therefore , it is reasonable to suggest that the fliA mutations identified in the tet-escape mutants likely affect the transcription of tetR from Pfla and consequently release the ureAB operon from tet-regulation . Additionally , some tet-escape mutants acquired non-synonymous mutations in fliE and flgE , genes that encode for components of the flagellum hook-basal body complex . Flagellar biosynthesis is a highly ordered and regulated process and transcription of late flagellar genes by fliA proceeds only once the hook-basal body complex is complete [53 , 54] . Mutation of fliE has been shown in other studies to affect the regulation of fliA dependent genes , leading to a two-fold reduction in flaA transcript levels [54] . Thus mutation of fliE and flgE may indirectly impact on the transcription of tetR from PflaA through the negative control of FliA [51 , 53] . Motility is essential for H . pylori colonization [19 , 55 , 56] , and therefore the isolation of tet-escape mutants with mutations in genes involved in flagellar biosynthesis ( fliA , fliE and flgE ) raises interesting questions . During the chronic stage of infection , once the bacteria are established in their environmental niche and have adhered to gastric epithelial cells [57] , can motility be compromised in preference for improved urease expression ? Further investigations with the appropriate conditional mutants are necessary to better understand the escape and in particular to test whether motility is still required once colonization is established . Another question is whether there are other mutations acquired by the tet-escape mutants that could compensate for mutations in fliA , fliE and flgE ? In-depth analysis of the whole genome sequencing data together with mutational studies needs to be undertaken to discern how these tet-escape mutants are able to survive in the host . The sequencing results underline the importance of urease for H . pylori to maintain persistence infection and reveal the high selective pressure for continuous expression of urease even after colonisation is successfully established . The mutations identified in the tet-escape mutants add support to the hypothesis that the genomic plasticity of H . pylori is an important mechanism for adaptation to new and changing environments [58] . Furthermore these findings also highlight the potential of using tet-based genetic tools together with whole genome sequencing to study H . pylori genetic plasticity and adaptation in a changing biological environment when the bacterium is put under duress . Using the conditional mutant , we have demonstrated that urease is essential for chronic infection and that repression of urease expression results in the loss of bacterial load within 5 to 7 days . The availability of these tools now allows for new questions to be asked regarding the reason behind the relatively rapid loss in colonization . One reason may be that the loss of urease activity results in the bacteria being more susceptible to clearance by phagocytic cells [24] . Another possibility is that the loss of urease activity negatively affects H . pylori ability to swim through gastric mucus , as the bacteria would no longer have the ability to decrease the viscosity of mucus through the elevation of local pH [59] , and consequently are cleared due to turnover of the mucus lining . The conditional urease mutant will serve as a valuable tool in further studies that pursue this line of investigation . In this study , H . pylori conditional urease mutants were generated by placing the expression of the urease subunits , UreA and UreB , under tet-control and have permitted the first direct testing of the hypothesis that urease is required by H . pylori for chronic infection . Furthermore , eventual escape from tet-regulated urease expression by H . pylori demonstrates that there is a very strong selective pressure on the bacterium to maintain urease expression during infection . Our data validates urease as a good target for therapeutic intervention . The conditional urease mutants generated here can also be used to gain more detailed insight into the role of urease in the persistence stage of infection including its interactions with MHC class II molecules [25] , induction of proinflammatory cytokine [27] and its potential role in motility [59] . Furthermore , this study demonstrates the need for conditional mutants , generated by using genetic tools such as the tet-system , to study H . pylori virulence factors , persistence and the bacterium’s influence on the host microbiota .
H . pylori X47 strains used in this study are listed in S2 Table . Bacteria were grown at 37°C under microaerobic conditions on Columbia blood agar ( CBA ) plates containing 5% horse blood and Dent’s antibiotic supplement ( Oxoid ) . When appropriate , antibiotic selection was carried out by supplementing media with chloramphenicol or streptomycin at a final concentration of 10 μg/ml . Microaerobic conditions were established in sealed jars using the Anoxomat MarkII system ( Mart Microbiology B . V . , the Netherlands ) after one atmosphere replacement with the following gas composition N2:H2:CO2 , 85:5:10 . All genetic manipulation of H . pylori strains was done using genomic insertion and replacement of a counter-selectable rpsL-cat cassette [60] . The use of the counterselectable streptomycin susceptibility ( rpsL-based ) system requires a host strain that possesses a streptomycin-resistant phenotype [61] . The H . pylori X47 host strain is naturally streptomycin-resistant and no modifications to this strain were required . The genotype of all mutants was confirmed by PCR and/or DNA sequencing . Oligonucleotides used in this study are listed in S3 Table . To place ureA and ureB under tet control , wild-type nucleotide sequences flanking the -35 and -10 promoter regions of the ureA promoter , PureA , were replaced with tetO sequences to generate five derivatives of PureA , urePtetO ( -I through -V ) ( Fig 1A and 1B ) . These promoter constructs were used to replace the native urease promoter , using the two-step rpsL-cat based transformation approach . A construct composed of the counterselection cassette flanked by DNAs homologous to regions of the ureA locus , ureA::rpsL-cat , was made by SOE PCR [62 , 63] ( S3 Fig ) and used to generate recipient strains in which PureA and ureA were replaced with rpsL-cat . Two 1 kb regions flanking PureA and ureA ( HP0073 ) , were amplified from 26695 genomic DNA using primers ureArcat1 and ureArcat2 , and ureArcat3 and ureArcat4 respectively . The rpsL-cat selection cassette was amplified using primers ureArcat5 and ureArcat6 . Nested primers ureArcat7 and ureArcat8 , were used to generate and amplify a final 3 . 4 kb PCR product , ureA::rpsL-cat . Natural transformation of the H . pylori strains with the ureA::rpsL-cat PCR construct was performed to obtain the recipient strain OND2017 . Transformants isolated on chloramphenicol plates were urease negative . Five tetO modified ureA promoter constructs urePtetO ( I-V ) , containing up to three tetO sites , were constructed by SOE PCR ( S4 Fig ) . The primer pairs used to make each urePtetO construct are listed in S4 Table . Briefly , a 1 kb fragment upstream , arm I , and a 1 . 5 kb fragment downstream , arm II , of PureA were amplified using 26695 genomic DNA as a template . Long primer tails were used to reconstruct the ureA promoter region upon fusion of arms I and II by SOE PCR . Primers ureArcat7 and ureArcat8 were used to amplify the final 2 . 5 kb products , urePtetO ( I-V ) , and sequencing confirmed that the modified ureA promoters were reconstructed correctly . Natural transformation of the recipient strain OND2017 with urePtetO PCR constructs resulted in replacement of the rpsL-cat with urePtetO and restoration of ureA , generating strains X47 urePtetOI through X47 urePtetOV ( OND2018—OND2022 ) . Correct allelic replacement of the resulting Strr transformants was confirmed by colony PCR using primers ureAP1 and ureArcat8 and by sequencing using primer urePseq . Conditional urease strains were generated by transforming the TetR expressing H . pylori strain , X47 mdaB::ptetR4 ( OND1987 ) [32] , with the ureA::rpsL-cat PCR construct to generate the recipient strain OND2026 . This urease negative strain was then transformed with each of the five urePtetO constructs to generate conditional urease mutant strains X47 mdaB::ptetR4; urePtetO ( I-V ) ( OND1954—OND1958 ) . Transformants were first screened for tetracycline dependent urease expression using the urease phenotype assay and additional characterization was done using the urease activity assay and immunoblot analysis . Urea culture plates ( Brucella broth , 7% NCS , 1 mM urea , phenol red 100 mg/l , vancomycin 6 mg/l , pH 6 ) were used to assay the urease phenotype of H . pylori clones . The pH of the media was adjusted with 1 M HCl before the addition of NCS and vancomycin . The pH was low enough to observe the colourimetric change of phenol red , from yellow to red , due to the catalytic activity of urease on urea , but not acidic enough to inhibit the growth of urease negative strains . To screen for tet-regulated urease activity , transformants and colonies re-isolated from infected animals were replica plated onto CBA plates with or without 50 ng/ml of ATc and cultured for 48 h . Bacteria were then patched onto urea plates and incubated under microaerobic conditions . Urea plates were examined after 16 h of incubation to identify clones that had switched urease phenotype upon exposure to ATc . Localized changes in colour around each growing colony identified urease positive clones . Conditional urease mutant strains grown on CBA plates without ATc remained urease negative , while strains grown on CBA plates with ATc became urease positive ( Example S5 Fig ) . The urease activity assay used in this study was adapted from the protocol previously described [28] . Strains were grown on CBA plates without or with 50 ng/ml ATc for two successive passages . Bacteria from 24 h plate cultures were collected and resuspended in cold buffer A ( 25 mM phosphate buffer , pH 6 . 8 ) and standardized to an OD600 = 4 . 0 . A 50 μl aliquot of the standardized bacterial suspension was then diluted with 50 μl of buffer B ( 25 mM phosphate buffer , pH 6 . 8 , 0 . 2% Tween-20 ) . A 25 μl aliquot of this diluted bacterial suspension was transferred into one well of a 96 well plate , diluted with 150 μl of buffer C ( 25 mM phosphate buffer , pH 6 . 8 , 250 μM phenol red ) and incubated for 5 min at 37°C . A 75 μl aliquot of urea solution ( 0 . 5 M ) was then added to the well and the absorbance at 560 nm was measured every 72 s for 75 cycles using a POLARstar Omega ( BGM Labtech ) plate reader . Activity was measured as the rate of change in absorbance over time and expressed as percent of urease activity of the wild-type X47 strain . All urease activity measurements were carried out in triplicate and experiments were repeated at least three times . Bacteria were grown in Heart Infusion ( HI ) medium supplemented with 10% Newborn Calf Serum ( NCS ) and vancomycin ( 6 μg/ml ) . Cultures were inoculated with bacteria suspended in PBS to give a starting OD600 = 0 . 05 , and grown under microaerobic conditions at 37°C and 120 rpm . For induction , H . pylori cultures were grown to mid-log phase in 10 ml of media . Cultures were induced with 200 ng/ml ATc and bacteria were incubated for another 12 h , with aliquots were taken at indicated time points . For gene silencing , conditional strains were cultured in the presence of 200 ng/ml ATc to mid-log phase . Fresh HI media , with or without 200 ng/ml ATc , was inoculated with pre-induced bacteria ( OD600 = 0 . 5 ) and grown for 12 h , with aliquots taken at indicated time points . Bacterial cells were collected by centrifugation and washed twice with PBS before processing for immunoblot analysis . Bacterial whole cell lysates were prepared as previously described [32] . The protein concentration of bacterial cell whole cell lysate samples was determined using the Micro BCA protein assay reagent kit ( Pierce ) with bovine serum albumin as the standard . Equal amounts of protein for each sample were mixed with 3x SDS-PAGE sample loading buffer , incubated at 95°C for 10 min , and proteins were separated by 10% SDS-PAGE and electrotransferred to a PVDF membrane . For detection of the UreB subunit of urease , mouse anti-UreB ( Austral biologicals ) was used at a dilution of 1:8000 . Secondary antibody rabbit anti-mouse-HRP ( Jackson ImmunoResearch Laboratories ) was used at a dilution of 1:10 , 000 and detection of the secondary HRP conjugate was accomplished by chemiluminescence ( Sigma ) using LAS 3000 ( Fujifilm ) ( software Image reader LAS 3000 V2 . 2 ) . For loading controls , duplicate gels were run in parallel and stained with Coomaise Brilliant Blue R-250 ( S1B Fig , S6 Fig and S7 Fig ) . Mouse procedures were reviewed and approved by the Institutional Animal Care and the Animal Ethics Committee of the University of Western Australia . 6–7 week old C57BL/6J female mice were challenged once by oral gavage with 200 μl of 1 x 109 CFU/ml of bacteria suspended in HI broth . Groups of infected mice received doxycycline ( Dox ) , anhydrotetracycline ( ATc ) or no supplement in drinking water containing 5% sucrose . Water was kept in light-protected bottles and changed every three days . Mice were sacrificed at indicated time points and stomachs were removed and homogenized in 1 ml HI using a tissue lyser ( Retch ) . Homogenates were serially diluted and plated out on H . pylori selective plates ( CBA containing 5% Horse blood , Dent , nalidixic acid 10 mg/l and Bacitracin 100 mg/l ) to determine the bacterial burden . Were appropriate , re-isolated clones were assayed for tet-responsive gene expression . Preparation of MiSeq library was performed using Illumina Nextera XT DNA sample preparation kit ( Illumina , San Diego , CA , USA ) as previously described with minor modifications [64] . In brief , 1 ng of genomic DNA was fragmented in 5 μl of Amplicon Tagment Mix and 10 μl of Tagment DNA buffer . Tagmentation reaction was performed by incubation at 55°C for 10 min followed by neutralisation with 5 μl of Neutralise Tagment Buffer for 5 min . Tagmented DNA ( 25 μl ) was indexed in a 50 μl limited-cycle PCR ( 12 cycles ) as outlined in the Nextera XT protocol and subsequently purified using 25 μl of AMPure XP beads ( Beckman Coulter Inc , Australia ) . The fragment size distribution of the purified DNA was analysed utilising a LabChip GXII 2100 Bioanalyser . DNA libraries were adjusted to 2 nM , pooled in equal volumes and then denatured with 0 . 2 N NaOH according to the Nextera protocol . The libraries were sequenced using 2 × 300 paired-end protocols on an Illumina MiSeq instrument ( MiSeq Reagent Kit v3 for 600 cycles ) . The draft genome sequence of H . pylori OND1954 has been deposited at DDBJ/ENA/GenBank under the accession MVFB00000000 . The version described in this paper is version MVFB01000000 . All raw sequence data generated in this study have been submitted to Sequence Read Archives ( SRA ) database with accession numbers listed in S5 Table . The generated MiSeq reads of H . pylori strain OND1954 was assembled using SPAdes genome assembler ( version 3 . 8 . 2 ) with careful option [65] . The draft genome sequence was subsequently annotated using Prokka ( version 1 . 11 ) with Swiss-Prot , Pfam ( release 30 . 0 ) , TIGRFAMs ( release 15 . 0 ) and Superfamily ( version 1 . 75 ) databases [66–70] . The annotation features are available in S1 File . The raw reads of OND1954-derivative strains were trimmed and mapped against the annotated draft genome using Bowtie2 on Geneious R7 platform [71 , 72] . Variants were called using the following parameters: minimum coverage = 10 and minimum variant frequency = 0 . 7 . For mouse colonization assays ( where n ≥ 5 ) the Mann-Whitney unpaired two-tailed test was used to compare colonization loads and the two sided Fisher’s exact test was used to compare infection rates . Bonferroni correction was used for multiple pairwise testing . Statistical analysis was performed using GraphPad Prism version 7 for Windows , ( GraphPad Software ) and Stata Statistical Software ( StataCorp . 2015 . Release 14 . College Station , TX: StataCorp LP . ) .
|
Helicobacter pylori is a bacterial pathogen that chronically infects half the global population and is a major contributor to the development of peptic ulcers and stomach cancer . H . pylori has evolved to survive in the stomach and one important adaptation is the enzyme urease . The bacteria cannot establish an infection in the host without this enzyme , and although widely postulated , the requirement of urease for chronic infection of the host has not been tested experimentally as conventional urease mutants are incapable of colonization . To overcome this constraint , a genetic system was introduced that allowed for the making of H . pylori strains in which urease expression could be turned off after the bacteria have colonised the stomach . We show for the first time that this enzyme is not only important for initial colonization but that it is also very important for maintaining chronic infection . We also show that if urease is turned off , the bacterium can mutate several different genes in order to restore urease expression . The genetic approach described here opens up opportunities to studying genes involved in the chronic stage of H . pylori infection to gain insight into how the bacterium is able to avoid clearance by the immune system and how it is able to adapt to changing biological environments .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"urea",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"ureases",
"enzymes",
"pathogens",
"drugs",
"enzymology",
"microbiology",
"organic",
"compounds",
"bacterial",
"diseases",
"helicobacter",
"tetracyclines",
"antibiotics",
"gastroenterology",
"and",
"hepatology",
"stomach",
"pharmacology",
"molecular",
"biology",
"techniques",
"bacteria",
"bacterial",
"pathogens",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"artificial",
"gene",
"amplification",
"and",
"extension",
"proteins",
"medical",
"microbiology",
"helicobacter",
"pylori",
"infection",
"microbial",
"pathogens",
"chemistry",
"gastrointestinal",
"infections",
"molecular",
"biology",
"gastrointestinal",
"tract",
"biochemistry",
"helicobacter",
"pylori",
"organic",
"chemistry",
"anatomy",
"polymerase",
"chain",
"reaction",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
] |
2017
|
Helicobacter pylori gene silencing in vivo demonstrates urease is essential for chronic infection
|
Fascioliasis ( or fasciolosis ) is a socioeconomically important parasitic disease caused by liver flukes of the genus Fasciola . Flukicide resistance has exposed the need for new drugs and/or a vaccine for liver fluke control . A rapidly improving ‘molecular toolbox’ for liver fluke encompasses quality genomic/transcriptomic datasets and an RNA interference platform that facilitates functional genomics approaches to drug/vaccine target validation . The exploitation of these resources is undermined by the absence of effective culture/maintenance systems that would support in vitro studies on juvenile fluke development/biology . Here we report markedly improved in vitro maintenance methods for Fasciola hepatica that achieved 65% survival of juvenile fluke after 6 months in standard cell culture medium supplemented with 50% chicken serum . We discovered that this long-term maintenance was dependent upon fluke growth , which was supported by increased proliferation of cells resembling the “neoblast” stem cells described in other flatworms . Growth led to dramatic morphological changes in juveniles , including the development of the digestive tract , reproductive organs and the tegument , towards more adult-like forms . The inhibition of DNA synthesis prevented neoblast-like cell proliferation and inhibited growth/development . Supporting our assertion that we have triggered the development of juveniles towards adult-like fluke , mass spectrometric analyses showed that growing fluke have an excretory/secretory protein profile that is distinct from that of newly-excysted juveniles and more closely resembles that of ex vivo immature and adult fluke . Further , in vitro maintained fluke displayed a transition in their movement from the probing behaviour associated with migrating stage worms to a slower wave-like motility seen in adults . Our ability to stimulate neoblast-like cell proliferation and growth in F . hepatica underpins the first simple platform for their long-term in vitro study , complementing the recent expansion in liver fluke resources and facilitating in vitro target validation studies of the developmental biology of liver fluke .
Fascioliasis ( or fasciolosis ) , a parasitic disease caused by liver flukes of the genus Fasciola , has significant economic and animal health impacts on the global agri-food industry . Global economic losses due to fascioliasis are estimated at around US$3 . 2 billion annually [1] , although a more recent study has identified impacts of up to US$4 . 78 billion in India alone [2] , while in the UK , fascioliasis costs the agri-food sector around £300 million [3] . Fasciola hepatica infection has become increasingly prevalent in humans , with up to 17 million people infected and 91 million at risk worldwide [4] , such that human fascioliasis is considered a Neglected Tropical Disease of major global and regional importance by the World Health Organization . Infection levels are set to rise further with increasing levels of resistance to triclabendazole ( the current drug of choice ) [5 , 6] and a potential explosion of fluke populations due to climate change [7] . Therefore , there is a pressing need to identify and evaluate novel diagnostic , therapeutic and preventative options for fascioliasis . Helminth parasitology has benefited from recent advances in transcriptomic , genomic and functional-genomic resources . These additions to the helminth ‘molecular toolbox’ have enhanced our ability to probe the fundamental biology of , and identify and validate therapeutic targets in , parasitic helminths . The F . hepatica toolset currently consists of a draft genome [8] , several developmentally-staged transcriptomes [9 , 10] and the ( RNA interference ( RNAi ) tools with which to functionally interrogate these datasets [11–14] . Similarly , advanced proteomics and sub-proteomic methods provide tools for advancing our understanding of fluke virulence and the host-parasite interface [15] . However , the effective use of these tools has been hindered by the absence of an effective in vitro maintenance system for F . hepatica juveniles and immature life stages ( the most pathogenic life stage of the fluke ) . In contrast , schistosome blood flukes can be maintained quite simply in vitro for many months in serum-supplemented culture medium , a method that has supported several functional genomic studies in Schistosoma mansoni [16] and S . japonicum [17] , while similarly simple methods have supported RNAi studies in Opisthorchis viverrini [18] and Clonorchis sinensis [19] . All of the reported F . hepatica culture methods are complex and of arguable utility , issues which have undoubtedly limited their adoption by the research community . Seminal studies from the 1960/70s reported successful in vitro maintenance of juvenile F . hepatica on a food source of cultured mammalian cell monolayers [20–22] , but there were notable discrepancies between these studies . However , in the late 1970s a comprehensive study by Davies and Smyth [23] evaluated 39 different combinations of basic media variously supplemented with animal sera and blood . The most effective combination in this study was NCTC 135 medium supplemented with 20% chick serum and ~0 . 01% sheep red blood cells , a combination which supported fluke growth , yielding development of reproductive tissue after 12 days in vitro . However , the authors stated that this outcome was only seen on one occasion and could not be repeated with different batches of serum . A subsequent study found that 50% human serum in RPMI medium with 2% human red blood cells promoted in vitro survival of up to 14 weeks , over which time the development of reproductive systems was observed , although no images were shown , and only limited quantitative data were presented to support these observations [24] . Similar inconsistencies exist regarding the most effective serum for inducing growth , with Davies and Smyth [23] favouring chick serum over human serum whilst Smith and Clegg [24] took the opposite stance . The latter study measured only the six fastest growing juveniles in each media tested [24] and did not consider the variation in growth capacity between individuals . These disparities and inconsistencies within and between studies have hindered the widespread uptake and refinement of in vitro maintenance methods for F . hepatica . As a result , experiments on F . hepatica juveniles and adults have tended to be performed over periods of hours or a few days . Consequently , studies of fluke gene expression and proteomics have been limited , with developmental studies only possible using non-contiguous systems , for example by comparing newly excysted juveniles ( NEJs ) grown in vitro with ex vivo liver-stage parasites and/or adults [25–27] . Here we have set out to develop a simple in vitro maintenance system for F . hepatica juveniles , using commonly available culture reagents . Our methods facilitate the survival of juvenile fluke in vitro for at least 6 months , during which time we observe rapid and consistent growth of newly-excysted juvenile fluke . These growing fluke exhibit the development of adult-like characteristics including reproductive structures , ultrastructural changes in the tegument towards an adult-like form , and an excretory/secretory ( E/S ) protein profile that is more reminiscent of immature liver or adult-stage fluke , than newly-excysted juveniles . We also demonstrate for the first time that F . hepatica growth and development is supported by the proliferation of neoblast-like stem cells . The methods that we describe permit long-term in vitro maintenance and development of F . hepatica juveniles to more mature forms and readily facilitate the study of developmental and temporal changes in biological parameters consistent with reduced animal use . These methods will support the exploitation of liver fluke poly-omics datasets in functional studies to interrogate gene function and seed the development of new drugs and vaccines . The work reported here also encourages the reduction of host animal use in studies of F . hepatica biology .
We employed two F . hepatica isolates in this study . All long-term maintenance experiments ( Figs 1–5 ) employed an Oregon strain ( Baldwin Aquatics , Oregon , USA ) . Cell proliferation experiments ( Figs 6 and 7 ) used an Italian strain ( Ridgeway Research ) . F . hepatica metacercariae were excysted as described by McVeigh et al . [13] . Prior to this excystment method , metacercariae of the Italian strain ( supplied on dialysis tubing ) were physically popped from their outer wall using a razor blade , before treatment in 10% bleach for 3–5 min . NEJs were held in RPMI before transfer to individual maintenance media as described below . NEJs were maintained in groups of 10 per well , in round bottomed 96 well plates ( Sarstedt ) in 250 μl relevant media . All experiments were handled using aseptic technique , and maintained in a humidified , 37°C incubator with 5% CO2 atmosphere . Media were changed three times per week , at which time the NEJs were also imaged and their survival rate recorded ( viability was assessed by the amount of movement and granulation that was seen in a juvenile; i . e . those juveniles not moving and heavily granulated were considered to be dead ) . Images were acquired using a Leica MZ125 microscope and attached Unibrain Fire-i digital camera , with measurements of the surface area of a random subset of worms ( i . e . those not obscured by other worms ) performed using measurement functions within ImageJ software ( http://imagej . nih . gov/ij/ ) , and calibrated against a 1 mm scale . We tested the following media: ( i ) Fasciola saline ( FS; Dulbecco’s modified Eagle’s medium ( Sigma-Aldrich ) , supplemented with 2 . 2 mM Ca[C2H3O2] ( Sigma-Aldrich ) , 2 . 7 mM MgSO4 ( Analab ) , 1 μM serotonin ( Sigma-Aldrich ) , 5 μg/ml gentamycin ( Sigma-Aldrich ) and 15 mM N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid ( Sigma-Aldrich ) ) ; ( ii ) RPMI 1640 ( ThermoFisher Scientific ) ; ( iii ) NCTC 135 ( ThermoFisher Scientific ) ; ( iv ) PBS ( 0 . 15 M NaCl , 0 . 03 M NaH2PO4H2O ( Sigma-Aldrich ) and 0 . 08 M Na2HPO4 ( Sigma-Aldrich ) pH 7 . 4 ) . Each medium was then supplemented with either foetal bovine serum ( FBS ) ( product #A15-151 , PAA—The Cell Culture Company ) , or chicken serum ( CS ) ( product #C5405 , Sigma-Aldrich ) , to varying proportions . All media included an antibiotic-antimycotic solution ( Sigma-Aldrich ) , to final concentrations of 100 U/ml penicillin , 0 . 1 mg/ml streptomycin and 0 . 25 μg/ml amphotericin ( the latter had no observable impact on worm growth or survival over a 24 h period ) . Fatty acids ( product F7050 , Sigma Aldrich ) , palmitic acid ( product P0500 , Sigma Aldrich ) and BSA ( product 05470 , Sigma Aldrich ) were tested at 0 . 25 ml/L , 10 μM and 20% w/vol , respectively . Prior to processing flukes for confocal microscopy , worms were paralysed by incubation in 7 . 14% MgCl2 for 1 min at RT . Juveniles cultured for up to 3 weeks post-excystment were free-fixed with rotation in 4% paraformaldehyde ( PFA: 4% PFA ( Sigma Aldrich ) in PBS , pH 7 . 4 ) for 4 h at room temperature ( RT , 21–25°C ) . Larger worms ( i . e . > 4 weeks after excystment ) were flat fixed in 4% paraformaldehyde between microscope slides for 2 h and then free-fixed as described above for a further 2 h , all at RT . Fixed worms were subsequently subjected to 3x 15 min washes in PBSTx ( PBS containing 0 . 5% Triton X-100 ( Sigma-Aldrich ) ) at RT with a final overnight wash at 4°C . Juveniles were then incubated in tetramethylrhodamine isothiocyanate ( TRITC ) -conjugated phalloidin ( Sigma-Aldrich; 200 ng/μl in Antibody Diluent ( AbD: PBS containing 0 . 1% bovine serum albumin ( Sigma-Aldrich ) and 0 . 1% Triton X-100 ( Sigma-Aldrich ) ) for 4 h in the dark at RT , before 3x 15 min washes in AbD at RT . Worms were mounted on slides in 8 μl of Vectashield ( Vector labs ) and viewed under a Leica AOBS SP5 confocal scanning laser microscope . Juveniles grown in 50% CS in RPMI were fixed in 4% glutaraldehyde ( Sigma-Aldrich ) , for 4 h at 4°C . Following this , juveniles were washed in 0 . 1 M sodium cacodylate buffer ( 0 . 1 M sodium cacodylate ( Sigma-Aldrich ) buffer ( pH 7 . 4 ) , containing 3% sucrose ( Sigma-Aldrich ) ) overnight ( ~16 h ) at 4°C . Juveniles were then stained in 1% OsO4 ( 90 min , 4°C ) , followed by three 15 min washes in H2O at RT . Juveniles were then washed twice in 70% ethanol for 30 min , twice in 90% ethanol for 20 min and twice in 100% ethanol for 5 min ( all at RT ) . Juveniles were covered with 200 μl hexamethyldisilazane ( Sigma-Aldrich ) and after 5 min this was removed and 200 μl of fresh hexamethyldisilazane was added and allowed to evaporate overnight ( ~16 h , RT ) . Juveniles were then transferred onto stubs and sputter coated for 5 min using a Polaron E5100 Series II before viewing under a FEI Quanta 200 scanning electron microscope . Image J software was used to measure the length of ten spines within the first three rings surrounding the oral suckers of 2–3 juveniles at each week of culture following excystment . Juveniles processed for TEM were fixed in 4% glutaraldehyde as described above , and then washed in sodium cacodylate buffer at 4°C ( 60 min ) . At this point juveniles were cut either longitudinally or transversely before fixation was continued for a further 11 h at 4°C . Juveniles were then washed in 0 . 1 M sodium cacodylate buffer containing 3% sucrose overnight ( ~16 h , 4°C ) and then stained in 1% OsO4 and processed through an ethanol series as described above . At this point juveniles were given two washes in propylene oxide ( Agar Scientific ) for 5 min at 4°C before embedding in resin ( 25 . 2% MNA , 25 . 2% DDSA , 49 . 6% agar resin and 1% DMP; Agar Scientific ) and propylene oxide in a 1:1 ratio . This was left overnight ( ~16 h ) at RT to allow excess propylene oxide to evaporate . Fresh resin was then placed on the samples and they were left for a further 24 h at RT . Juveniles embedded in resin were polymerised at 60°C for 48 h . Ultrathin sections , 60–70 nm in thickness , were cut on a Reichert Ultracut E ultramicrotome , mounted on bare 200-mesh copper grids , double-stained with alcoholic uranyl acetate ( 5 min ) and aqueous lead citrate ( 3 min ) and viewed in a FEI CM 100 TEM operating at an accelerating voltage of 100 kV . Visualisation of proliferating cells in F . hepatica juveniles was achieved by labelling nuclei undergoing DNA synthesis with 5-ethynyl-2-deoxyuridine ( EdU; ThermoFisher Scientific ) . We performed EdU labelling by incorporating 500 μM EdU in the worm maintenance medium ( RPMI 1640 +/- 50% CS , as indicated below ) . EdU exposure of juvenile fluke was performed for 24 h or 7 days , in either +CS or -CS media according to the pulse-chase exposure protocols as follows: ( i ) 24 h control: 24 h +EdU +CS; ( ii ) Protocol 1: 7 days +EdU +CS; ( iii ) Protocol 2: 7 days +EdU–CS; ( iv ) Protocol 3: 24 h +EdU +CS , 6 days–EdU +CS; ( v ) Protocol 4: 24 h +EdU +CS , 6 days–EdU–CS . Labelled worms were then processed for detection of EdU-labelled nuclei using the Click-iT EdU Alexa Fluor 488 Imaging Kit ( ThermoFisher Scientific ) , including labelling of nuclear DNA with Hoechst 3342 . Our only variation from the manufacturer’s labelling instructions was to fix the worms in 4% paraformaldehyde in PBS for 4 h at RT . To visualise the morphology of individual labelled cells , we adapted the cell dispersal method described by Collins et al . [28] . NEJs incubated overnight in 50% CS containing 500 μM EdU ( 37°C ) were dissociated by incubation in 3 . 5x Trypsin EDTA in RPMI 1640 , for 3 h , with occasional vigorous agitation through a 100 μl pipette tip . Isolated cells were strained by sequential passage through 100 μm and 40 μm cell strainers ( Corning ) , pelleted ( 200 g , 5 min ) , and resuspended in 4% PFA/PBS in which they were fixed for 30 min . This cell solution was then spotted onto Superfrost slides ( ThermoFisher Scientific ) and dried at 37°C for approx 15 min . EdU detection was then performed following manufacturer’s instructions , as described above . Samples were mounted on slides in Vectashield ( Vector Laboratories ) , and viewed on a Leica TCS SP5 confocal microscope . To assess the link between active cell division and EdU labelling , we incubated worms in 50% CS in RPMI containing 500 μM EdU and up to 30 mM hydroxyurea ( HU; stocks prepared in H2O ) for 7 days . In these experiments , HU-containing media were replaced daily because of the instability of HU in solution . In recovery experiments , we incubated juveniles in 50% CS in RPMI containing 500 μM EdU and 25 mM HU for 7 days , followed by removal of HU for a 3 day recovery period . EdU labelling was imaged on a Leica SP5 confocal microscope . Whole worms were imaged as maximally projected z-stacks , each generated from 15 optical sections gathered between dorsal and ventral surfaces . To quantify proliferation , we counted the EdU-labelled nuclei in our images using a cell counter plugin for ImageJ . Confocal figures were generated in GIMP ( www . gimp . org ) . Juveniles grown for 29 weeks in RPMI containing 20% , 50% or 100% CS were washed five times with 250 μl of fresh RPMI at 37°C . Live worms were selected and then collectively incubated ( 1 worm from 20% CS , 17 worms from 50% CS and 4 worms from 100% CS ) for 4 h in RPMI without CS . For proteomic analyses , supernatants containing each E/S sample were pooled and lyophilised before being reconstituted in 50 μL MilliQ H2O , with 5% of each sample analysed by 1DE ( 4–20% Mini-PROTEAN TGX Gel Biorad ) using SilverQuest Silver staining kit ( Invitrogen ) . The remaining samples were subjected to MS/MS analysis using an Orbitrap Elite 1410 Mass Spectrometer ( Thermo Scientific ) at the LaTrobe Institute for Molecular Sciences ( LIMS ) Proteomic Mass Spectrometry Facility . Proteins were identified using an in-house F . hepatica transcriptome database and by performing a BLASTp of the NCBI and GenBank databases to reveal associated accession numbers . All graphs were produced and statistical tests were carried out in GraphPad Prism 6 for Windows ( GraphPad Software , La Jolla California USA , www . graphpad . com ) . In cases where sample variances were equal ( determined by F test or Brown-Forsythe test ) parametric tests were used ( t test and ANOVA ) but when variances were not equal non parametric tests were used ( Mann Whitney U-test and Kruskal-Wallis ) . Post-hoc tests were used to compare means/medians of multiple groups and were chosen based on whether comparisons were needed between all groups or against a control group .
In parallel to our attempts to develop RNAi methods for Fasciola spp . [11 , 13] , we have also made efforts to improve in vitro maintenance methods to enable long-term study of phenotypic changes following RNAi or drug treatments . From initial experiments employing a DMEM-based “Fasciola saline” ( FS ) , performed over less than 24 h [11] , we subsequently reported that unsupplemented RPMI enabled in vitro study of viable juvenile fluke for up to 3 weeks [13] . Here , we tested both of these media as well as NCTC 135 , in the presence/absence of varying levels of foetal bovine serum ( FBS ) and chicken serum ( CS ) ( S1 Fig and S1 Table ) . Early experiments using un-supplemented base media illustrated that FS and NCTC were a much less effective maintenance media than RPMI ( 50% survival time: FS , 9 days; NCTC , 9 days; RPMI 26 days ) , leading us to disregard FS and NCTC from further experiments . Success of previous studies in using RPMI as a medium for schistosome maintenance [29] , alongside the significantly lower purchase cost of RPMI , led us to focus on RPMI in remaining experiments . Supplementation of RPMI with FBS did not improve worm survival , and in fact seemed to impede longevity ( 50% survival time: RPMI , 26 days; +5% FBS , 23 days; +10% FBS , 22 days; +20% FBS 19 days; +50% FBS , 14 days; 100% FBS , 15 days; S1 Fig ) . At this stage , we tested chicken serum ( CS ) , in line with a previous study [23] and found that RPMI supplemented with CS was an extremely effective promoter of fluke longevity . Fig 1 shows that 50% CS in RPMI supported 65% juvenile survival after 29 weeks ( at which time the experiment was purposely terminated ) . We found 50% to be the optimal concentration of those we tested , since 100% CS was less effective ( 50% survival at 4 weeks; 13% survival at 29 weeks ) , with dilutions below 50% CS displaying concentration-dependent impacts ( 50% survival: +5% CS , 4 weeks; +10% CS , 5 weeks; +20% CS , 8 weeks ) . Note that one juvenile in 20% CS in RPMI did survive to the end of the trial but was considered an outlier in terms of both survival and growth ( see below ) . All of these parameters represent significant improvements on fluke survival in RPMI±FBS , and to our knowledge represent the longest reported periods of F . hepatica maintenance in vitro . In additional experiments not detailed here , we have since achieved maintenance/growth of fluke for 13 months using these methods . Finally , we tested whether supplementation of RPMI with a fatty acid mixture , palmitic acid or bovine serum albumin would stimulate fluke growth but no effect was observed relative to the growth observed with RPMI+50% CS ( S2 Fig ) . In addition to the improved survival imparted by CS supplemented RPMI , we also noted more rapid growth of fluke compared to those maintained in FBS . Fig 1B illustrates that even after 1 week juveniles maintained in RPMI containing either 50% or 100% CS were significantly larger than those in any other media ( mean Week 1 area: RPMI , 0 . 02 μm2; 50% CS in RPMI , 0 . 04 μm2; 100% CS , 0 . 05 μm2 ) ( Week 1 Kruskal-Wallis: H = 143 . 9 , 241 d . f . , P < 0 . 0001; S2B Table ) . By 29 weeks post-excystment , juveniles maintained in 50% CS in RPMI had grown to 38 . 5x their original size ( worm area: week 0: 0 . 02 mm2; week 29: 0 . 77 mm2 ) , although there was no significant growth after 20 weeks . Notably , we observed a considerable increase in the variability of worm sizes at later time points ( distance between 95% confidence intervals: 0 weeks , 0 . 00137 μm2; 10 weeks , 0 . 1328 μm2; 20 weeks , 0 . 4381 μm2; 29 weeks , 0 . 5519 μm2 ) , suggesting differences in the growth capacities of individual worms . In growing worms , we used confocal microscopy to observe increased gut complexity and the development of reproductive tissue , both of which are indicative of transit towards more adult-like characteristics . NEJs and early ( up to a week post excystment ) in vitro cultured juvenile fluke exhibited simple digestive caeca and no reproductive structures . From two weeks onwards , both branched digestive caecae and uterine tubing were visible in the region posterior to the acetabulum ( Fig 2A and 2B ) . By 2 to 3 weeks post-excystment all juveniles exhibited primary branching of digestive caeca ( Fig 2A and 2D ) , with 66% of fluke developing uterine tubing posterior to the ventral sucker ( Fig 2B , 2D and 2E ) . We consider this tubing uterine in nature because it is consistent with the location of the uterus in mature F . hepatica [30] , and is therefore indicative of developing female reproductive structures . At 4 weeks post excystment , secondary gut branches had appeared in 55% of fluke ( Fig 2A and 2F ) . After 29 weeks maintenance in vitro , all surviving juveniles had undergone considerable growth ( Fig 1B ) with all individuals exhibiting secondary branching of the digestive caeca ( Fig 2A ) and uterine tubing posterior to the acetabulum ( Fig 2B ) . Due to the larger size of these worms , we were able to view the latter structures in greater detail , observing an extended uterus ( Fig 2G ) . An extreme example of the variable growth capacities of individual worms was exhibited by a single fluke , maintained in 20% chicken serum for 29 weeks . This fluke grew larger than any other in this study . Although an outlier in terms of growth rate , it was nonetheless instructive in terms of the potential for in vitro development in this system . Specifically , this individual grew to 239x its original size , reaching 4 . 3 μm2 in area and 3 mm in length ( Fig 3A ) , whilst juveniles in 50% CS in RPMI reached ~38 . 5x their original size ( 0 . 77 μm2 ) with a mean length of 1 . 45 mm . This fluke had distinctive external morphology more representative of an adult than a juvenile fluke , characterised by: ( i ) a more anterior position for the acetabulum relative to the central positioning seen in juveniles ( Fig 2C ) ; ( ii ) development of an oral cone and ‘shoulders’ , with a leaf-like shape , distinct from the more vermiform juvenile; and , ( iii ) the loss of the probing movement displayed by migrating stage fluke and the adoption of a wave–like movement characteristic of adults ( S1 , S2 Movies ) . This specimen had the most well developed gut of any observed in this study , with distinct tertiary branches of the caecae ( Fig 3B ) . In addition to considerably longer and more elaborate uterine tubing , this worm also displayed a putative ootype ( Fig 3B ) . These structures were connected to the gonopore via a gonoduct , tracking dorsally to the acetabulum ( Fig 3B and S3 Fig ) . We did not observe testes per se , but this specimen was unique amongst our samples in exhibiting tubing putatively associated with male gonads in the anterior region where testes exist in adult worms ( Fig 3B ) . In addition to the gross internal developmental changes described above , our maintenance method also triggered changes to the ultrastructure of the inner and outer tegument that signify development of juvenile fluke towards more immature- and adult-like forms . We detected these changes using both SEM and TEM . SEM observation of the fluke surface revealed that tegumental spines grew in length significantly over the course of the study ( mean spine length around oral cone: 0 weeks , 0 . 52 μm; 2 weeks , 1 . 37 μm; 4 weeks , 2 . 26 μm; 29 weeks , 3 . 28 μm ) ( Fig 4A; Kruskal-Wallis: H = 134 , 157 d . f . , P < 0 . 0001; S3 Table ) . Tegumental spines also changed in morphology during development; Fig 4B illustrates the progression from the short , stunted spines of the NEJ , towards the more distinct , regular arrangement of spines as rings circling the fluke visible at 2 weeks post-excystment ( Fig 4C ) . Similarly , NEJs lacked spines on the posterior ventral surface , developing spines in this region by 2 weeks post excystment . This occurred at an even slower rate on the posterior dorsal surface , where spines were visible only after 29 weeks post excystment ( S4 Fig ) . By 4 weeks post excystment spines began to develop multiple tips ( Fig 4D ) . This process was first visible in spines on the ventral anterior surface , which at this point had two tips ( Fig 4D ) . By 29 weeks post excystment we observed spines with two , three or four tips ( Fig 4E ) compared to the 8-tipped spines seen in ex vivo adult fluke . The development of cultured juveniles in our maintenance medium also triggered changes in the internal ultrastructure of the fluke tegument , most evident from the increase in depth of the tegument syncytial layer to ~1800 nm ( 29 weeks ) ( Fig 5A; Kruskal-Wallis: H = 38 . 07 , 45 d . f . , P < 0 . 0001; S4A Table ) , which is comparable to the tegument depth observed in immature fluke recovered from the liver of mice 1–2 weeks post-infection [31] . In addition , a significant increase in the depth of surface invaginations ( tegumental in-folds ) was observed over the weeks following excystment ( mean invagination length: 0 weeks , 218 . 1 nm; 2 weeks , 308 . 4 nm; 4 weeks , 1053 nm; 29 weeks , 2146 nm ) ( Fig 5B; Kruskal Wallis: H = 46 . 82 , 84 d . f . , P < 0 . 0001; S4B Table ) . Beyond these gross changes , we also observed differences in the organellar composition of the tegumental syncytia over time . Fig 5C shows that the NEJ tegument is barely discernible from underlying tissues . However , by 1 week post-excystment there had been a rapid change in the tegument ultrastructure , with the development of a clear syncytium and the appearance of tegument-specific secretory vesicles: T0 ( large , spherical electron dense structures ) and T2 bodies ( biconcave discoid structures with electron lucent contents ) ( Fig 5D ) . This process remained visible at 2 weeks post-excystment but with the additional presence of T1-like bodies ( i . e . vesicles containing both electron-dense and–light regions; Fig 5E and 5F ) which are known to appear in immature flukes in the liver [31] . By 4 weeks post-excystment the tegument appeared devoid of T0 bodies , with the T1-like bodies having developed into mature T1 bodies with their typical ‘cartwheel’ appearance ( Fig 5G ) . By 29 weeks post-excystment , T1 bodies were tightly packed in the syncytium with evidence of T2 bodies also present ( Fig 5H ) . To visualise patterns of cell division associated with fluke growth and development , we labelled worms with EdU , enabling the detection of cells undergoing active DNA synthesis . Fig 6A and 6B show that EdU-labelled ( EdU+ ) nuclei accumulated at different rates in growing ( +CS ) and non-growing ( -CS ) worms when incubated continuously in EdU for 7 days . EdU accumulation over this period was higher in growing than non-growing fluke ( -CS +EdU # nuclei 17±1 , n = 8; +CS +EdU # nuclei 88±6 , n = 13; p<0 . 0001; Fig 6C ) , suggesting that proliferating EdU+ cells directly contribute to fluke growth by increasing the total cell count . The spatial pattern of EdU+ accumulation in growing worms is notable: in non-growing worms , EdU+ nuclei were distributed throughout the parenchyma , but appeared absent from the anterior 1/3 of the worm ( Fig 6B ) . In growing 7 day old fluke , EdU+ nuclei accumulated in a distinct pattern of three clusters of nuclei in ( i ) the oral sucker , ( ii ) the anterior-posterior midline and ( iii ) the lateral margins ( Fig 6A ) . Only in +CS worms did we observe EdU+ nuclei in the anterior 1/3 of the fluke . The appearance in growing worms of EdU+ cells in locations not labelled in non-growing worms , suggests expansion and movement of these cells into new areas , and probably their differentiation in those areas . Taken alongside the increased expression of typical neoblast marker genes ( argonaut and nanos ) in growing compared to non-growing worms ( S5 Fig ) [32] , we consider that the EdU+ cells of F . hepatica juveniles resemble flatworm neoblasts . To examine the morphology of EdU+ cells , we performed tryptic digests of whole juveniles labelled in vitro for 18h ( +CS+EdU ) , with EdU detection and confocal analysis of dispersed cells fixed onto microscope slides . All of the EdU+ cells we detected displayed the gross morphology characteristic of neoblasts as described in other flatworms—relatively small , rounded cells with a prominent nucleolus , and scant cytoplasm surrounding the nucleus ( Fig 6D and 6E ) . Some EdU- cells also had a similar appearance , but most of these had distinctive morphologies suggestive of differentiation ( Fig 6F–6H ) . In non-growing worms , while EdU+ nuclei are visible , accumulation of EdU labelling is not significantly different between worms incubated in EdU for 24h or 7 days ( 24h EdU+ nuclei 16±3 , n = 3; 7 day EdU+ nuclei 17±1 , n = 8; Fig 6B and 6M ) ; neither does the localisation of EdU+ nuclei change between these time points in non-growing samples ( Fig 6B and 6M ) . This suggests that in addition to the population of proliferating cells that contribute directly to worm growth , presumably through expansion and differentiation , there is a sub-population of EdU+ cells that proliferate , possibly for maintenance or self-renewal , but which progress towards differentiation only upon exposure to an appropriate developmental signal . Following inhibition of DNA synthesis and neoblast proliferation by exposure to HU ( 7 days; see below ) , we tested the self-renewal potential of neoblasts after the removal of HU by following EdU re-accumulation in worms over a 3 day recovery period . Although we observed the significant recovery of growth in these worms , we did not detect expansion of the few remaining neoblasts following this recovery period . It is possible that the 25 mM HU that we employed triggered cell death in our neoblast population [33]; future work should titrate HU to the lowest possible concentration for use in renewal experiments . Neoblasts are presumed pluripotent when they originate , only subsequently differentiating into mature cell types within specific tissue or organ systems ( although there is evidence for several sub-populations with potentially distinct fates amongst planarian neoblasts [34] ) . This developmental process requires neoblasts to migrate from their origins in the parenchyma to their mature site of residence in a differentiated tissue . To investigate whether F . hepatica EdU+ cells behaved similarly , we performed pulse-chase experiments with the aim of determining the spatial fate of EdU+ cells labelled in a 24h pulse , following a 6 day chase period ( protocols described in Fig 6I ) . Both growing and non-growing pulse-chase worms displayed a spatial shift in EdU+ localisation compared to 24h controls ( Fig 6J , 6K , 6N and 6O ) , including movement into previously unlabelled areas , consistent with previous reports of flatworm neoblasts [35 , 36] . To further examine the link between neoblasts and worm growth , we examined the impact of inhibiting DNA synthesis ( using hydroxyurea , HU ) on neoblast proliferation and worm growth rate . Fig 7A illustrates that HU has a concentration-dependent impact on growth of the juvenile flukes over a 7 day observation period , with statistically significant inhibition of growth vs untreated controls at ≥10 mM ( worm area: 0 mM HU , 0 . 045±0 . 002 mm2 , n = 24; 25 mM HU , 0 . 03±0 . 001 mm2 , n = 26 , p<0 . 0001 ) . HU simultaneously inhibits EdU+ proliferation , such that 10- fold fewer EdU+ nuclei are visible in worms treated with 25 mM HU than in untreated controls over a 7 day incubation ( Fig 7D , 7E and 7F; Mean±SEM # EdU+ nuclei/worm: 0 mM HU , 41±3 . 9; 25 mM HU , 5 . 8±0 . 7; Mann Whitney U test , n = 21 p<0 . 0001 ) . Upon removal of HU , worm growth rate increases ( Fig 7B ) with a significant increase in size over the following 3 days ( Fig 7C; Mean±SEM worm area: Day 7 , 0 . 026±0 . 0005; Day 10 , 0 . 031±0 . 001; Mann Whitney U test , n = 97 , p<0 . 001 ) . However , this significant increase in growth does not correlate with a significant increase in the number of EdU-labelled nuclei ( Fig 7D and 7G ) . The correlation between worm growth and neoblast proliferation is consistent with neoblasts representing the major source of cellular proliferation in support of worm growth and development . We performed MS/MS analysis of E/S samples from juveniles maintained for 29 weeks in either 20% CS in RPMI ( n = 1 fluke ) , 50% CS in RPMI ( n = 17 fluke ) or 100% CS ( n = 4 fluke ) . Our aim was to compare the protein content of E/S gathered from in vitro maintained juveniles with the previously described distinct E/S profiles of NEJ , immature and adult F . hepatica . We hypothesised that this would provide proteomic evidence of fluke development . We identified peptides from 61 different F . hepatica proteins , with distinct profiles in each of our juvenile groups: 17 proteins in the juvenile grown in 20% CS in RPMI , 52 proteins in those juveniles grown in 50% CS in RPMI and 27 proteins identified in the juveniles from 100% CS . Within these juvenile profiles we discovered ten proteins known to be solely expressed in the E/S of immature and/or adult fluke , including cathepsin L 1A , L2 and L5 , cathepsin B2 and B7 , Legumain 1 and 5 , a prolyl-carboxypeptidase , Thioredoxin H type 1 and GST sigma 1 ( Table 1; [25 , 27] ) . Since our in vitro juvenile E/S profiles more closely resemble the known profile of immature/adult worms than NEJs , we consider that our in vitro maintenance system enables study of the temporal development of E/S output in F . hepatica .
While recent progress has been made in the development of sequence datasets and functional genomics tools for liver fluke , in vitro maintenance methods to support the application of these resources for the study of F . hepatica biology are lacking . In an attempt to address this we have developed a simple method for the in vitro maintenance of F . hepatica juveniles . Using these methods to study growth and developmental processes , we show here that: ( i ) our in vitro system recapitulates aspects of known morphological and behavioural indicators of F . hepatica development towards adulthood; ( ii ) F . hepatica growth and development is likely supported by the proliferation of somatic neoblast-like cells; and , ( iii ) the proteomic profile of E/S material released by the cultured flukes changes to a more immature/adult fluke–like profile providing evidence that significant fluke development is occurring . This work originated from our attempts to improve the survival of juvenile fluke during RNAi experiments , where our aim was to maintain fluke for a period long enough to detect phenotypic changes following transcriptional silencing . The primary finding here was that although worms would grow and survive for up to 3 weeks in FBS , the survival and growth rate of juvenile worms was vastly improved by substituting chicken serum ( CS ) for FBS . CS has previously been used in the cultivation of Maritrema novaezealandensis , a trematode parasite of Red-Billed Gulls in New Zealand [37] as well as for the culture of Philophthalmus sp . and Gynaecotyla adunca which both have avian definitive hosts [38 , 39] , so its use in helminth culture media is not without precedent . Why CS represents such an improvement on FBS , a product of F . hepatica’s definitive host , remains unclear . In addition to likely differences in the complements of growth hormones/promoters between the two sera that would impact on their relative capacities to stimulate growth , there exists a previously reported compositional difference , where CS is known to contain higher levels of triglycerides and cholesterols [40] than FBS . This is an intriguing factor given that fluke are thought to lack the capacity to synthesise fatty acids de novo [41] so it is possible that the higher lipid content of CS may reflect its greater capacity to support growth . Attempting to improve upon earlier studies in this area [23 , 24] we quantified 2 dimensional area ( rather than length ) as a measure of worm growth , reasoning that this measure would be less susceptible to fluctuation caused by the considerable changes in length that occur during normal motility in juvenile fluke . If we do compare lengths between juveniles in this study and previous work , the largest juvenile in our trials reached 3 mm in length , similar to the 3 . 1 mm fluke obtained after 14 weeks fluke culture by Smith and Clegg ( 1981 ) using 50% human sera in RPMI with 2% human red blood cells . However , in that study , survival was less than 50% at 14 weeks and only 96 of the 240 total juveniles were measured ( specifically , the six fastest growing juveniles in each group ) [24] . Additionally , these authors reported that seven flukes grew to 6–7 mm after 14 weeks . The data presented in our study encompass the mean growth of all fluke ( across two independent trials; note that in two other trials in which fluke growth was not monitored throughout , worms had reached sizes similar to those reported here after 6 months ) , thereby representing a more complete overview of the growth platform . More significantly , fluke survival was greater in our trials with 65% survival after 29 weeks . Growing juveniles exhibited considerable variation in growth capacity , as exhibited by the range of sizes seen , particularly during the latter stages of the study . These differences may be due to their relative abilities to respond to chemical growth promoting signals in medium , to extract adequate nutrition from the medium , or to physical differences in their cellular capacities for growth and differentiation . Regardless , it is worth noting that the flukes used here in long-term maintenance experiments originated from a wild type population of metacercariae obtained from the USA , the genetic diversity of which may be responsible for the range of growth phenotypes reported . Comparison to in vivo juveniles , which in mice grow to an average of 1 mm in length after 8–9 days [30] and 3 mm after 15–20 days in the liver , shows that growth clearly occurs more slowly in our in vitro system than is the case in vivo . These comparisons suggest that our in vitro fluke maintenance system facilitates the development of fluke comparable to immature liver stage in vivo parasites . To determine whether morphological measures support this conclusion , we employed confocal and electron microscopy to examine changes to gut , reproductive and tegument tissues . Interestingly , the in vitro development of caecal branching in the juveniles corresponded with that of size-matched in vivo juveniles; in vitro juveniles of ~0 . 5 mm in length ( 3 weeks growth post-excystment ) , 1 . 45 mm ( 29 weeks growth ) and 3 mm ( 29 weeks growth , 20% CS outlier ) , respectively , exhibited primary , secondary and tertiary caecal branching , as did their size-matched in vivo relatives [30] . These observations indicate that development of the digestive tract is linked to worm size , rather than to developmental age . A similar observation was made regarding the development of the reproductive system . In vitro juveniles of ~0 . 5 mm in length ( 3 weeks post-excystment ) exhibited some development of uterine tubing , of a similar extent seen in size matched ( 1 week post infection ) in vivo parasites [30] . However , where in vivo parasites possessed a visible gonoduct and testes at this stage , our in vitro juveniles did not . By 29 weeks , our 1 . 45 mm in vitro juveniles had fallen considerably behind the development of 1 . 5 mm in vivo parasites , which have a well-developed gonoduct and significant oviduct , ovary and testes development [30] . None of these features were visible in our 1 . 45 mm in vitro juveniles; even though our 3 mm outlier fluke exhibited much more reproductive tissue development than any other fluke in this study , gonads were not detected . This individual displayed extensive uterine development , including the complex folds noted for this structure in vivo [30] . There was also development of the gonoduct , anterior to the ventral sucker , although a gonopore was not visible . This specimen also possessed tube-like structures in the region of the testes , potentially representing the vas deferens leading from the testes to the ootype . Although our in vitro fluke can be considered stunted versions of comparable in vivo parasites , our juveniles do compare favourably with the development achieved in vitro in previous studies , where uterine development was reported in 6–7 mm juveniles , albeit in the absence of ovarian development or evidence of eggs [24] . We did not observe development of a cirrus , consistent with previous reports [23 , 24] . Previous in vitro culture studies did not examine tegumental development . Our SEM analyses revealed that development of the tegumental surface under our in vitro assay system largely resembles that seen in vivo . In contrast to the rather poorly developed spines of NEJs [42] , we observed growth and the increased distribution of spines across the surface of fluke growing in vitro . After one week of growth in vitro spines were present across the whole ventral surface and showed a significant increase in length as well as an increased density , especially on the posterior dorsal surface . This growth-related increase in spine density has been observed previously in fluke recovered from murine hosts [43] . During the subsequent three weeks of growth in vitro we observed no further increases in tegumental spine distribution although , after 29 weeks , spines were found across essentially the entirety of the fluke surface , including the dorsal , posterior surface from which they had been absent at earlier time points . We also noted morphological changes to the spines , characterised by the development from single-tipped to multi-tipped spines in later stages . We first noted multi-tipped spines at 4 weeks post-excystment , firstly on the anterior ventral spines where secondary tips began to appear . This change occurs at around 2 weeks post infection in vivo , although after 3 weeks in vivo spines had as many as 8-points [43] . Even after 29 weeks in vitro , our juveniles in 50% CS in RPMI had spines bearing a maximum of 3–4 tips , again suggesting that although our system triggers in vivo-like developmental changes , these changes occur at a slower rate in vitro than in vivo . The subsurface ultrastructure of the tegument also showed changes in line with development in our growing fluke . Developmental changes to the composition and diversity of various organelles and structures in the tegumental syncytium have been described in some detail in F . hepatica [31] . One of the most well studied aspects of this process is the dynamics of the secretory vesicles ( termed T0 , T1 , T2 bodies [31] ) contributing to tegumental turnover and renewal . In this process , T0 bodies are replaced by T1 bodies as the fluke develops towards adulthood , while there is a concurrent increase in the number of T2 bodies [31] . These bodies are thought to play a key role in tegument development and defence from the host immune system as the juvenile fluke migrates through the liver parenchyma . TEM studies of our in vitro fluke revealed that T0 to T1 transformation occurs in vitro over a similar timescale to that seen in vivo . Juveniles cultured in vitro in our trials exhibited T0 and T2 bodies after 1-week post-excystment with T1 bodies appearing after 2 weeks . T1 bodies began to outnumber the T0 bodies after 3 weeks of growth , with only T1 and T2 bodies visible by four weeks post-excystment . In vivo , the process occurs only slightly faster , with T1 bodies first appearing after 5 days of infection in mice and representing the most numerous tegumental body after 2 weeks when the fluke are in the liver parenchyma . This suggests that tegument development is timed from excystment , rather than being related to fluke growth/size . The appearance of T2 bodies provides further evidence as to the development of these juveniles as it again suggests that the ultrastructural development of these bodies is progressing in a manner comparable to that seen in vivo . T2 bodies appear in vivo after two days of infection in mice ( in syncytial cell bodies with only a few at the surface of the tegument syncytium ) but only predominate in mature , adult fluke [31] . To investigate the cellular basis of the growth/development phenotypes described here , we used confocal microscopy to examine the nature of proliferating cells in juvenile F . hepatica . Given that neoblasts ( pluripotent stem cells ) appear to be the only proliferating somatic cells in other flatworms [32 , 35 , 44 , 45] , our hypothesis was that similar progenitor cells would also drive growth and development in F . hepatica . We tested this hypothesis using a commercially available EdU labelling kit to examine the location , morphology and behaviour of proliferating cells , and here provide the first description of putative neoblasts in F . hepatica , showing that EdU+ cells in juvenile F . hepatica: ( i ) have neoblast-like morphology; ( ii ) display migratory behaviour consistent with flatworm neoblasts; ( iii ) support worm growth . By detecting incorporation of EdU , a thymidine analogue , into newly synthesised DNA , we identified strong labelling of cellular nuclei ( co-stained with Hoechst 3342 ) scattered through the posterior two thirds of juvenile fluke . F . hepatica EdU+ cells arise within the parenchyma below the tegument and muscle layers , and posterior to the cerebral ganglia . These cells show the typical morphology of neoblasts described in other flatworms ( round cell , large nucleus , prominent nucleolus , scant cytoplasm ) . To explicitly link the presence of neoblasts with fluke growth , we first examined the correlation between neoblast numbers and worm growth . Fig 7 shows that in worms maintained in 50% CS in RPMI , alongside worms in unsupplemented RPMI , in the presence of 500 μM EdU for 7 days we saw growth only in 50% CS worms , while the accumulation of EdU+ nuclei ( indicating the presence of cell division ) , was much greater in RPMI plus CS than in RPMI-only worms . In fact , the numbers of EdU+ nuclei did not increase in our non-growing worms during 7 days incubation in RPMI . These data suggest that enhanced neoblast proliferation in CS-containing media contributes to worm growth . To test this link further we used HU ( an inhibitor of ribonucleotide reductase , the enzyme that catalyses production of DNA from RNA ) to inhibit DNA synthesis in , and subsequent mitosis of , EdU+ cells . At concentrations of 10 mM or greater , HU inhibited worm growth and lowered neoblast abundance in CS-maintained worms . This suggests that neoblast proliferation provides the increased cellular mass that drives worm growth . Using pulse-chase microscopy , we were able to confirm that the localisation pattern of fluke neoblasts changes over time , suggesting that they migrate away from their site of origin . Following a pulse/chase regime of 1 day pulse , 6 days chase , we detected a distinct shift in spatial localisation of EdU+ labelling in growing worms . Fig 6 illustrates this shift where EdU+ nuclei in “pulsed” worms were seen across the core and lateral margins , but only within the posterior two thirds of the worms . This pattern changed in “chased” worms where EdU+ nuclei were concentrated around the flanks , including within the anterior one third of the flukes . We hypothesise that these may represent differentiated tegumental nuclei as described in S . mansoni [28] . These data represent the first description of neoblast-like cells in F . hepatica , and will support further investigations into the molecular genetics and functional genomics of these pluripotent cells . Such studies will provide further biological insight into the intriguing stem cell system of parasitic flatworms , and may highlight developmental targets for future control interventions . Lastly , proteomic analysis of E/S proteins provides further insight into the maturation of the in vitro cultured fluke . The presence of the cathepsins CL1A and CL2 in all three E/S samples , and CL5 in two of the E/S samples , clearly indicates that the in vitro cultured fluke have developed significantly from the juvenile stage with the supplementation of 20–100% CS . CL1 , CL2 , and CL5 are the main proteases known to be produced by adult F . hepatica [46 , 47] . Other proteins known to be produced by the immature and/or adult flukes were also identified , such as CL3 , CB2 , putative cathepsin B7 , Legumain 1 , putative prolylcarboxypeptidase , Thioredoxin H-type 1 and GST-Sigma 1 [25 , 27] . The E/S protein profile of flukes grown in 50% CS in RPMI exhibited the closest similarity to the known profile of immature/adult E/S . The protein profiles observed indicate that the flukes had not fully switched from the juvenile stage as two proteins that are markers for juvenile E/S ( CB3 and putative legumain 2 ) were observed in the in vitro developed worms . It is possible that the variation in E/S profile observed in our in vitro fluke , relative to that observed from flukes recovered from animals , may be due to the lack of host cues required for complete maturation . Further careful manipulation of the culture conditions may allow the production of flukes that better mimic those recovered in vivo . In summary , this study describes a set of simple methods enabling long-term in vitro maintenance of F . hepatica juveniles , which also permit constant monitoring of development , survival , growth and other phenotypic measures in maintained fluke . We have used this system to: ( i ) profile developmental changes in several fluke tissues that resemble those processes reported in in vivo parasites; and , ( ii ) provide the first description of neoblast like putative stem cells in F . hepatica , implicating these cells as essential for fluke growth and development . These methods will therefore support the development of in vitro assays for flukicidal drug and vaccine target validation screens , including the use of functional genomics tools such as RNAi [11–14] . Given that our in vitro methods recapitulate several aspects of fluke development in vivo , albeit at a slower rate , our system also has clear potential to reduce animal use that is currently unavoidable for the production of late juvenile or early adult parasites .
|
Parasitic worms require a host organism in order to survive and reproduce . As such , it is difficult to study them outside of a host . Some parasites can be maintained in vitro using cell culture methods; in the case of F . hepatica , previously-reported methods are unsatisfactory because they are difficult to reproduce and unable to support long term growth and development . Here we have developed a new set of methods for maintaining F . hepatica juveniles in vitro . These methods use simple , commonly available reagents and techniques , enabling us to keep fluke alive in vitro for at least 6 months , as well as stimulating the development of characteristics that resemble adult parasites . Over time , our in vitro fluke show changes in the structure and complexity of individual tissues , and the proteins they produce , such that they are more reminiscent of adult , than juvenile fluke . Additionally , we demonstrate that fluke growth is supported by the division of cells resembling stem cells , which have not been reported previously for F . hepatica . This work will support the study of liver fluke , enabling the development of new drugs and vaccines for the treatment of liver fluke infections of humans and animals .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"livestock",
"medicine",
"and",
"health",
"sciences",
"nucleic",
"acid",
"synthesis",
"rna",
"interference",
"helminths",
"tropical",
"diseases",
"vertebrates",
"fascioliasis",
"light",
"microscopy",
"parasitic",
"diseases",
"animals",
"cell",
"differentiation",
"trematodes",
"developmental",
"biology",
"microscopy",
"neglected",
"tropical",
"diseases",
"epigenetics",
"dna",
"confocal",
"microscopy",
"dna",
"synthesis",
"chemical",
"synthesis",
"research",
"and",
"analysis",
"methods",
"birds",
"genetic",
"interference",
"flatworms",
"fasciola",
"gene",
"expression",
"gamefowl",
"fowl",
"agriculture",
"poultry",
"biosynthetic",
"techniques",
"biochemistry",
"helminth",
"infections",
"rna",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"chickens",
"amniotes",
"organisms"
] |
2016
|
Stimulating Neoblast-Like Cell Proliferation in Juvenile Fasciola hepatica Supports Growth and Progression towards the Adult Phenotype In Vitro
|
The protozoan parasite Trypanosoma cruzi , the causative agent of Chagas disease , causes severe morbidity and mortality in afflicted individuals . About 30% of T . cruzi-infected individuals present with cardiac , gastrointestinal tract , and/or neurological disorders . Megacolon , one of the major pathologies of Chagas disease , is accompanied by gastrointestinal motility disorders . The molecular mechanism of T . cruzi-mediated megacolon in Chagas disease is currently unknown . To decipher the molecular mechanism of T . cruzi-induced alteration in the colon during the early infection phase , we exposed primary human colonic epithelial cells ( HCoEpiC ) to invasive T . cruzi trypomastigotes at multiple time points to determine changes in the phosphoprotein networks in the cells following infection using proteome profiler Human phospho-kinase arrays . We found significant changes in the phosphorylation pattern that can mediate cellular deregulations in colonic epithelial cells after infection . We detected a significant increase in the levels of phosphorylated heat shock protein ( p-HSP ) 27 and transcription factors that regulate various cellular functions , including c-Jun and CREB . Our study confirmed significant upregulation of phospho ( p- ) Akt S473 , p-JNK , which may directly or indirectly modulate CREB and c-Jun phosphorylation , respectively . We also observed increased levels of phosphorylated CREB and c-Jun in the nucleus . Furthermore , we found that p-c-Jun and p-CREB co-localized in the nucleus at 180 minutes post infection , with a maximum Pearson correlation coefficient of 0 . 76±0 . 02 . Increased p-c-Jun and p-CREB have been linked to inflammatory and profibrotic responses . T . cruzi infection of HCoEpiC induces an increased expression of thrombospondin-1 ( TSP-1 ) , which is fibrogenic at elevated levels . We also found that T . cruzi infection modulates the expression of NF-kB and JAK2-STAT1 signaling molecules which can increase pro-inflammatory flux . Bioinformatics analysis of the phosphoprotein networks derived using the phospho-protein data serves as a blueprint for T . cruzi-mediated cellular transformation of primary human colonic cells during the early phase of T . cruzi infection .
The protozoan parasite Trypanosoma cruzi is the causative agent of Chagas disease , a neglected tropical disease which causes severe morbidity and mortality worldwide . Originally endemic in South American countries where it still constitutes a severe socioeconomic burden , Chagas disease has spread around the world and become a global health crisis [1 , 2] . Currently , the disease is present in all major economically advanced countries due to modern globalization and migration [3] . As many as 30% of afflicted individuals eventually present with cardiac , gastrointestinal tract and/or neurological disorders [4] . The development of megacolon , as one of the pathologies of T . cruzi infection , is usually accompanied by unwanted changes in gastrointestinal ( GI ) tract motility which is thought to be due to decrease in the efficiency of the enteric nervous system [5 , 6] . GI motility disorders have been attributed to alterations in the number of interstitial cells of Cajal and enteric nervous system defects . Although it is generally agreed that the enteric neurons [7 , 8] and interstitial cells of Cajal [8 , 9] decrease in numbers in megacolon , it is unclear what roles they play in the pathophysiology of chagasic megacolon . The presence of more natural killer and cytotoxic T-cells in colon lesions from patients with megacolon suggest that immune responses also play a role in the neuronal loss in chagasic megacolon patients [6] . A study using a murine model of chagasic megacolon showed that megacolon was accompanied by increases in colon wall thickness , hypertrophy , and collagen deposition , which are hallmarks of fibrosis [7] . This report correlates with others showing an increase in fibrotic lesions in smooth muscle and myenteric plexus of chagasic megacolon tissue sections [8] . The fibrotic lesions observed in megacolon tissue sections can be caused by increased deposition of extracellular matrix ( ECM ) and matricellular proteins including TSP-1 . The interactions between T . cruzi and colon cells including colon epithelium cells can deregulate cell signaling pathways leading to increased expression of transcription factors that upregulate the synthesis of ECM proteins [10 , 11] causing fibrogenesis and cellular transformation reported in megacolon tissue sections . The role played by colon epithelium in the onset of chagasic megacolon remains unknown . To understand the pathogenesis of chagasic megacolon , researchers will need to study the role of parasite-induced signaling molecules including cytokines , chemokines , neurotransmitters , and neurotrophic factors in mediating signal transfer within and among colon cells . Currently , studies elucidating parasite-induced changes in colon cells involved in the onset and development of chagasic megacolon is lacking . The fact that there have been reports of Chagas disease outbreaks caused by consumption of T . cruzi contaminated foods where the infected individuals presented with clinical manifestations comparable to those infected via other types of transmission routes shows that the parasite can infect intestinal epithelial cells [12 , 13] . Furthermore , murine experimental T . cruzi infections have been established using oral cavity and gastrointestinal gavage inoculation routes [14] . We hypothesize that the molecular interactions between T . cruzi and colon epithelium will deregulate cell signaling cascades , leading to an increase in active transcription factors that can cause increased deposition of matrix proteins , fibrogenesis , loss of colon elasticity and eventually megacolon pathology . To delineate the molecular mechanisms of T . cruzi-induced changes in colon epithelium during the early stage of infection , we challenged primary human colonic epithelial cells with invasive T . cruzi trypomastigotes . This infection model closely mimics the physiological state and changes in host phospho-proteome profile induced by the parasite are essential in understanding parasite mediated pathology . A predominance of megacolon pathology was reported in T . cruzi infected individuals from Chile , which is the origin of the Tulahuen strain , suggesting a colon tropism for the strain [15] . Additionally , T . cruzi Tulahuen strain has been shown to generate experimental murine Chagas disease infection through the gastrointestinal tract [14] . In our study , we show that T . cruzi Tulahuen trypomastigotes induce significant changes in phosphorylation patterns of a variety of signaling proteins and transcription factors including c-Jun and CREB during the early phase of cellular infection . We also observed that these phosphorylated transcription factors are translocated and colocalized in the nucleus in a time-dependent manner . These T . cruzi induced changes during the early infection phase can lead to a fibrogenic response in host colon epithelial cells , subsequently resulting in the development of colon pathology .
The proteome profiler human phospho-kinase array kit ( #ARY003B ) , as well as antibodies against p-AktT308 ( #MAB7419 ) , p-AktS473 ( #AF887 ) , c-Jun ( #MAB2670 ) , STAT1 ( #MAB1490 ) , ( RelA/NF kappa B p65 ( #MAB5078 ) , p-HSP27 ( #MAB23141 ) , HSP27 ( #AF1580 ) , TSP-1 ( #MAB3074 ) and SP600125 ( Inhibitor ) were purchased from R&D Systems ( Minneapolis , MN , USA ) . Antibodies against RSK ( #9355S ) , p-CREBS133 ( #9198S ) , CREB ( #9197S ) , p-JNKT183/Y185 ( #9255S ) , JNK2 ( #9258S ) , p-c-JunS73 ( #3270S ) , and pan Akt ( #4691S ) , as well as p-CREBS133 , p-NF-κB p65 ( #3031 ) , p-IKKα/β ( # 2697 ) , p-STAT1 ( #7649 ) , pTAK ( #9339 ) , p-IRAK ( #11927 ) , IKKα ( #2682 ) , p-CREBS133 ( 87G3 ) Rabbit mAb Alexa Fluor 647 Conjugate ( #14001S ) , p-JunS73 rabbit mAb with Alexa Fluor 488 conjugate ( #12714S ) , anti-Rb HRP ( #7074S ) , and anti-mouse HRP ( #7076S ) were purchased from Cell Signaling Technology ( Danvers , MA , USA ) , and beta actin antibody ( #sc-69879 ) from Santa Cruz ( Dallas , Texas , USA ) . SDS ( 10% ) , acrylamide ( 29:1 ) , blotting-grade blocker , precision plus protein marker , 2-mercaptoethanol , 10x Tris/Glycine/SDS , 10x Tris Buffered Saline , 10% Tween 20 , Trans-Blot Turbo , Mini Nitrocellulose Transfer Packs were purchased from Bio-Rad Laboratory ( Hercules , CA , USA ) . NP40 cell lysis buffer , Phalloidin and Prolong Gold antifade mounting reagent containing DAPI ( #P36935 ) were purchased from Life Technologies ( Carlsbad , CA , USA ) . The comprehensive kit of human colonic epithelial cells , ( HCoEpiC ) ( #2950 ) containing primary cells and all required culturing reagents and supplements was purchased from ScienCell Research Laboratory ( Carlsbad , CA , USA ) . Protease inhibitor cocktail set III was purchased from Calbiochem ( Gibbstown , NJ , USA ) . Phosphatase inhibitor cocktails 2 and 3 , Trizma , ethyl alcohol , and acetic acid were purchased from Sigma Aldrich ( St . Loius , MO , USA ) . HCoEpiC were grown in colonic epithelial cell medium ( CoEpiCM , Cat . #2951 ) supplemented with the accompanying supplements as recommended by the manufacturer ( ScienCell Research Laboratory , CA ) . Briefly , tissue culture flasks were incubated with poly L-lysine solution ( 2 μg/cm2 ) at 37ºC overnight . HCoEpiC were seeded in the coated flasks and cultured at 37ºC and in 5% CO2 . Confluent cell monolayers ( about 80% ) were used in our assays . For inhibition of JNK-c-Jun pathway , the cells were pretreated overnight with 100µM SP600125 . Heart myoblast monolayers ( 80% confluence ) grown in complete DMEM containing glutamax , 10% fetal bovine serum , 1% penicillin/streptomycin , multivitamins , and MEM non-essential amino acids ( Life Technologies , Carlsbad , CA , USA ) were infected with invasive T . cruzi trypomastigotes Tulahuen strain clone MMC 20A [16] . Transgenic T . cruzi Tulahuen trypomastigotes expressing green fluorescent protein ( GFP ) , generated as previously described [17] were used to analyze cellular infection through confocal microscopy . Highly invasive T . cruzi trypomastigotes were harvested from tissue culture supernatants as previously described [16] . The parasites were washed in Hanks Balanced Salt Solution ( HBSS ) and resuspended in CoEpiCM without supplements at a concentration of 1x107 parasites/ml . HCoEpiC were starved in CoEpiCM without supplements for 1h at 37°C , 5% CO2 prior to the addition of trypomastigotes . The starved cells were challenged with pure population of T . cruzi trypomastigotes at a ratio of 10 parasites per cell for different time points ( 60 , 90 , 120 , 180 minutes ) essentially as described [10] . Parasites were washed off with 1X DPBS ( without calcium/magnesium ) and the cells were either processed immediately or stored at -80ºC for further experimentation . To analyze the T . cruzi infection patterns of HCoEpiC , the GFP trypomastigotes were washed off and the cells were incubated with fresh complete CoEpiCM with daily changes up to 72h . The cells were fixed with 4% paraformaldehyde for 5 minutes at room temperature , washed with 1X DPBS , stained with phalloidin ( 1:2000 ) at 4°C overnight and mounted with mounting media containing DAPI to stain the nuclei for microscopy . The number of internalized parasites per cell were evaluated by screening no less than 200 cells per well . To analyze the phosphorylation profiles of kinases and their protein substrates in T . cruzi-infected primary colon cells , we used a commercially available human phospho-kinase array kit following the manufacturer's protocol ( R&D Systems ) . This is a membrane-based sandwich immunoassay kit where the capture antibodies against 43 kinase phosphorylation sites , two other proteins and control proteins are spotted in duplicates on nitrocellulose membranes to bind specific target proteins present in the cell lysates ( Product Datasheet ARY003B , R&D Systems ) . Briefly , blocked array membranes were incubated with T . cruzi infected and the same time points uninfected cell protein lysates , and with the same amount of T . cruzi trypomastigotes protein lysate ( ~250µg ) , at 4°C overnight on a platform shaker . Washed membranes were further incubated with biotinylated antibody cocktails for 2h at room temperature on a rocking platform . The membranes were washed , probed with streptavidin-HRP and visualized by chemiluminescence using X-ray films . The films were scanned , and the density of each spot was quantified against the average of the internal controls as recommended by the manufacturers ( R&D Systems ) . Densitometric data analyses were done using the free Image J software available on the National Institutes of Health ( NIH ) website ( https://imagej . nih . gov/ij/ ) . In order to validate the data obtained using the phospho-kinase array kit , we used immunoblotting assays to evaluate the phosphorylation profile of selected phosphoproteins as previously described [10] . Briefly , serum-starved HCoEpiC incubated with invasive T . cruzi trypomastigotes were lysed with the NP40 cell lysis buffer containing phosphatase inhibitor cocktails 2 and 3 , and protease inhibitor cocktail set III , each at a ratio of 1:100 . The cell lysates ( 20μg/lane ) were separated by SDS-PAGE and blotted onto nitrocellulose membranes using the Trans-Blot Turbo system . The membranes were stained with Ponceau S staining solution ( #P7170 , Sigma Aldrich ) to verify protein transfer . The membranes were washed and blocked with blocking buffer ( 1X TBS pH 7 . 4 , 5% nonfat dry milk and 0 . 1% Tween-20 ) for 1h at room temperature . Blocked membranes were incubated with the respective phospho-primary antibodies in antibody buffer ( 1X TBS pH 7 . 4 , 1% nonfat dry milk and 0 . 1% Tween-20 ) at 4°C overnight . The blots were washed to remove excess antibodies , probed with the corresponding HRP-conjugated secondary antibodies , visualized by chemiluminescence , and scanned . The membranes were stripped and re-probed with antibodies against the corresponding total proteins or beta actin to normalize the data . Data were collected from three independent sets of experiment and analyzed by densitometry using Image J software . Immunofluorescence assays were used to detect phosphoproteins of interest in the nucleus . HCoEpiC seeded on Lab-Tek chamber slides were used for immunofluorescence assays as previously described [10] . Briefly , T . cruzi trypomastigotes ( 10 parasites per cell ) were incubated with the cells for various lengths of time . The parasites were washed off . The cells were fixed with 4% paraformaldehyde for 5 minutes at room temperature and washed with 1X DPBS . Fixed cells were perforated with 0 . 1% Triton-X100 in TBS for 5 minutes and blocked with 3% BSA-PBS for 30 minutes at room temperature . Slides were incubated with anti-human-p-c-JunS73 Alexa 488 conjugate ( 1:100 ) , anti-p-CREB Alexa 647 conjugate ( 1:100 ) , and phalloidin ( 1:2000 ) at 4°C overnight . The slides were washed with 1% BSA-PBS and mounted with mounting media containing DAPI to stain the nuclei . For colocalization assays , the fixed , perforated and blocked slides were incubated with a mixture containing anti-human-p-c-JunS73 Alexa 488 conjugate and anti-human-p-CREB Alexa 647 conjugate at a dilution of 1:100 each , washed and mounted with mounting media containing DAPI to stain the nuclei . Stained slides were analyzed using the Nikon A1R confocal microscope at the Morphology Core Facility at Meharry Medical College . The functional analysis of biological pathways of altered phosphoproteins were elucidated using bioinformatics approach . Protein annotation was done using GENECARDS ( www . genecards . org ) and DAVID ( david . ncifcrf . gov ) to find common reference and to achieve de-aliasing . Biological Pathways were constructed using Ingenuity Pathway Analysis Path Explorer ( IPA , QIAGEN ) . Significantly enriched canonical pathways were determined using Fisher’s Exact test for enrichment . Significantly enriched canonical pathways constructed by IPA were grouped into five meta-level categories: 1 ) immunological response , 2 ) fibrotic extracellular signaling , 3 ) neuronal response , 4 ) intracellular signaling , and 5 ) stress associated response . Visualization of proteins to pathway group mapping was done using table2net ( http://tools . medialab . sciences-po . fr/table2net/ ) for graph file parsing and formatting followed by Gephi ( https://gephi . org/ ) for visualization . All data were collected from three independent sets of experiments . Alterations in phosphorylated protein levels were analyzed using Student’s t-test or one-way analysis of variance ( ANOVA ) for multiple groups of data . The statistical analyses were performed using SPSS software for phospho-array measurements . For altered protein expression , a fold change of ≥1 . 5 with p-value ≤ 0 . 01 was considered significant . For biological pathway analysis and mapping , a Fisher’s exact test was used to identify enriched pathways and a p-value ≤ 0 . 001 was considered significant . AKt1/2/3 ( AKT1 [207] , AKT2 [208] , AKT3[10000] ) , AMPKα1 ( PRKAA1 [5562] ) , AMPKα2 ( PRKAA2 [5563] ) , Chk2 ( CHEK2 [11200 ) , c-Jun ( JUN [3725 ) , CREB ( CREB1 [1385] ) , EGFR ( EGFR [1956] ) , Enos ( NOS3 [4846] ) , ERK1/2 ( MAPK1 [5594] ) , FAK ( PTK2 [5747] ) , Fgr ( FGR [2268] ) , Fyn ( FYN [2534] ) , GSK-3α/β ( GSK3A [2931] , GSK3B [2932] ) , HcK ( HCK [3055] ) , HSP27 ( HSPB1 [3315] , HSPB2 [3316] ) , HSP60 ( HSP60 [3329] ) , JNK1/2/3 ( MAPK8 [5599] , MAPK9 [5601] , MAPK10[5602] ) , Lck ( LCK [3932] ) , Lyn ( LYN [4067] ) , MSK1/2 ( RPS6KA5 [9252] , RPS6KA4 [8986] ) , p27 ( IFI27 [3429] ) , p38α ( MAPK14 [1432] ) , p53 ( TP53 [7157] ) , p70 S6 ( RPS6KB1 [6198] , RPS6KB2 [6199] ) , PDGF Rβ ( PDGFRB [5159] ) , PLC-γ1 ( PLCG1 [5335] ) , PRAS40 ( AKT1S1 [84335] ) , PYK2 ( PTK2B [2185] ) , RSK1/2/3 ( RPS6KA1 [6195] , RPS6KA3 [6197] , RPS6KA2 [6196] ) , Src ( SRC [6714] ) , STAT2 ( STAT2 [6773] ) , STAT3 ( STAT3 [6774] ) , STAT5a ( STAT5A [6776] ) , STAT5a/b ( STAT5A [6776] , STAT5B[6777] ) , STAT5b ( STAT5B [6777] ) , STAT6 ( STAT6 [6778] ) , TOR ( MTOR [2475] ) , WNK1 ( WNK1 [65125] ) , Yes ( YES [7525] ) , β-Catenin ( CTNNB1 [1499] ) , TSP1 ( THBS1 [7057] ) , β-ACTIN ( ACTB [60] ) , IKKβ ( IKBKB [3551] ) , IRAK4 ( IRAK4 [51135] ) , TAK-1 ( MAP3K7 [6885] ) , JAK2 ( JAK2 [3717] ) , STAT1 ( STAT1 [6772] ) , NFKβ-P65 ( RELA [5970] ) , IKKα ( CHUK [1147] ) .
To evaluate T . cruzi invasiveness of HCoEpiC , we analyzed the percentage of infection after challenging the cells with transgenic T . cruzi Tulahuen trypomastigotes expressing GFP at different time points . The percentage of infected HCoEpiC was maximum at 180 minutes , where about 30% of the cells are infected with 1 . 8±0 . 05 parasites per cell ( Fig 1A and 1B ) . HCoEpiC sustained regular T . cruzi infection where we observed that more than 80% of the cells contained multiplying amastigotes at 72h post infection ( Fig 1C ) . The data is represented microscopically ( Fig 1D ) . Our data represent the first report of the kinetics of primary human colon epithelial cells in vitro infection by T . cruzi Tulahuen strain . To gain insight into the early regulation of phosphoproteins and their associated signaling cascades mediated by T . cruzi trypomastigote infection in colonic epithelial cells , we utilized the human phospho-kinase array . We challenged HCoEpiC with T . cruzi for different lengths of time ( 0 , 60 , 90 , 120 and 180 minutes ) and analyzed the protein phosphorylation profiles following infection ( Fig 2A , S1 Table ) . The time intervals selected were adequate to accommodate important downstream phosphorylation patterns . We also analyzed the protein phosphorylation profiles of HCoEpiC in the starved condition in the absence of parasites at the same time points . The phosphorylation profile of the 43 kinase phosphorylation sites was conserved over time in uninfected cells and did not affect the T . cruzi induced phosphorylation pattern ( S1 Fig , S2 Table ) . Furthermore , we also observed no cross reactivity on the membranes using the same amount of T . cruzi trypomastigote lysate ( Fig 2A ) . The reference spots on the arrays ( Fig 2A and S1 Fig ) are included to align the array membrane and to show that the array has been incubated with streptavidin-HRP during the assay procedure . We found that during the early phase of infection , T . cruzi infection significantly altered the phosphorylation pattern of 21 kinase phosphorylation sites ( S1 Table ) . Our interest here is to group and map the altered phosphoproteins based on their functional role; including fibrosis , neurological signaling , and immune signaling among others . We employed bioinformatics analyses to determine the deregulated pathways associated with modulated phosphoproteins . Our analyses uncovered several phosphoproteins that are involved with cellular transformation pathways as well as other functional pathways associated with proinflammatory responses and several disease pathogenesis pathways . Specifically , we found that the parasite activated the JNK and c-Jun signaling pathways ( Fig 2B ) and CREB associated signaling pathway ( Fig 2C ) . The level of p-c-Jun was significantly increased by 2 . 4±0 . 06 fold at the 120-minutes but it is 1 . 7±0 . 02 at 180 minutes compared to control . Phosphorylation of JNK2 , a known upstream molecule of c-Jun , was also upregulated during infection . The phosphorylation of transcription factor CREB was significantly increased by at least two-fold compared to the control . This was accompanied by a significant increase in the phosphorylated levels of HSP27 , as well as kinases Src , Akt S473 and Fyn at different time points . The levels of p-Akt T308 and p-RSK were not significantly increased ( Fig 2C ) . Additionally , phosphorylated levels of several proteins including ERK1/2 , p27 , TOR and STAT5A were downregulated at differing time points ( S1 Table ) . Taken together , these results provide us with the foundation to map the phospho-proteomic network that operates in human colon cells challenged with T . cruzi . To understand the functions of the altered phosphoproteins during T . cruzi infection and their role in Chagas disease progression , we developed a biological interaction network that to indicate the relationship among individual proteins in the colon cells as mediated by T . cruzi . We were able to map the phosphoproteins to multiple cellular processes and pathways that are deregulated in many disease states ( Fig 3 ) . Specifically , we mapped the altered phosphoproteins to stress , immunological , intracellular , fibrotic , and neuronal signaling responses , indicating the interaction among different yet overlapping arms of cellular responses that can contribute to the onset of colon pathology induced by T . cruzi infection . The canonical pathways involved during the early phase of T . cruzi infection of primary HCoEpiC is shown ( S3 Table ) . The phosphoproteomic array revealed the upregulation of p-CREB in the host cells . The array data also showed that the phosphorylated levels of Akt S473 and HSP27 were significantly upregulated . We found that the level of pHSP27 was increased more than two-fold at 120 and 180 min ( Fig 4A ) . Our western blot data showed that the phosphorylated level of Akt T308 was increased to 1 . 34±0 . 03 fold at 120 minutes and then decreased to 1 . 27±0 . 01 fold at 180 minutes ( Fig 4B ) . The phosphorylated level of Akt S473 was significantly upregulated at all-time points during the infection to a maximum of 2 . 07±0 . 04 fold at 120 minutes , and then decreased to 1 . 38±0 . 11 at 180 minutes ( Fig 4C ) . Our data also showed that the regulatory and kinase domains of Akt were activated by T . cruzi infection . These results indicate that the Akt signaling pathway in the host cells were activated by the parasite in the early infection phase . Furthermore , our array data also showed that pRSK was upregulated in the infected host cells . Our western blot data revealed a steady gradual upregulation of pRSK to 1 . 19±0 . 01 fold at 180 minutes compared to uninfected control cells ( Fig 4D ) . Since p-Akt and p-RSK are upregulated and they are upstream signaling molecules of CREB , we decided to validate the fold change of phosphorylated CREB . Immunoblot analysis of cellular lysates showed that p-CREB was significantly upregulated at least two-fold at several time points ( Fig 4E ) . The phosphoproteome data analysis and the bioinformatics analysis of network interactions revealed that T . cruzi infection promotes the activation of pro-inflammatory signaling pathways . We are particularly interested in JNK signaling , a pathway that is well-known to be activated by external stimuli but has yet to be explored in Chagas disease progression . Phosphorylation of c-Jun , which is downstream of JNK signaling , can mediate cellular transformations implicated in many severe disease states . Our phospho-array data showed significant increases in the level of phosphorylated JNK and c-JunS63 in T . cruzi-infected colonic cells . We evaluated the levels of phosphorylated JNK and c-JunS73 ( not present in the array ) at various time points by western blot analysis using lysates of colonic epithelial cells challenged with T . cruzi . Our immunoblot data showed that the level of phosphorylated JNK , an upstream molecule of c-Jun , was upregulated by 1 . 24±0 . 02 fold at 180 minutes compared to uninfected control colonic cells ( Fig 5A ) . The level of p-c-JunS73 was also upregulated at 90 , 120 and 180 minutes compared to control cells ( Fig 5B ) . Hence , the western blot results match with the phospho-array data . To evaluate whether T . cruzi infection in these cells lead to changes in the expression of c-Jun target genes , we analyzed the level of thrombospondin-1 ( TSP-1 ) protein in the infected cells at the different time points . We observed that the level of TSP-1 protein increases with the infection of HCoEpiC up to a maximum of 2 . 78±0 . 05 fold at 180 minutes compared to uninfected control ( Fig 5C ) indicating that the expression of TSP-1 protein is increased during T . cruzi infection of colon epithelial cells . To confirm this , we preincubated the cells with SP600125 ( 100µM ) , a pharmacological inhibitor of the JNK-c-Jun pathway and observed a decrease in p-JNK and p-c-Jun ( S2 Fig ) . The amount of downstream TSP-1 in the cells pretreated with the inhibitor was significantly decreased in the presence of T . cruzi ( Fig 5D ) . In order to evaluate the kinetics of T . cruzi induced proinflammatory responses in HCoEpiC , we investigated the regulation of NF-kB associated signaling molecules in T cruzi infected HCoEpiC . We found that the p-p65 level was increased up to a maximum of 2 . 48±0 . 04 at 180 minutes compared to uninfected control ( Fig 6A and 6B ) . The level of p-IKKα/β was significantly increased to a maximum of 1 . 91±0 . 02 at 90 minutes with a corresponding decrease in IKKα levels to a minimum of 0 . 59±0 . 05 at 180 minutes ( Fig 6A and 6B ) . Furthermore , we analyzed p-TAK1 and p-IRAK4 , upstream regulators of NF- kB pathway . We observed that p-TAK1 was increased to a maximum of 2 . 30±0 . 08 at 60 minutes and p-IRAK4 was upregulated by T . cruzi infection to a maximum of 1 . 43±0 . 07 at 60 minutes ( Fig 6A and 6B ) . We also evaluated the levels of p-JAK2 and p-STAT1 , signaling molecules that play roles in the regulation of interferon signaling pathway . We observed that levels of p-JAK2 increased to a maximum of 1 . 73±0 . 07 at 180 minutes and that of p-STAT1 increased to a maximum of 2 . 21±0 . 11 at 120 minutes in T . cruzi infected HCoEpiC ( Fig 6C and 6D ) . Our phospho-array data showed that p-c-Jun and p-CREB were significantly upregulated in colonic cells during the early phase of T . cruzi infection . These increase in the phosphorylation of several key proteins was validated by immunoblot assays . To understand the importance of upregulated p-CREB and p-c-Jun in colonic epithelial cells during early T . cruzi infection , we performed immunofluorescence assays to evaluate nuclear translocation and colocalization of both transcription factors . To do this , we measured the localization of each phosphoprotein using the mean fluorescence intensity ( MFI ) . Our data showed that p-CREB was translocated to the nucleus . Specifically , the nuclear translocation of p-CREB significantly increased over time to a maximum of 112 . 6±10 . 9 MFI at 180 minutes ( Fig 7A and 7B ) . We also performed immunofluorescence assays to evaluate the nuclear translocation of p-c-JunS73 . Our results show nuclear staining of this phosphoprotein during early phase of T . cruzi infection . We found that the nuclear localization of p-c-JunS73 significantly increased over time to a maximum of 123 . 2±7 . 5 MFI at 180 minutes ( Fig 7C and 7D ) . These data agree with our array analysis that indicates that T . cruzi infection increased the level of active transcription factors p-CREB and p-c-JunS73 in the nuclei of infected primary human colon cells . Since we observed that the nuclear translocation of p-c-Jun and p-CREB are increasing with time during T . cruzi infection , we next evaluated if both active transcription factors colocalized in the nucleus . Our confocal microscopy data showed that both phosphorylated transcription factors also colocalized in the nucleus , and that the extent of this colocalization increased over time during T . cruzi infection ( Fig 8A and 8B ) . The Pearson’s correlation value of this colocalization significantly increased with time to 0 . 70±0 . 02 at 120 minutes and a maximum of 0 . 76±0 . 02 at 180 minutes ( Fig 8B ) . Taken together , these results show for the first time that T . cruzi infection enhances the colocalization of both transcription factors into the nuclei of infected colon cells .
Chagasic megacolon is a major pathology associated with severe morbidity and mortality in T . cruzi-infected patients . The molecular mechanisms that cause megacolon in Chagasic patients remains largely undefined . Primary colonic epithelial cells constitute a good model for studying the mechanisms of T . cruzi infection of colon cells . We found that T . cruzi can infect colonic epithelial cells with more than 80% cellular infection after 72h ( Fig 1 ) . We hypothesize that during T . cruzi infection , the parasite deregulates host signal transduction and eventually the gene transcription profiles to cause symptoms associated with Chagas disease [10 , 18 , 19] . To elucidate the molecular mechanism of T . cruzi-induced molecular alterations in colon cells , we challenged primary human colonic epithelial cells with invasive T . cruzi trypomastigotes for various lengths of time ( 0 , 60 , 90 , 120 and 180 minutes ) to evaluate the pattern of altered phosphoproteins after infection using phospho-proteomic arrays and the mapping of biological network interactions . Analyzing the phosphorylation profiles of kinases and their protein substrates is essential for understanding how cells recognize and respond to the presence of T . cruzi in their micro-environment . Normal cellular activities in colon cells are highly susceptible to alterations following exposure to T . cruzi trypomastigotes . The analysis of phosphorylation profiles of kinases and their respective protein substrates may facilitate our understanding of the cellular response mediated by T . cruzi trypomastigotes in host colonic cells . In the present study , our goal was to decipher the phospho-signaling pathways that operate during early T . cruzi infection in human colonic epithelial cells . We investigated the phosphorylation activity of several axes of signaling cascades that may mediate T . cruzi-induced colon pathology . Our results showed that the parasite significantly altered the pattern of phosphorylated kinases and phosphoprotein levels during infection . Specifically , we found that T . cruzi deregulated the phosphorylated levels of 21 kinase phosphorylation sites , in the infected cells at different time points compared to uninfected controls . Bioinformatics analyses revealed that these phosphoproteins can be mapped to a variety of signaling pathways including those involved in neuronal , inflammatory , and fibrotic responses . Currently , it has been proposed that intense degenerative inflammatory changes occur in chagasic megacolon tissues , accompanied by increased collagen deposition and fibrosis . These changes are thought to mediate neuronal damage as well as intramuscular denervation , leading to chagasic megacolon [7] . The bioinformatics network developed in our study revealed significant alterations in the phosphoproteome associated with the signaling pathways that may cause the colon pathology observed in chagasic patients . The altered phosphoproteins we observed in our study have been reported to play a role in cellular pathways that can cause tissue fibrosis , neuronal damages , and negative immunological responses . The upregulated phosphoproteins involved in neurological and inflammatory pathways can damage colonic myenteric neurons during T . cruzi infection [20] . Bioinformatics analysis of array data also showed an overall increase in the proinflammatory response in HCoEpiC during T . cruzi infection . Hence , we further analyzed the associated proinflammatory pathways and found that NF-kB and JAK2-STAT1 pathways are activated by T . cruzi infection . The increased level of NF-kB ( p-NF-kB , p-IKKα/β , p-TAK1 and p-IRAK4 ) revealed the high probability of activation of Toll like receptors by T . cruzi infection of HCoEpiC [21] . The upregulated level of JAK2-STAT1 pathway in HCoEpiC also correlates with the activation of interferon signaling pathway during T cruzi infection . Our observation complements previous studies suggesting the activation of NF-kB and JAK-STAT signaling pathways by T . cruzi [22–24] . Some of the phosphoproteins modulated in array during the infection , including p-c-Jun , have been correlated with increased expression of TSP-1 . Previous reports show that the transcription factor c-Jun enhances TSP-1 promoter activity resulting in increase in TSP-1 [25 , 26] . These findings support our previous reports showing that the parasite increased the level of TSP-1 during cellular infection [27] . Increased TSP-1 can mediate fibrotic responses [28] including those seen in megacolon [29] . The above observations support our findings that the level of TSP-1 increases during T . cruzi infection of colonic epithelial cells . This increase in TSP-1 protein is inhibited by a JNK inhibitor SP600125 suggesting that TSP-1 protein increase is regulated by p-c-Jun during T . cruzi infection ( Fig 5C and 5D ) . In the current study we focused on transcription factors CREB and c-Jun whose dysregulation have been suggested to be involved in several major pathologies . Deregulated levels of p-CREB have been reported in brain pathology [30] and defects in neuronal homeostasis [31] . c-Jun has been reported to play a role in sclerosis [32] , neuronal defects [33] , and several other pathologies . Our array data showed significant upregulation of p-Akt S473 in T . cruzi-infected colonic cells compared to uninfected control . Our data also showed a significant increase in upstream p-JNK and downstream p-c-Jun during T . cruzi infection . This is important as an increase in JNK phosphorylation has been reported to play a role in endothelial dysfunction [34] . The activation of JNK followed by CREB as shown in our study was previously reported to be important in enhancing the transcription of sphingosine kinase 2 ( SPHK2 ) , which is elevated in colon cancer [35] . Studies have also shown that RSK1 may regulate CREB phosphorylation to perform various biological functions [36 , 37] . Our data emphasized the potential role of CREB and JNK in causing the colon pathology observed in some T . cruzi-infected individuals . Our data also showed the upregulation of p-HSP27 , an anti-apoptotic protein [38] and HSP60 , a molecular chaperone which facilitates cyto-protection through stabilization of mitochondrial proteins [39] during T . cruzi infection . These proteins collinearly matched the increase in Akt phosphorylation in the infected cells . Our results show upregulation of p-HSP27 during T . cruzi infection , in agreement with other reports , suggesting that T . cruzi infection enhances cell survival [40 , 41] . Additionally , Akt activation in stressed cells results in HSP27 phosphorylation [42] , and the prosurvival effect of this protein occurs through its interaction with Akt . Interestingly , c-Jun and CREB play a role in the expression of several proteins that are essential in regulating extracellular matrix ( ECM ) protein interactions [43 , 44] . The deregulation of both these phosphorylated proteins during early T . cruzi infection results in immune modulations and fibrogenic responses . These fibrogenic responses increase fibrosis and muscular hypertrophy reported in the colonic sections from Chagas disease patients suffering from megacolon [8 , 9 , 45] . For instance , p-CREB has been reported to transactivate TGF-β1 expression to mediate hepatic fibrosis [46]; it also plays a key role in several immune responses [47] . JNK signaling plays an integral role in several key mechanisms involved in fibrogenic responses via direct phosphorylation of SMAD3 , a profibrotic transcription factor [10 , 48] . We also previously explained how ECM proteins are involved in T . cruzi-mediated Chagas pathology [11 , 49] . CREB promotes an antiapoptotic survival signal in macrophages , leading to enhanced host immune responses during infection [50 , 51] . Others also showed that CREB phosphorylation can induce TNF-α production during bacterial infections [52] . Activated CREB promotes a survival signal response in macrophages [47] , leading to immune inflammatory responses during T . cruzi infection . Additionally , c-Jun has diverse functions in colon disease as well as in cancer progression [53 , 54] . Data from our phosphoproteomic assays showed a significant increase in Src phosphorylation . These observations support a previously published conclusion that the remodeling of the ECM occurs through increased focal adhesion kinase and Src signaling [55] . Therefore , we postulate in our model that this may be one of the ways that T . cruzi causes pathology observed in chagasic individuals ( Fig 9 ) . Our data show that T . cruzi infection of colonic cells significantly enhanced the phosphorylation of c-Jun and CREB in a time-dependent manner during the early phase of infection . Furthermore , nuclear colocalization of both these activated transcription factors increased in the infected cells compared to uninfected controls . Our data complement findings from previous studies showing that CREB binds to consensus cAMP-response element ( CRE; TGACGTCA ) and the overlapping activator protein-1 ( AP-1 ) motif ( TGACTGA , designated CRE2/AP-1 ) [56] . The phosphorylation and nuclear colocalization of both CREB and c-Jun during early T . cruzi infection may promote heterodimer formation between these two transcription factors and/or competitive binding during T . cruzi infection . Therefore , these transcription factors may play a novel role in mediating molecular deregulations in T . cruzi-infected colon cells . However , the detailed mechanisms remain to be deciphered . Our model suggests that T . cruzi increases cellular levels of phosphorylated kinases , thus activating transcription factors including p-c-Jun and p-CREB that are all translocated into the nucleus to regulate inflammatory responses and cellular transformations . The colocalization of p-c-Jun and p-CREB in HCoEpiC challenged with T . cruzi is a novel observation in the context of T . cruzi infection ( Fig 9 ) . It has been previously reported that p-c-Jun and p-CREB cross-talk to cause cellular transformation [56–58] . Furthermore , increased TSP-1 expression induced by increased p-c-Jun can lead to a fibrogenic response which can cause colon pathology that could be implicated in the onset of megacolon . Fibrogenesis , due increased deposition of TSP-1 , can lead to a marked thickening of the colon epithelium as observed in tissue sections of specimens from chagasic megacolon patients compared to controls [59] . Overall , we conclude that the colocalization of upregulated p-c-Jun and p-CREB is a putative key feature in T . cruzi-mediated colon pathology , warranting further investigation .
|
Trypanosoma cruzi is a hemoflagellate that is now considered a global health threat in all industrialized regions of the world . Some chagasic patients present with digestive , neurological , and/or cardiac disorders . The mechanisms of T . cruzi-induced pathology remain to be elucidated . In this study , we challenged primary human colonic epithelial cells with T . cruzi and evaluated changes in the phosphorylated kinases and phosphoprotein levels that may induce cellular and molecular alterations leading to cellular transformations during the early phase of infection . The parasite induced significant increases in levels of phosphorylated kinases and phosphoproteins that govern multiple cellular pathways associated with immunological , stress , neuronal , and intercellular interactions as well as fibrogenic responses . The parasite also enhanced the levels of p-AKT , p-HSP27 , p-JNK , and downstream transcription factors like p-c-Jun and p-CREB during the early infection phase . Additionally , we observed that the phosphorylated transcription factors are translocated to and colocalized in the nucleus in a time-dependent manner . These transcription factors regulate the expression of genes , including genes encoding extracellular matrix proteins , which play a role in the onset of colon pathology observed in some chagasic patients . Our study provides novel insights into the interactome that occurs during acute phase of T . cruzi infection of primary human colon cells .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"gene",
"regulation",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"epithelial",
"cells",
"megacolon",
"protozoans",
"gastroenterology",
"and",
"hepatology",
"transcription",
"factors",
"digestive",
"system",
"animal",
"cells",
"proteins",
"gene",
"expression",
"biological",
"tissue",
"gastrointestinal",
"tract",
"trypanosoma",
"cruzi",
"biochemistry",
"trypanosoma",
"eukaryota",
"anatomy",
"post-translational",
"modification",
"cell",
"biology",
"genetics",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"phosphoproteins",
"colon",
"organisms"
] |
2018
|
Phospho-proteomic analysis of primary human colon epithelial cells during the early Trypanosoma cruzi infection phase
|
Genome-wide association studies ( GWAS ) have uncovered numerous genetic variants ( SNPs ) that are associated with blood pressure ( BP ) . Genetic variants may lead to BP changes by acting on intermediate molecular phenotypes such as coded protein sequence or gene expression , which in turn affect BP variability . Therefore , characterizing genes whose expression is associated with BP may reveal cellular processes involved in BP regulation and uncover how transcripts mediate genetic and environmental effects on BP variability . A meta-analysis of results from six studies of global gene expression profiles of BP and hypertension in whole blood was performed in 7017 individuals who were not receiving antihypertensive drug treatment . We identified 34 genes that were differentially expressed in relation to BP ( Bonferroni-corrected p<0 . 05 ) . Among these genes , FOS and PTGS2 have been previously reported to be involved in BP-related processes; the others are novel . The top BP signature genes in aggregate explain 5%–9% of inter-individual variance in BP . Of note , rs3184504 in SH2B3 , which was also reported in GWAS to be associated with BP , was found to be a trans regulator of the expression of 6 of the transcripts we found to be associated with BP ( FOS , MYADM , PP1R15A , TAGAP , S100A10 , and FGBP2 ) . Gene set enrichment analysis suggested that the BP-related global gene expression changes include genes involved in inflammatory response and apoptosis pathways . Our study provides new insights into molecular mechanisms underlying BP regulation , and suggests novel transcriptomic markers for the treatment and prevention of hypertension .
Systolic and diastolic blood pressure ( SBP and DBP ) are complex physiological traits that are affected by the interplay of multiple genetic and environmental factors . Hypertension ( HTN ) is a critical risk factor for stroke , renal failure , heart failure , and coronary heart disease [1] . Genome-wide association studies ( GWAS ) have identified numerous loci associated with BP traits [2 , 3] . These loci , however , only explain a small proportion of inter-individual BP variability . In aggregate the 29 loci reported by the International Consortium of Blood Pressure ( ICBP ) consortium GWAS account for about one percent of BP variation in the general population [3] . Most genes near BP GWAS loci are not known to be mechanistically associated with BP regulation [3] . Therefore , further studies are needed to determine whether the genes implicated in GWAS demonstrate functional relations to BP physiology and to uncover the molecular actions and interactions of genetic and environmental factors involved in BP regulation . Alterations in gene expression may mediate the effects of genetic variants on phenotype variability . We hypothesized that characterizing gene expression signatures of BP would reveal cellular processes involved in BP regulation and uncover how transcripts mediate genetic and environmental effects on BP variability . We additionally hypothesized that by integrating gene expression profiling with genetic variants associated with altered gene expression ( eSNPs or eQTLs ) and with BP GWAS results , we would be able to characterize the genetic architecture of gene expression effects on BP regulation . Several previous studies have examined the association of global gene expression with BP [4 , 5] or HTN [6 , 7] . Most of these studies , however , were based on small sample sizes and lacked replication [4 , 5 , 6 , 7] . To address this challenge , we conducted an association study of global gene expression levels in whole blood with BP traits ( SBP , DBP , and HTN ) in six independent studies . In order to avoid the possibility that the differentially expressed genes we identified reflect drug treatment effects , we excluded individuals receiving anti-hypertensive treatment . The eligible study sample included 7017 individuals: 3679 from the Framingham Heart Study ( FHS ) , 972 from the Estonian Biobank ( EGCUT ) , 604 from the Rotterdam Study ( RS ) [8] , 597 from the InCHIANTI Study , 565 from the Cooperative Health Research in the Region of Augsburg [KORA F4] Study [9] , and 600 from the Study of Health in Pomerania [SHIP-TREND] [10] . We first identified differentially expressed BP genes in the FHS ( n = 3679 ) followed by external replication in the other five studies ( n = 3338 ) . Subsequently , we performed a meta-analysis of all 7017 individuals from the six studies , and identified 34 differentially expressed genes associated with BP traits using a stringent statistical threshold based on Bonferroni correction for multiple testing of 7717 unique genes . The differentially expressed genes for BP ( BP signature genes ) were further integrated with eQTLs and with BP GWAS results in an effort to differentiate downstream transcriptomic changes due to BP from putatively causal pathways involved in BP regulation .
After excluding individuals receiving anti-hypertensive treatment , the eligible sample size was 7017 ( FHS , n = 3679; EGCUT , n = 972; RS , n = 604; InCHIANTI , n = 597; KORA F4 , n = 565 and SHIP-TREND , n = 600 ) . Clinical characteristics of participants from the four studies are presented in Table 1 . The mean age varied across the cohorts ( FHS = 51 , EGCUT = 36 , RS = 58 , InCHIANTI = 71 , KORA F4 = 72 and SHIP-TREND = 46 years ) as did the proportion of individuals with hypertension ( 11% in FHS , 19% in EGCUT , 35% in RS , 45% in InCHIANTI , 26% in KORA , and 12% in SHIP ) . At a Bonferroni corrected p<0 . 05 , we identified 73 , 31 , and 8 genes that were differentially expressed in relation to SBP , DBP , and HTN , respectively in the FHS , which used an Affymetrix array for expression profiling , and 6 , 1 , and 1 genes in the meta-analysis of the 5 cohorts that used an Illumina array ( Illumina cohorts ) : EGCUT , RS , InCHIANTI , KORA F4 and SHIP-TREND ( S1 Table ) . For each differentially expressed BP gene in the FHS or in the Illumina cohorts , we attempted replication in the other group . At a replication p<0 . 05 ( Bonferroni corrected ) , 13 unique genes that were identified in the FHS were replicated in the Illumina cohorts , including 10 for SBP ( CD97 , TAGAP , DUSP1 , FOS , MCL1 , MYADM , PPP1R15A , SLC31A2 , TAGLN2 , and TIPARP ) , 5 for DBP ( CD97 , BHLHE40 , PRF1 , CLC , and MYADM ) , and 2 for HTN ( GZMB and MYADM ) ( Table 2 ) . Each of the unique BP signature genes in the Illumina cohorts , 6 for SBP ( TAGLN2 , BHLHE40 , MYADM , SLC31A2 , DUSP1 , and MCL1 ) , 1 for DBP ( BHLHE40 ) and 1 for HTN ( SLC31A2 ) , replicated in the FHS . All 6 Illumina cohorts BP signature genes that replicated in the FHS were among the 13 FHS BP signature genes that replicated in the Illumina cohorts . The BP signature genes identified in the FHS showed enrichment in the Illumina cohorts at pi1 = 0 . 88 , 0 . 75 , and 0 . 99 for SBP , DBP , and HTN respectively ( pi1 value indicates the proportion of significant signals among the tested associations [11]; see details in the Methods section ) . Fig . 1 shows that the mean gene expression levels of the top BP signature genes were consistent with the BP phenotypic changes observed in the FHS and the Illumina cohorts . The 73 SBP signature genes in the FHS ( 55 of these 73 genes were measured in the Illumina cohorts ) at a Bonferroni corrected p<0 . 05 in aggregate explained 9 . 4% of SBP phenotypic variance in the Illumina cohorts , and the 31 DBP signature genes from the FHS ( 22 of these 31 genes were measured in the Illumina cohorts ) in aggregate explained 5 . 3% of DBP phenotypic variance in the Illumina cohorts . These results suggest that in contrast to common genetic variants identified by BP GWAS , which explain in aggregate only about 1% of inter-individual BP variation [3] , changes in gene expression levels explains a considerably larger proportion of phenotypic variance in BP . A meta-analysis of differential expression across all six cohorts revealed 34 differentially expressed BP genes at p<0 . 05 ( Bonferroni corrected for 7717 genes that were measured and passed quality control in the FHS and Illumina cohorts ) , including 21 for SBP , 20 for DBP , and 5 for HTN ( Table 2 and S2 Fig . ) . All of the 34 differentially expressed BP signature genes showed directional consistency in the FHS and the Illumina cohorts ( Table 2 ) . The 34 BP signature genes included all 13 genes that were cross-validated between the FHS and the Illumina cohorts . Of the 34 BP signature genes , 27 were positively correlated with BP and only 7 genes were negatively correlated . MYADM and SLC31A2 were top signature genes for SBP , DBP , and HTN . At FDR<0 . 2 , 224 unique genes were differentially expressed in relation BP phenotypes including 142 genes for SBP , 137 for DBP , and 45 for HTN ( details are reported in the S1–S2 Text , and S3–S5 Table ) . We used gene set enrichment analysis ( GSEA ) to identify the biological process and pathways associated with gene expression changes in relation to SBP , DBP , and HTN in order to better understand the biological themes within the data . As shown in Table 3 , the GSEA of genes whose expression was positively associated with BP showed enrichment for antigen processing and presentation ( p<0 . 0001 ) , apoptotic program ( p<0 . 0001 ) , inflammatory response ( p<0 . 0001 ) , and oxidative phosphorylation ( p = 0 . 0018 ) . The negatively associated genes showed enrichment for nucleotide metabolic process ( p<0 . 0001 ) , positive regulation of cellular metabolic process ( p<0 . 0001 ) , and positive regulation of DNA dependent transcription ( p = 0 . 0021 ) . Among the 34 BP signatures genes from the meta-analysis of all 6 studies , 33 were found to have cis-eQTLs and 26 had trans-eQTLs ( Fig . 2A and S2 Table ) based on whole blood profiling [12 , 13] . Of these , six master trans-eQTLs mapped to either five or six BP signature genes ( no master cis-eQTL was identified ) . Five master trans-eQTLs ( rs653178 , rs3184504 , rs10774625 , rs11065987 , and rs17696736 ) were located on chromosome 12q24 within the same linkage disequilibrium ( LD ) block ( r2 >0 . 8 , Fig . 2B ) . We retrieved a peak cis- and trans-eQTL for each BP signature gene . The peak cis-eQTL explained 0 . 2–20% of the variance in the corresponding transcript levels , in contrast , the peak trans-eQTL accounted for very little ( 0 . 02–2% ) of the corresponding transcript variance . Westra et al . also reported a similar small proportion of variance in transcript levels explained by trans-eQTLs [12] . We then linked the cis- and trans-eQTLs of the 34 BP signature genes with BP GWAS results from the ICBP Consortium [3] and the NHGRI GWAS Catalog [14] ( Fig . 2 and S2 Table ) . We did not find any cis-eQTLs for the top BP signature genes that also were associated with BP in the ICBP GWAS [3] . However , the 6 master trans-eQTLs were all associated with BP at p<5e-8 in the ICBP GWAS [3] and were associated with multiple complex diseases or traits ( Table 4 ) . For example , rs3184504 , a nonsynonymous SNP in SH2B3 that was associated in GWAS with BP , coronary heart disease , hypothyroidism , rheumatoid arthritis , and type 1 diabetes [12] , is a trans-eQTL for 6 of our 34 BP signature genes from the meta-analysis ( FOS , MYADM , PP1R15A , TAGAP , S100A10 , and FGBP2; Fig . 2A-B and Table 4 ) . These 6 genes are all highly expressed in neutrophils , and their expression levels are correlated significantly ( average r2 = 0 . 04 , p<1e-16 ) . rs653178 , intronic to ATXN2 and in perfect LD with rs3184504 ( r2 = 1 ) , also is associated with BP and multiple other diseases in the NHGRI GWAS Catalog [14] . It also is a trans-eQTL for the same 6 BP signature genes ( Table 4 ) . These two SNPs are cis-eQTLs for expression SH2B3 in whole blood ( FDR<0 . 05 ) , but not for ATXN2 ( FDR = 0 . 4 ) . We found that the expression of SH2B3 is associated with expression of MYADM , PP1R15A , and TAGAP ( at Bonferroni corrected p<0 . 05 ) , but not with FOS , S100A10 , or FGBP2 . The expression of ATXN2 was associated with expression of 5 of the 6 genes ( PP1R15A was not associated ) . S3 Fig . shows the coexpression levels of the eight genes that were cis- or trans- associated with rs3184504 and rs653178 genotypes . These results suggest that there may be a pathway or gene co-regulatory mechanism underling BP regulation involving these genes that is driven by this common genetic variant ( rs3184504; minor allele frequency 0 . 47 ) or its proxy SNPs . We further checked whether the cis- or trans-eQTLs for the top 34 BP signature genes are associated with other diseases or traits in the NHGRI GWAS catalog [14] . We identified 12 cis-eQTLs ( for 8 genes ) and 6 trans-eQTLs ( for 6 genes ) that are associated with other diseases or traits in the NHGRI GWAS catalog [14] ( Table 4 ) .
Our meta-analysis of gene expression data from 7017 individuals from six studies identified and characterized whole blood gene expression signatures associated with BP traits . Thirty-four BP signature genes were identified at Bonferroni corrected p<0 . 05 ( 224 genes were identified at FDR<0 . 2 , reported in the S1 Text ) . Thirteen BP signature genes replicated between the FHS and Illumina cohorts . The top BP signature genes identified in the FHS ( 55 genes for SBP and 22 genes for DBP ) explained 5–9% of interindividual variation in BP in the Illumina cohorts on average . Among the 34 BP signature genes ( at Bonferroni corrected p<0 . 05 ) , only FOS [15] and PTGS2 [16] have been previously implicated in hypertension . We did not find literature support for a direct role of the remaining signature genes in BP regulation . However , we found several genes involved in biological functions or processes that are highly related to BP , such as cardiovascular disease ( GZMB , ANXA1 , TMEM43 , FOS , KCNJ2 , PTGS2 , and MCL1 ) , angiogenesis ( VIM and TIPARP ) , and ion channels ( CD97 , ANXA1 , S100A10 , PRF1 , ANTXR2 , SLC31A2 , TIPARP , and KCNJ2 ) . We speculate that these genes may be important for BP regulation , but further experimental validation is needed . Seven of the 34 signature genes , including KCNJ2 , showed negative correlation of expression with BP . KCNJ2 is a member of the potassium inwardly-rectifying channel subfamily; it encodes the inward rectifier K+ channel Kir2 . 1 , and is found in cardiac , skeletal muscle , and nervous tissue [17] . Most outward potassium channels are positively correlated with BP . Loss-of-function mutations in ROMK ( KCNJ1 , the outward potassium channel ) are associated with Bartter's syndrome , and ROMK inhibitors are used in the treatment of hypertension [18 , 19] . Previous studies reported that greater potassium intake is associated with lower blood pressure [20 , 21 , 22 , 23] . These data suggest that KCNJ2 up-regulation may be a means of lowering BP . By linking the BP signature genes with eQTLs and with BP GWAS results , we found several SNPs that are associated with BP in GWAS and that also are trans associated with several of our top BP signature genes . For example , rs3184504 , a non-synonymous SNP located in exon 3 of SH2B3 , is associated in GWAS with BP , coronary heart disease , hypothyroidism , rheumatoid arthritis , and type I diabetes [12] . rs3184504 is a common genetic variant with a minor allele frequency of approximately 0 . 47; the rs3184504-T allele is associated with an increment of 0 . 58 mm Hg in SBP and of 0 . 48 mm Hg in DBP [2] . rs3184504 is a cis-eQTL for SH2B3 , expression of this gene was not associated with BP or hypertension in our data . However , rs3184504 also is a trans-eQTL for 6 of our 34 BP signature genes: FOS , MYADM , PP1R15A , TAGAP , S100A10 , and FGBP2 . These 6 genes are highly expressed in neutrophils [12] , and are coexpressed . Prior studies have suggested an important role of neutrophils in BP regulation [24] . We speculate that these 6 BP signature genes , all driven by the same BP-associated eQTL , point to a critical and previously unrecognized mechanism involved in BP regulation . Further experimental validation is needed . One limitation of our study is the use of whole blood derived RNA for transcriptomic profiling . GSEA showed that the top enriched biological processes for the differentially expressed BP genes include inflammatory response . Numerous studies have shown links between inflammation and hypertension [25 , 26 , 27] . The top ranked genes in inflammatory response categories provide a guide for further experimental work to recognize the contributions of inflammation to alterations in BP regulation . We speculate that using similar approaches in other tissues might identify additional differentially expressed BP signature genes . In conclusion , we conducted a meta-analysis of global gene expression profiles in relation to BP and identified a number of credible gene signatures of BP and hypertension . Our integrative analysis of GWAS and gene expression in relation to BP can help to uncover the genetic and genomic architecture of BP regulation; the BP signature genes we identified may represent an early step toward improvements in the detection of susceptibility , and in the prevention and treatment of hypertension .
This investigation included six studies ( the Framingham Heart Study ( FHS ) , the Estonian Biobank ( EGCUT ) , the Rotterdam Study ( RS ) [8] , the InCHIANTI Study , the Cooperative Health Research in the Region of Augsburg ( KORA F4 ) Study [9] , and the Study of Health in Pomerania ( SHIP-TREND ) [10] , each of which conducted genome-wide genotyping , mRNA expression profiling , and had extensive BP phenotype data . Each of the six studies followed the recommendations of the Declaration of Helsinki . The FHS: Systems Approach to Biomarker Research ( SABRe ) in cardiovascular disease is approved under the Boston University Medical Center’s protocol H-27984 . Ethical approval of EGCUT was granted by the Research Ethics Committee of the University of Tartu ( UT REC ) . Ethical approval of the InCHIANTI study was granted by the Instituto Nazionale Riposo e Cura Anziani institutional review board in Italy . Ethical approval of RS was granted by the medical ethics committee of the Erasmus Medical Center . The study protocol of SHIP-TREND was approved by the medical ethics committee of the University of Greifswald . KORA F4 is a population-based survey in the region of Augsburg in Southern Germany which was performed between 2006 and 2008 . KORA F4 was approved by the local ethical committees . Informed consent was obtained from each study participant . Hypertension ( HTN ) was defined as SBP ≥140 mm Hg or DBP ≥90 mm Hg . We excluded individuals receiving anti-hypertensive treatment because of the possibility that some of the differentially expressed genes we identified would reflect treatment effects . The eligible study sample included 7017 individuals: 3679 from FHS , 972 from EGCUT , 604 from RS , 597 from InCHIANTI , 565 from KORA F4 , and 600 from SHIP-TREND . RNA was isolated from whole blood samples that were collected in PaxGene tubes ( PreAnalytiX , Hombrechtikon , Switzerland ) in FHS , RS , InCHIANTI , KORA F4 and SHIP-TREND , and in Blood RNA Tubes ( Life Technologies , NY , USA ) in EGCUT . Gene expression in the FHS samples used the Affymetrix Exon Array ST 1 . 0 . EGCUT , RS , InCHANTI , KORA F4 , and SHIP-TREND used the Illumina HT12v3 ( EGCUT , InCHANTI , KORA F4 , and SHIP-TREND ) or HT12v4 ( RS ) array . Raw data from gene expression profiling are available online ( FHS [http://www . ncbi . nlm . nih . gov/gap; accession number phs000007] , EGCUT [GSE48348] , RS [GSE33828] , InCHIANTI [GSE48152] , KORA F4 [E-MTAB-1708] and SHIP-TREND [GSE36382] ) . The details of sample collection , microarrays , and data processing and normalization in each cohort are provided in the S2 Text . The association of gene expression with BP was analyzed separately in each of the six studies ( Equation 1 ) . A linear mixed model was used in the FHS in order to account for family structure . Linear regression models were used in the other five studies . In each study , gene expression level , denoted by geneExp , was included as the dependent variable , and explanatory variables included blood pressure phenotypes ( SBP , DBP , and HTN ) , and covariates included age , sex , body mass index ( BMI ) , cell counts , and technical covariates . A separate regression model was fitted for each gene . The general formula is shown below , and the details of analyses for each study are provided in the S2 Text and S6 Table . geneExp=BP+∑j=1mcovariates The overall analysis framework is provided in S1 Fig . . We first identified differentially expressed genes associated with BP ( BP signature genes ) in the FHS samples ( Set 1 ) and attempted replication in the meta-analysis results from the Illumina cohorts ( Set 2 , see Methods , Meta-analysis ) . We next identified BP signature genes in the Illumina cohorts ( Set 2 ) , and then attempted replication in the FHS samples ( Set 1 ) . The significance threshold for pre-selecting BP signature genes in discovery was at Bonferroni corrected p = 0 . 05 ( in FHS , corrected for 17 , 318 measured genes [17 , 873 transcripts] , and in illumina cohorts , corrected for 12 , 010 measured genes [14 , 222 transcripts] that passed quality control ) . Replication was established at Bonferroni corrected p = 0 . 05 , correcting for the number of pre-selected BP signatures genes in the discovery set . We computed the pi1 value to estimate the enrichment of significant p values in the replication set ( the Illumina cohorts ) for BP signatures identified in the discovery set ( the FHS ) by utilizing the R package Qvalue [11] . Pi1 is defined as 1-pi0 . Pi0 value provided by the Qvalue package , represents overall probability that the null hypothesis is true . Therefore , pi1 value represents the proportion of significant results . For genes passing Bonferroni corrected p<0 . 05 in the discovery set for SBP , DBP and HTN , we calculated pi1 values for each gene set in the replication set . We performed meta-analysis of the five Illumina cohorts ( for discovery and replication purposes ) , and then performed meta-analysis of all six cohorts . An inverse variance weighted meta-analysis was conducted using fixed-effects or random-effects models by the metagen ( ) function in the R package Meta ( http://cran . r-project . org/web/packages/meta/index . html ) . At first , we tested heterogeneity for each gene using Cochran’s Q statistic . If the heterogeneity p value is significant ( p<0 . 05 ) , we will use a random-effects model for the meta-analysis , otherwise use a fixed-effects model . The Benjamini-Hochberg ( BH ) method [28] was used to calculate FDR for differentially expressed genes in relation to BP following the meta-analysis of all six cohorts . We also used a more stringent threshold to define BP signature genes by utilizing p<6 . 5e-6 ( Bonferroni correction for 7717 unique genes [7810 transcript] based on the overlap of FHS and illumina cohort interrogated gene sets ) . To estimate the proportion of variances in SBP or DBP explained by a group of differentially expressed BP signature genes ( gene 1 , gene 2 , … , gene n ) , we used the following two models: Full model: BP=∑i=1ngenei+∑j=1mcovariates Null model: BP=∑j=1mcovariates The proportion of variance in BP attributable to the group of differentially expressed BP signature genes ( hBP_sig2 ) was calculated as: hBP_sig2=max ( 0 , σG . null2+σerr . null2−σG . full2−σerr . full2σBP2 ) where σBP2 is the total phenotypic variance of SBP or DBP , σG . full2 and σerr . full2 are the variance and error variance when modeling with the tested group of gene expression traits ( gene 1 , gene 2 , … , gene n ) , and σG . null2 and σerr . null2 are the variance and error variance when modeling without the tested group of gene expression traits . The proportion of the variance in BP phenotypes attributable to the FHS BP signature genes was estimated in the five Illumina cohorts , respectively , and then the average proportion values were reported . In turn , the proportion of the variance in BP phenotypes attributable to the Illumina BP signature genes was estimated in the FHS . SNPs associated with altered gene expression ( i . e . eQTLs ) were identified using genome-wide genotype and gene expression data in all available FHS samples ( n = 5257 ) at FDR<0 . 1 ( Joehanes R , submitted , 2014 , and a brief summary of methods and results are provided in the S2 Text ) . A cis-eQTL was defined as an eQTL within 1 megabase ( MB ) flanking the gene . Other eQTLs were defined as trans-eQTLs . We combined the eQTL list generated in the FHS with the eQTLs generated by meta-analysis of seven other studies ( n = 5300 ) that were also based on whole blood expression[12] . For every BP signature gene , we estimated the proportion of variance in the transcript attributable to the corresponding cis- or trans-eQTLs ( heQTL2 ) using the formula: heQTL2=max ( 0 , σeQTL . null2+σerr . null2−σeQTL . full2−σerr . full2σgene2 ) where σgene2 was the total phenotypic variance of a gene expression trait; σeQTL . full2 and σerr . full2 were the variance and the residual error , respectively , when modeling with the tested eQTL; σeQTL . null2 and σerr . null2 were the variance and the residual error when modeling without the tested eQTL . In order to understand the biological themes within the global gene expression changes in relation to BP , we performed gene set enrichment analysis[29] to test for enrichment of any gene ontology ( GO ) biology process[30] or KEGG pathways[31] . “Metric for ranking gene” parameters were configured to the beta value of the meta-analysis , in order to look at the top enriched functions for BP associated up-regulated and down-regulated gene expression changes respectively . One thousand random permutations were conducted and the significance level was set at FDR≤ 0 . 25 to allow for exploratory discovery [29] . Steering Committee ( alphabetical ) Gonçalo Abecasis , Murielle Bochud , Mark Caulfield ( co-chair ) , Aravinda Chakravarti , Dan Chasman , Georg Ehret ( co-chair ) , Paul Elliott , Andrew Johnson , Louise Wain , Martin Larson , Daniel Levy ( co-chair ) , Patricia Munroe ( co-chair ) , Christopher Newton-Cheh ( co-chair ) , Paul O'Reilly , Walter Palmas , Bruce Psaty , Kenneth Rice , Albert Smith , Harold Snider , Martin Tobin , Cornelia Van Duijn , Germaine Verwoert . Members Georg B . Ehret1 , 2 , 3 , Patricia B . Munroe4 , Kenneth M . Rice5 , Murielle Bochud2 , Andrew D . Johnson6 , 7 , Daniel I . Chasman8 , 9 , Albert V . Smith10 , 11 , Martin D . Tobin12 , Germaine C . Verwoert13 , 14 , 15 , Shih-Jen Hwang6 , 16 , 7 , Vasyl Pihur1 , Peter Vollenweider17 , Paul F . O'Reilly18 , Najaf Amin13 , Jennifer L Bragg-Gresham19 , Alexander Teumer20 , Nicole L . Glazer21 , Lenore Launer22 , Jing Hua Zhao23 , Yurii Aulchenko13 , Simon Heath24 , Siim Sõber25 , Afshin Parsa26 , Jian'an Luan23 , Pankaj Arora27 , Abbas Dehghan13 , 14 , 15 , Feng Zhang28 , Gavin Lucas29 , Andrew A . Hicks30 , Anne U . Jackson31 , John F Peden32 , Toshiko Tanaka33 , Sarah H . Wild34 , Igor Rudan35 , 36 , Wilmar Igl37 , Yuri Milaneschi33 , Alex N . Parker38 , Cristiano Fava39 , 40 , John C . Chambers18 , 41 , Ervin R . Fox42 , Meena Kumari43 , Min Jin Go44 , Pim van der Harst45 , Wen Hong Linda Kao46 , Marketa Sjögren39 , D . G . Vinay47 , Myriam Alexander48 , Yasuharu Tabara49 , Sue Shaw-Hawkins4 , Peter H . Whincup50 , Yongmei Liu51 , Gang Shi52 , Johanna Kuusisto53 , Bamidele Tayo54 , Mark Seielstad55 , 56 , Xueling Sim57 , Khanh-Dung Hoang Nguyen1 , Terho Lehtimäki58 , Giuseppe Matullo59 , 60 , Ying Wu61 , Tom R . Gaunt62 , N . Charlotte Onland-Moret63 , 64 , Matthew N . Cooper65 , Carl G . P . Platou66 , Elin Org25 , Rebecca Hardy67 , Santosh Dahgam68 , Jutta Palmen69 , Veronique Vitart70 , Peter S . Braund71 , 72 , Tatiana Kuznetsova73 , Cuno S . P . M . Uiterwaal63 , Adebowale Adeyemo74 , Walter Palmas75 , Harry Campbell35 , Barbara Ludwig76 , Maciej Tomaszewski71 , 72 , Ioanna Tzoulaki77 , 78 , Nicholette D . Palmer79 , CARDIoGRAM consortium80 , CKDGen Consortium80 , KidneyGen Consortium80 , EchoGen consortium80 , CHARGE-HF consortium80 , Thor Aspelund10 , 11 , Melissa Garcia22 , Yen-Pei C . Chang26 , Jeffrey R . O'Connell26 , Nanette I . Steinle26 , Diederick E . Grobbee63 , Dan E . Arking1 , Sharon L . Kardia81 , Alanna C . Morrison82 , Dena Hernandez83 , Samer Najjar84 , 85 , Wendy L . McArdle86 , David Hadley50 , 87 , Morris J . Brown88 , John M . Connell89 , Aroon D . Hingorani90 , Ian N . M . Day62 , Debbie A . Lawlor62 , John P . Beilby91 , 92 , Robert W . Lawrence65 , Robert Clarke93 , Rory Collins93 , Jemma C Hopewell93 , Halit Ongen32 , Albert W . Dreisbach42 , Yali Li94 , J H . Young95 , Joshua C . Bis21 , Mika Kähönen96 , Jorma Viikari97 , Linda S . Adair98 , Nanette R . Lee99 , Ming-Huei Chen100 , Matthias Olden101 , 102 , Cristian Pattaro30 , Judith A . Hoffman Bolton103 , Anna Köttgen104 , 103 , Sven Bergmann105 , 106 , Vincent Mooser107 , Nish Chaturvedi108 , Timothy M . Frayling109 , Muhammad Islam110 , Tazeen H . Jafar110 , Jeanette Erdmann111 , Smita R . Kulkarni112 , Stefan R . Bornstein76 , Jürgen Grässler76 , Leif Groop113 , 114 , Benjamin F . Voight115 , Johannes Kettunen116 , 126 , Philip Howard117 , Andrew Taylor43 , Simonetta Guarrera60 , Fulvio Ricceri59 , 60 , Valur Emilsson118 , Andrew Plump118 , Inês Barroso119 , 120 , Kay-Tee Khaw48 , Alan B . Weder121 , Steven C . Hunt122 , Yan V . Sun81 , Richard N . Bergman123 , Francis S . Collins124 , Lori L . Bonnycastle124 , Laura J . Scott31 , Heather M . Stringham31 , Leena Peltonen119 , 125 , 126 , 127 , Markus Perola125 , Erkki Vartiainen125 , Stefan-Martin Brand128 , 129 , Jan A . Staessen73 , Thomas J . Wang6 , 130 , Paul R . Burton12 , 72 , Maria Soler Artigas12 , Yanbin Dong131 , Harold Snieder132 , 131 , Xiaoling Wang131 , Haidong Zhu131 , Kurt K . Lohman133 , Megan E . Rudock51 , Susan R Heckbert134 , 135 , Nicholas L Smith134 , 136 , 135 , Kerri L Wiggins137 , Ayo Doumatey74 , Daniel Shriner74 , Gudrun Veldre25 , 138 , Margus Viigimaa139 , 140 , Sanjay Kinra141 , Dorairajan Prabhakaran142 , Vikal Tripathy142 , Carl D . Langefeld79 , Annika Rosengren143 , Dag S . Thelle144 , Anna Maria Corsi145 , Andrew Singleton83 , Terrence Forrester146 , Gina Hilton1 , Colin A . McKenzie146 , Tunde Salako147 , Naoharu Iwai148 , Yoshikuni Kita149 , Toshio Ogihara150 , Takayoshi Ohkubo149 , 151 , Tomonori Okamura148 , Hirotsugu Ueshima152 , Satoshi Umemura153 , Susana Eyheramendy154 , Thomas Meitinger155 , 156 , H . -Erich Wichmann157 , 158 , 159 , Yoon Shin Cho44 , Hyung-Lae Kim44 , Jong-Young Lee44 , James Scott160 , Joban S . Sehmi160 , 41 , Weihua Zhang18 , Bo Hedblad39 , Peter Nilsson39 , George Davey Smith62 , Andrew Wong67 , Narisu Narisu124 , Alena Stančáková53 , Leslie J . Raffel161 , Jie Yao161 , Sekar Kathiresan162 , 27 , Chris O'Donnell163 , 27 , 9 , Stephen M . Schwartz134 , M . Arfan Ikram13 , 15 , W . T . Longstreth Jr . 164 , Thomas H . Mosley165 , Sudha Seshadri166 , Nick R . G . Shrine12 , Louise V . Wain12 , Mario A . Morken124 , Amy J . Swift124 , Jaana Laitinen167 , Inga Prokopenko51 , 168 , Paavo Zitting169 , Jackie A . Cooper69 , Steve E . Humphries69 , John Danesh48 , Asif Rasheed170 , Anuj Goel32 , Anders Hamsten171 , Hugh Watkins32 , Stephan J . L . Bakker172 , Wiek H . van Gilst45 , Charles S . Janipalli47 , K . Radha Mani47 , Chittaranjan S . Yajnik112 , Albert Hofman13 , Francesco U . S . Mattace-Raso13 , 14 , Ben A . Oostra173 , Ayse Demirkan13 , Aaron Isaacs13 , Fernando Rivadeneira13 , 14 , Edward G Lakatta174 , Marco Orru175 , 176 , Angelo Scuteri174 , Mika Ala-Korpela177 , 178 , 179 , Antti J Kangas177 , Leo-Pekka Lyytikäinen58 , Pasi Soininen177 , 178 , Taru Tukiainen180 , 181 , 177 , Peter Würtz177 , 18 , 180 , Rick Twee-Hee Ong56 , 57 , 182 , Marcus Dörr183 , Heyo K . Kroemer184 , Uwe Völker20 , Henry Völzke185 , Pilar Galan186 , Serge Hercberg186 , Mark Lathrop24 , Diana Zelenika24 , Panos Deloukas119 , Massimo Mangino28 , Tim D . Spector28 , Guangju Zhai28 , James F . Meschia187 , Michael A . Nalls83 , Pankaj Sharma188 , Janos Terzic189 , M . J . Kranthi Kumar47 , Matthew Denniff71 , Ewa Zukowska-Szczechowska190 , Lynne E . Wagenknecht79 , F . Gerald R . Fowkes191 , Fadi J . Charchar192 , Peter E . H . Schwarz193 , Caroline Hayward70 , Xiuqing Guo161 , Charles Rotimi74 , Michiel L . Bots63 , Eva Brand194 , Nilesh J . Samani71 , 72 , Ozren Polasek195 , Philippa J . Talmud69 , Fredrik Nyberg68 , 196 , Diana Kuh67 , Maris Laan25 , Kristian Hveem66 , Lyle J . Palmer197 , 198 , Yvonne T . van der Schouw63 , Juan P . Casas199 , Karen L . Mohlke61 , Paolo Vineis200 , 60 , Olli Raitakari201 , Santhi K . Ganesh202 , Tien Y . Wong203 , 204 , E Shyong Tai205 , 57 , 206 , Richard S . Cooper54 , Markku Laakso53 , Dabeeru C . Rao207 , Tamara B . Harris22 , Richard W . Morris208 , Anna F . Dominiczak209 , Mika Kivimaki210 , Michael G . Marmot210 , Tetsuro Miki49 , Danish Saleheen170 , 48 , Giriraj R . Chandak47 , Josef Coresh211 , Gerjan Navis212 , Veikko Salomaa125 , Bok-Ghee Han44 , Xiaofeng Zhu94 , Jaspal S . Kooner160 , 41 , Olle Melander39 , Paul M Ridker8 , 213 , 9 , Stefania Bandinelli214 , Ulf B . Gyllensten37 , Alan F . Wright70 , James F . Wilson34 , Luigi Ferrucci33 , Martin Farrall32 , Jaakko Tuomilehto215 , 216 , 217 , 218 , Peter P . Pramstaller30 , 219 , Roberto Elosua29 , 220 , Nicole Soranzo119 , 28 , Eric J . G . Sijbrands13 , 14 , David Altshuler221 , 115 , Ruth J . F . Loos23 , Alan R . Shuldiner26 , 222 , Christian Gieger157 , Pierre Meneton223 , Andre G . Uitterlinden13 , 14 , 15 , Nicholas J . Wareham23 , Vilmundur Gudnason10 , 11 , Jerome I . Rotter161 , Rainer Rettig224 , Manuela Uda175 , David P . Strachan50 , Jacqueline C . M . Witteman13 , 15 , Anna-Liisa Hartikainen225 , Jacques S . Beckmann105 , 226 , Eric Boerwinkle227 , Ramachandran S . Vasan6 , 228 , Michael Boehnke31 , Martin G . Larson6 , 229 , Marjo-Riitta Järvelin18 , 230 , 231 , 232 , 233 , Bruce M . Psaty21 , 135* , Gonçalo R Abecasis19* , Aravinda Chakravarti1 , Paul Elliott18 , 233* , Cornelia M . van Duijn13 , 234* , Christopher Newton-Cheh27 , 115 , Daniel Levy6 , 16 , 7 , Mark J . Caulfield4 , Toby Johnson4 Affiliations
|
The focus of blood pressure ( BP ) GWAS has been the identification of common DNA sequence variants associated with the phenotype; this approach provides only one dimension of molecular information about BP . While it is a critical dimension , analyzing DNA variation alone is not sufficient for achieving an understanding of the multidimensional complexity of BP physiology . The top loci identified by GWAS explain only about 1 percent of inter-individual BP variability . In this study , we performed a meta-analysis of gene expression profiles in relation to BP and hypertension in 7017 individuals from six studies . We identified 34 differentially expressed genes for BP , and discovered that the top BP signature genes explain 5%–9% of BP variability . We further linked BP gene expression signature genes with BP GWAS results by integrating expression associated SNPs ( eSNPs ) and discovered that one of the top BP loci from GWAS , rs3184504 in SH2B3 , is a trans regulator of expression of 6 of the top 34 BP signature genes . Our study , in conjunction with prior GWAS , provides a deeper understanding of the molecular and genetic basis of BP regulation , and identifies several potential targets and pathways for the treatment and prevention of hypertension and its sequelae .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
A Meta-analysis of Gene Expression Signatures of Blood Pressure and Hypertension
|
In this study , we evaluated the long-term efficacy of a two-component subunit vaccine against Trypanosoma cruzi infection . C57BL/6 mice were immunized with TcG2/TcG4 vaccine delivered by a DNA-prime/Protein-boost ( D/P ) approach and challenged with T . cruzi at 120 or 180 days post-vaccination ( dpv ) . We examined whether vaccine-primed T cell immunity was capable of rapid expansion and intercepting the infecting T . cruzi . Our data showed that D/P vaccine elicited CD4+ ( 30-38% ) and CD8+ ( 22-42% ) T cells maintained an effector phenotype up to 180 dpv , and were capable of responding to antigenic stimulus or challenge infection by a rapid expansion ( CD8>CD4 ) with type 1 cytokine ( IFNγ+ and TFNα+ ) production and cytolytic T lymphocyte ( CTL ) activity . Subsequently , challenge infection at 120 or 180 dpv , resulted in 2-3-fold lower parasite burden in vaccinated mice than was noted in unvaccinated/infected mice . Co-delivery of IL-12- and GMCSF-encoding expression plasmids provided no significant benefits in enhancing the anti-parasite efficacy of the vaccine-induced T cell immunity . Booster immunization ( bi ) with recombinant TcG2/TcG4 proteins 3-months after primary vaccine enhanced the protective efficacy , evidenced by an enhanced expansion ( 1 . 2-2 . 8-fold increase ) of parasite-specific , type 1 CD4+ and CD8+ T cells and a potent CTL response capable of providing significantly improved ( 3-4 . 5-fold ) control of infecting T . cruzi . Further , CD8+T cells in vaccinated/bi mice were predominantly of central memory phenotype , and capable of responding to challenge infection 4-6-months post bi by a rapid expansion to a poly-functional effector phenotype , and providing a 1 . 5-2 . 3-fold reduction in tissue parasite replication . We conclude that the TcG2/TcG4 D/P vaccine provided long-term anti-T . cruzi T cell immunity , and bi would be an effective strategy to maintain or enhance the vaccine-induced protective immunity against T . cruzi infection and Chagas disease .
Chagas disease is prevalent in almost all Latin American countries , including Mexico and Central America [1] . Currently , ~11–18 million individuals are infected worldwide , and ~13 , 000 children and adults die annually because of the clinical complications of T . cruzi-induced heart disease and lack of effective treatments [2] . The vectorial , autochthonous , and congenital transmission of T . cruzi exists in the United States , where >300 , 000 infected individuals can potentially transfer infection through blood or organ donation [3–5] . When considered from a global perspective , Chagas disease represents the third greatest tropical disease burden after malaria and schistosomiasis [6] . Before setting the goal of vaccine development against any disease , an important question is whether vaccination is an economically viable approach with desirable health benefits . With regard to T . cruzi infection , the research community has pushed for a vaccine that can achieve complete parasite elimination from the host . However , several studies , including our published reports ( reviewed in [7] ) , testing the efficacy of subunit vaccines have resulted in findings that vaccine-induced immunity can provide a reduction in tissue parasite burden associated with variable degrees of control of acute or chronic disease symptoms . The vaccine mediated control of infection and disease in experimental studies generally resembled that noted in 60–70% of the chagasic patients that remained seropositive and maintained residual parasites for their entire lives , but did not develop a clinically symptomatic form of the disease [2] . Further , recent computer simulation modeling of the impact of a prophylactic vaccine for Chagas disease showed that a vaccine would provide net cost savings ( along with health benefits ) , even when the risk of infection is only 1% , vaccine efficacy is only 25% , and the cost of a vaccine is US$20 or lower [8] . Thus , it is ethically appropriate to consider a satisfactory vaccination goal to reduce the frequency and severity of clinical disease by decreasing the extent of persistent parasite burden; and accordingly , continuing efforts towards developing a vaccine against T . cruzi infection and Chagas disease are economically justifiable . We have employed a computational/bioinformatics approach for unbiased screening of the T . cruzi genome database and identification of 11 potential candidates [9 , 10] . Through rigorous analysis over a period of several years , we determined that three candidates ( TcG1 , TcG2 , TcG4 ) were maximally relevant for vaccine development [11] . These candidates were highly conserved in clinically relevant T . cruzi strains , expressed ( mRNA/protein ) in infective trypomastigote and intracellular amastigote stages of T . cruzi , and recognized by immunoglobulins and CD8+T cells in multiple T . cruzi-infected hosts [10 , 11] . We have examined the protective efficacy of TcG1 , TcG2 and TcG4 ( individually or in combination ) in mice . Our data showed that co-delivery of the three antigens elicited additive immunity and protection from T . cruzi infection than was noted with individual candidate antigens [11] . Delivery of the 3-component vaccine by a DNA-prime/DNA-boost approach was less effective than the heterologous DNA-prime/protein-boost ( D/P ) approach in eliciting protective immunity [11–13] . Mice challenged with T . cruzi immediately after immunization with the 3-component D/P vaccine were capable of controlling 90–97% of the acute parasitemia and tissue parasite burden , and , subsequently , inflammatory infiltrate and tissue fibrosis were particularly absent in the heart and skeletal muscle of vaccinated mice [13] . In this study , we have sought to determine the long-term efficacy of the subunit vaccine against T . cruzi infection . We included TcG2 and TcG4 in the vaccine , as these antigens were most potent in eliciting parasite-specific antibody and CD8+T cell immunity [11–13] . Mice were immunized with TcG2/TcG4 vaccine delivered by the D/P approach , and we examined whether a ) the 2-component D/P vaccine primed TH1 CD4+T cells and generated a stable pool of CD8+T memory cells , and b ) the vaccine-primed T cells were capable of rapid expansion and intercepting the infecting T . cruzi . We also determined if c ) the 2-component D/P vaccine primed immunity is enhanced by co-delivery of IL-12 and GM-CSF cytokine adjuvants , or d ) a booster immunization ( bi ) several months after the second vaccine dose was effective in providing better protection from T . cruzi infection .
D/P vaccination resulted in 60–70% expansion of splenic cell number , observed at day 14 and 120 post-vaccination ( S1 Table ) . To examine the T cell profile primed by the TcG2- and TcG4-encoding D/P vaccine , splenocytes from immunized mice were submitted to flow cytometry analysis before and after in vitro stimulation with recombinant antigens . Splenocytes were gated for CD4+ and CD8+ T cells and analyzed for surface markers of effector/effector memory ( TEM , CD44+CD62L- ) and central memory ( TCM , CD44+CD62L+ ) phenotype . The immediate early T cell response to D/P vaccine , measured at 14 days post vaccination ( dpv ) , was evidenced by a significant increase in ex vivo levels of CD4+ ( 36–39% of total , 3 . 8–4 . 7-fold ) and CD8+ ( 34–43% of total , 2 . 5–3 . 3-fold ) TEM and TCM cells , respectively , in D/P vaccinated mice when compared to that noted in non-vaccinated controls ( Fig 1A and 1B , p<0 . 01 ) . The vaccine-primed CD4+ and CD8+ T cells exhibited antigen-specific activation and proliferation ( Ki67+ ) , and were positive for IFNγ ( 10–12% and 18–20% , respectively ) and TNFα ( 3 . 2–4 . 4% and 7 . 6–9 . 2% , respectively ) cytokines ( Fig 1C and 1D , p<0 . 05–0 . 01 ) . Transport of CD107a/b integral membrane proteins to the plasma membrane of effector T cells is required for the cytolytic activity mediated by perforin and granzyme and the release of IFNγ which exerts pleiotropic effects to suppress intracellular pathogens . Flow cytometry studies showed a high frequency of the vaccine-primed CD8+T cells were CD107a+IFNγ+perforin+ ( 30–34% , Fig 1E , p<0 . 001 ) . Next , we examined the stability and effector phenotype of T cells in vaccinated mice at 120 dpv . The ex vivo frequency of CD4+ TEM and TCM ( 28–35% ) was slightly decreased , while that of CD8+ TEM cells ( 48–52% ) was increased , with a simultaneous decline in CD8+ TCM cells ( 22–26% ) in vaccinated mice harvested at 120 dpv ( compare Fig 2A and 2B with Fig 1A and 1B ) . The vaccine-elicited long-lived T cells proliferated upon antigenic stimulation ( Ki67+ , insets in Fig 2A and 2B ) , though the frequency of cytokine-producing CD4+T cells ( Ki67+IFNγ+: 6–8% versus 10–12%; Ki67+TNFα+: 2 . 2–2 . 4% versus 3–4% ) and CD8+T cells ( Ki67+IFNγ+: 12–14% versus 18–20%; Ki67+TNFα+: 3 . 6–5 . 2% versus 8–9% ) at 120 dpv , as compared to that noted at 14 dpv , was significantly decreased ( compare Fig 2C and 2D with Fig 1C and 1D ) . Yet , 62–68% of the IFNγ producing CD4+ and CD8+ T cells were CD44+CD62L ( TEM phenotype , Fig 2E and 2F , p<0 . 01–0 . 001 ) , and 25% and 16–18% of the CD8+IFNγ+ T cells were CD107a+ and CD107a+perforin+ , respectively , indicating their cytolytic phenotype ( Fig 1G , p<0 . 05–0 . 01 ) . Together , the data presented in Figs 1 and 2 suggested that the 2-component D/P vaccine primed CD4+ and CD8+ TEM/TCM cells were long-lived , and responded to antigenic stimulus with type 1 cytokine production and CTL activity . Co-delivery of cytokine adjuvants did not change the frequency or phenotype of the vaccine-primed T cells in immunized mice . No antigen-specific expansion of CD4+ and CD8+ T cells was observed in control mice given empty vector or cytokine adjuvants only . To determine if the vaccine-elicited , long-lived T cells were responsive to T . cruzi , mice were challenged at 120 dpv and harvested at 10 days pi . Irrespective of vaccination status , all mice exhibited a significant ( 15-20-fold ) expansion of splenic cell number , the maximum expansion being observed in vaccinated/infected mice ( S1 Table ) . Ex vivo flow cytometry analysis of splenocytes showed the expansion of proliferating ( Ki67+ ) CD4+ and CD8+ TEM cells by 1 . 3–1 . 6-fold in response to T . cruzi infection in vaccinated mice ( Fig 3A and 3B and insets , compare with Fig 2A and 2B , p<0 . 05–0 . 01 ) . Further , vaccinated mice exhibited up to 57% , 29% and 79% increase in antigen-specific IFNγ+ , TNFα+ , and IFNγ+TNFα+ CD4+T cells , respectively , following challenge infection ( compare Fig 3C with 2C ) ; and a majority ( 70–72% ) of the IFNγ+CD4+T cells were proliferative ( Ki67+ ) with effector ( CD44+CD62L- ) phenotype ( Fig 3C and 3E , p<0 . 05–0 . 01 ) . An expansion of vaccine-induced CD8+T cells in response to challenge infection was evidenced by a >2-fold increase in IFNγ+ and TNFα+ and a 4-fold increase in IFNγ+TNFα+ CD8+T cells ( compare Fig 3D with Fig 2D , p<0 . 05–0 . 01 ) ; of which 80–84% exhibited proliferative ( Ki67+ ) and effector ( CD44+CD62L- ) phenotype ( Fig 3D and 3F , p<0 . 01 ) . Likewise , vaccinated mice , in response to challenge infection , exhibited a 40–91% expansion of antigen-specific CD8+IFNγ+T cells that were also CD107a+ or CD107a+perforin+ ( compare Fig 3G with Fig 2G , p<0 . 05 ) . When compared to non-vaccinated/infected mice , vaccinated/infected mice exhibited an 8- to 10-fold higher frequency of antigen-specific IFNγ+ effector ( CD44+CD62L- ) CD4+ and CD8+ T cells and CD8+ cytolytic T cells ( Fig 3E–3G , p<0 . 001 ) . Quantitative PCR showed , respectively , 2 . 2-fold and 2-3-fold decline in peripheral ( S1A Fig ) and tissue ( spleen , heart , skeletal muscle; Fig 3H , p<0 . 05–0 . 01 ) levels of T . cruzi in vaccinated/infected mice when compared to that noted in non-vaccinated/infected controls . Together , these data suggested that the D/P vaccine-elicited , long-lived T cells rapidly expanded ( CD8>CD4 ) with a type 1 effector phenotype in response to challenge infection , and were capable of controlling T . cruzi infection . Next , we determined if a bi would enhance the vaccine elicited T cell immunity and provide better control of T . cruzi infection . For this , mice were immunized with the 2-component D/P vaccine as above , and given a booster dose of TcG2/TcG4 recombinant proteins before challenge infection . The booster-immunized mice exhibited 2-3-fold higher number of splenic cells as compared to non-vaccinated mice ( S1 Table ) . When analyzing splenic T cell frequency , we observed an overall decline in the splenic frequency of CD4+ ( TEM: 21 . 3–22 . 2% versus 28–35% , TCM: 10 . 5–12 . 8% versus 30–34% ) and CD8+ ( TEM: 34 . 5–36 . 8% versus 48–52% , TCM: 14 . 4–16% versus 22–26% ) T cells at 14 days post bi , when compared to that noted before bi ( compare Fig 4A and 4B with Fig 2A and 2B ) . However , splenocytes from vaccinated/bi mice in vitro stimulated with recombinant TcG2/TcG4 proteins exhibited 1 . 6–3 . 4-fold and 1 . 4–3 . 2-fold higher frequencies of the cytokine ( IFNγ and/or TNFα ) producing CD4+ and CD8+ T cells , respectively , as compared to that noted in vaccinated ( but not bi ) mice ( compare Fig 4C and 4D with Fig 2C and 2D , all p<0 . 05–0 . 01 ) . Approximately , 60–83% of the antigen-specific IFNγ- or TNFα-producing CD4+T cells and 80–90% of the IFNγ-producing CD8+T cells exhibited a proliferative ( Ki67+ ) , effector ( CD44+CD62L- ) phenotype ( Fig 4E and 4F , p<0 . 01 ) . Importantly the frequency of CD8+CD107a+ and CD8+CD107a+perforin+ cells secreting IFNγ was increased by 37 . 5–45% post bi ( compare Fig 4G with Fig 2G , p<0 . 05–0 . 01 ) . To validate the antigen-specific cytotoxic activity of the CD8+T lymphocytes in vaccinated/bi mice , spleen cells in vitro stimulated with recombinant antigens were used as effectors and tested for their ability to lyse EL4 cells exposed to GFP+ rMVA encoding TcG2 or TcG4 antigens or empty MVA . A significant level of antigen-specific cytolytic activity ( range 56–69% , effector-to-target cell ratio , 80: 1 or 40:1 ) was evident in vaccinated/bi mice ( Fig 4H , p<0 . 01 . Splenocytes from mice immunized with vector only or cytokines only did not show CTL activity against any of the antigen-sensitized target cells ( Fig 4H ) . When mice were challenged with T . cruzi immediately after bi , a potent expansion of parasite-specific T cells was evidenced by a 1 . 4-2-fold increase in ex vivo frequency of CD4+ and CD8+ TEM cell populations ( compare Fig 5A and 5B with Fig 4A and 4B , p<0 . 05–0 . 01 ) . Further , splenocytes of vaccinated/bi mice , in response to challenge infection , exhibited a strong increase in T . cruzi-specific IFNγ+ or TNFα+ ( 2 . 0–3 . 1-fold ) and IFNγ+TNFα+ ( 4 . 3–7 . 8-fold ) CD4+ and CD8+ T cells , that were also Ki67+ with effector ( CD44+CD62L- ) phenotype ( compare Fig 5C–5F with Fig 4C–4F , p<0 . 05–0 . 01 ) . The D/P-vaccinated/bi mice exhibited a larger expansion of total splenic cells ( S1 Table ) and T cells than was noted in the D/P-vaccinated mice in response to challenge infection , as was evidenced by a significantly higher ( 1 . 6–2 . 8-fold , p<0 . 05–0 . 01 ) frequency of proliferating and non-proliferating IFNγ+ , TNFα+ , and IFNγ+TNFα+ CD4+ and CD8+ T cells ( compare Fig 5 with Fig 3 ) . Subsequently , vaccinated/bi mice exhibited a 2 . 5-fold decline in peripheral ( S1B Fig ) and 3–3 . 8-fold decline in tissue ( spleen , heart and skeletal muscle ) levels of Tc18SrDNA in comparison to that noted in non-vaccinated/infected mice ( Fig 5H , p<0 . 05–0 . 01 ) . Together , the data presented in Figs 4 and 5 , along with that presented in Figs 2 and 3 , suggested that bi was effective in expanding the 2-component D/P vaccine-elicited type 1 cytokine producing T cells and CD8+ CTLs that provided better protection from challenge infection than was observed with 2-component D/P vaccine only . To evaluate the stability and effector phenotype of the T cells elicited by the 2-component D/P vaccine and bi , mice were harvested at 120 or 180 days post bi . As above , the booster-immunized mice continued to maintain 2-3-fold higher number of splenic cells at 120 or 180 days post bi as compared to non-vaccinated mice ( S1 Table ) . The ex vivo frequency of CD4+ TCM cells was increased by 3 . 1–3 . 4-fold and 2 . 1–3 . 2-fold at 120 and 180 days post-bi , respectively , while that of CD4+ TEM cells was increased modestly ( 1 . 4-2-fold ) at 120 days post bi only , when compared to that noted in mice harvested at 14 days post bi ( compare Figs 6A and 7A with Fig 4A ) . Ex vivo CD8+T cells of TEM phenotype at 120 days post bi ( 72–76% , Fig 6B ) contracted by 2 . 4-fold at 180 days post bi ( Fig 7B ) . In contrast , CD8+T cells of TCM phenotype were increased by 1 . 4-fold at 120–180 days post bi compared to that noted in mice harvested at 14 days post bi ( compare Figs 6B and 7B with Fig 4B ) . To determine if long-lived vaccine/bi elicited T cells are functional , splenocytes from vaccinated/bi mice were harvested at 120 or 180 days post bi , in vitro stimulated with recombinant antigens , and analyzed by flow cytometry . We noted a modest decline in the frequency of antigen-specific Ki67+CD4+T cells at 120 days ( 56–58% ) and 180 days ( 52–54% ) post bi when compared to that noted at 14 days ( 60–64% ) post bi ( compare Figs 6A and 7A with Fig 4A ) . The splenic cells from vaccinated mice harvested at 120 days post bi exhibited a ~2-4-fold larger expansion of proliferative ( Ki67+ ) and non-proliferative ( Ki67- ) IFNγ+ and IFNγ+TNFα+ CD4+T cells in response to in vitro antigenic stimulus than was observed with splenic cells from mice harvested at 14 or 180 days post bi ( compare Fig 6C with Figs 7C and 4C , p<0 . 05–0 . 01 ) . The CD8+T cells , at 120 days post bi , exhibited a similar level of antigen-specific proliferative ( Ki67+ ) capacity , but a significantly higher capacity for type 1 cytokine ( IFNγ and/or TNFα ) production and lytic activity ( IFNγ+CD107a+perforin- or IFNγ+CD107a+perforin+ ) than was noted at 14 days post bi ( compare Fig 6D , 6F , and 6G with Fig 4D , 4F , and 4G , p<0 . 05–0 . 01 ) . At 180 days post bi , the percentage of CD8+T cells capable of proliferating and producing type 1 cytokines with lytic activity contracted by at least 30% in comparison to that noted at 120 days post bi ( compare Fig 7D , 7F , and 7G with Fig 6D , 6F , and 6G ) . Together , the data presented in Figs 6 and 7 ( along with Fig 4 ) suggested that the D/P vaccine/bi elicited CD4+ and CD8+ T cells were stable , the frequency of TCM cells was enhanced at 120–180 days post bi; and these CD4+ and CD8+ T cells were capable of rapidly responding to antigenic stimulus with proliferation and activation of type 1 cytokine production and cytolytic profile . Finally , we determined whether long-lived T cells present in vaccinated/bi mice were capable of responding to T . cruzi infection . For this , at 120 or 180 days post bi , mice were challenged with T . cruzi and splenic cell characterization performed at day 10 pi . When challenged at 120 days post bi , mice exhibited a similar expansion of CD4+TEM cells capable of producing IFNγ and/or TNFα cytokines as was noted in mice challenged at 14 days post bi ( compare Fig 8A , 8C , and 8E with Fig 5A , 5C and 5E , p<0 . 05–0 . 01 ) . The CD8+T cells were maintained at a very high frequency ( 72–76% ) in vaccinated mice at 120 days post bi ( Fig 6B ) , and expanded only by 15–21% post challenge infection ( Fig 8B ) . Importantly , in mice challenged at 120 days post bi , 36–38% , 24–28% , and 23–30% of the antigen-specific CD8+T cells were IFNγ+ , TNFα+ or IFNγ+TNFα+ , respectively , and 48–52% of the IFNγ+CD8+T cells were also CD107+perforin+ ( Fig 8D , 8F , and 8G , p<0 . 05–0 . 01 ) . These results suggested a significant expansion of cytokine producing ( 35–77% ) CD8+T cells with a cytolytic phenotype ( 2-fold increase ) in mice challenged at 120 days post bi in comparison to that noted in mice challenged at 14 days post bi ( compare Fig 8D , 8F , and 8G with Fig 5D , 5F , and 5G , p<0 . 05–0 . 01 ) . Subsequently , vaccinated/bi mice challenged 120 days post bi exhibited a significant control ( 1 . 5–2 . 3-fold ) of peripheral ( S1B Fig ) and tissue ( Fig 8H , p<0 . 05–0 . 01 ) parasites in comparison to the non-vaccinated/infected mice . At 180 days post bi , splenic ex vivo CD4+ and CD8+ TEM cells increased by 20–44% only in response to challenge infection ( with or without in vitro antigenic stimulation ) with no change in the TCM population ( Fig 9A and 9B ) . Yet , the effector CD4+ and CD8+ T cells generated post-challenge infection were poly-functional as was evidenced by 1 . 8-4-fold increase in IFNγ- and/or TNFα- producing CD4+T cells ( compare Fig 9C with Fig 7C , p<0 . 01 ) and up to 2-fold increase in type 1 cytokine-producing CD8+T cells of cytolytic phenotype ( compare Fig 9D , 9F , and 9G with Fig 7D , 7F , and 7G , p<0 . 01 ) . The vaccinated mice , challenged at 180 days post bi , were able to achieve a 1 . 5–1 . 8-fold control of T . cruzi infection as compared to that noted in non-vaccinated/infected mice ( Figs 9H and S1B , p<0 . 05 ) . Together , the data presented in Figs 8 and 9 suggested that the D/P vaccine/bi induced T cell immunity was stable , and capable of rapidly expanding with a poly-functional phenotype in response to challenge infection at least until 180 days post bi , and provided significant control of parasite dissemination and replication .
The pathology of Chagas disease presents a complicated and diverse picture in humans [14] . The major complications and destructive evolutionary outcomes of chronic infection by T . cruzi in humans include ventricular fibrillation , thromboembolism , and congestive heart failure [15 , 16] . Studies in animal models and human patients have revealed the pathogenic mechanisms found during disease progression , and the features of protective immunity [17] . Parasite-specific CD4+T cells are suggested to assist in T . cruzi control through secretion of Th1 cytokines ( e . g . , IFNγ , IL2 ) , amplification of the phagocytic activity of macrophages , stimulation of B cell proliferation , and antibody production and differentiation and activation of CD8+ T cells ( reviewed in [17] ) . T . cruzi antigen-specific CD8+T cells contribute to parasite control , either by cytolysis of the infected cells or by secretion of Th1 cytokines ( IFNγ ) that induce trypanocidal activity ( reviewed in [18 , 19] ) . Accordingly , several antigens ( e . g . GP90 , cruzipain , GP82 , ASP2 , TSA1 , Tc24 ) have been tested to elicit protective immunity to T . cruzi in experimental animals ( reviewed in [7 , 20 , 21] ) . A majority of the candidate antigens were selected based upon their potential to be recognized by antibodies in infected mice and have proved to be efficacious as vaccine in providing some degree of protection from T . cruzi infection . In parallel , efforts to enhance the protective efficacy of subunit vaccines against T . cruzi have included testing the use of adjuvants , e . g . saponin , CpGODN , IL-12 and GMCSF cytokines; [22] attenuated strain of Salmonella [23] or adenovirus [24] for antigen delivery , and heterologous prime-boost protocols [25] . Based upon several studies that we have conducted , we believe TcG2 and TcG4 candidate antigens are an excellent choice for subunit vaccine development , and a heterologous prime/boost approach for vaccine delivery is highly efficacious against T . cruzi infection . The selected candidates TcG2 and TcG4 tested in this study ( and TcG1 tested in other studies ) are highly conserved in clinically relevant T . cruzi strains , expressed ( mRNA/protein ) in infective trypomastigote and intracellular amastigote stages of T . cruzi , and released during parasite differentiation in host cell cytoplasm , a characteristic required for antigen presentation for T cell activation [10] . These antigens showed antigen-specific antibody ( IgG1 , IgG2a and IgG2b ) and/or CD8+ T cell responses in T . cruzi-infected mice and dogs . Three candidate antigens ( TcG1 , TcG2 and TcG4 ) were also recognized by antibody response in chagasic patients from distinct study sites ( Argentina-Bolivia and Mexico-Guatemala ) and expressed in diverse strains of the circulating parasites [12 , 13 , 26] . Further , we noted that immunization with candidate antigens as a DNA vaccine provided T cell immunity ( TcG2 = TcG4>TcG1 ) that was additive when the antigens were co-delivered and achieved a significant ( but modest ) control of challenge infection in mice and dogs [10–12] . The delivery of candidate antigens as heterologous prime/boost vaccine [13 , 27 , 28] provided protective immunity consisting of parasite- and antigen-specific lytic antibodies and type 1 CD8+ cytotoxic T lymphocytes against challenge infection and chronic disease that was significantly better than that observed with DNA-prime/DNA-boost vaccine [10 , 11] . The enhanced efficacy of a heterologous prime/boost approach for vaccine delivery could be because delivery of antigens as DNA vaccines elicits robust T-cell responses , which are critical for the development of T-cell-dependent antibody responses [29 , 30] , and DNA immunization is also highly effective in priming antigen-specific memory B cells . Delivery of vaccine candidates as recombinant proteins is generally more effective at eliciting antibody responses and may directly stimulate antigen-specific memory B cells to differentiate into antibody-secreting cells , resulting in production of high-titer , antigen-specific antibodies [31 , 32] . Studies testing the protective efficacy of Tc24 , TSA1 ( individually or in combination , [33] ) or other antigens ( reviewed in [7 , 20 , 21] ) have shown the protection associated with the induction of CD8+ T cell activity and IFN-γ production; however , in these studies , challenge infection was conducted immediately after vaccination . To the best of our knowledge , this is the first report documenting that a ) a subunit vaccine can be useful in achieving long-term protection against T . cruzi infection and Chagas disease , and b ) the effector T cells can be long-lived and play a role in vaccine elicited protection from parasitic infection . Approximately , ~30–40% of the infected individuals develop symptomatic clinic disease presented as severe inflammatory myocarditis , and extensive destruction and fibrosis of the heart . T . cruzi antigen-specific T lymphocytes producing inflammatory cytokines ( e . g . IFNγ , IL17 ) can be detected in most individuals during chronic infection [34]; however , their role in host resistance versus chronic pathology is debated [35–38] . In experimental studies using a variety of genetically modified mice , both innate and adaptive immune responses were found to play an important role in providing resistance to T . cruzi infection . With regard to innate immune responses , a compromised signaling through toll-like receptors ( TLR-3 , -7 , -9 ) and their adaptor molecules MyD88 and TRIF , and inhibition of IL-1-activating inflammasomes resulted in increased susceptibility to T . cruzi infection [39–41] . Likewise , mice genetically deficient in CD4+ or CD8+ T cell function were extremely susceptible to infection [42] , and it was shown that T lymphocytes act by producing type 1 cytokines , such as IFNγ , and CD8+ T cell cytotoxicity mediated by perforin is important for resistance against infection [9 , 25 , 43 , 44] . The question then arises as to how to determine that vaccination was effective since both vaccine and T . cruzi infection elicited potent T cell responses . With respect to this issue , our data in the present study provide clues to the mechanisms relevant to vaccine efficacy . One , immunization with the 2-component D/P vaccine resulted in the generation of antigen-specific , IFNγ-producing CD4+ and CD8+ T cells and CD8+ cytotoxic T lymphocytes that were long-lived and could be further enhanced by bi ( Figs 2 and 4 ) . Two , vaccinated mice exhibited an expansion of T cell responses within 7–10 days post-infection . Immune responses elicited by vaccination followed by infection were >3-orders of magnitude higher than that generated by infection only ( at 10 dpi ) and were mediated by multifunctional lymphocytes abundantly secreting pro-inflammatory cytokines , such as IFNγ and TNFα , capable of translocating the CD107a and perforin molecules to the cell surface , and producing cytotoxic activity against infected target cells . The T cell immunity elicited in D/P vaccinated ( ± booster-immunized ) mice was similar to that noted against self-limiting infections . For example , influenza , LCMV , or Listeria have been reported to result in the highest T cell immunity , as measured by the frequency and function of specific CD8+ T lymphocytes , within 7–15 days pi [45] . In comparison , in experimental models of T . cruzi infection and human patients , parasite-specific T cell immunity is delayed ( peaks at ≥30 day pi [44 , 46] ) ; and when improved protective immunity was generated , it correlated with significantly higher numbers of antigen-specific IFN-γ producing total and CD8+ T cells that were better able to expand after in vitro re-stimulation [47] . Thus , we propose that vaccine-induced protection to T . cruzi infection and Chagas disease is associated with an increased frequency of highly competent CD8+ T lymphocytes at an early stage of parasite infection and replication . Further studies will be required to identify the underlying mechanisms for a delayed T cell response in natural T . cruzi infection , though a recent study suggests that accelerated apoptosis of CD8+T cells , resulting in inability to clear parasites , might be the key reason for chronic disease development [48] . Finally , we propose that our studies challenge the immunological paradigm of vaccine development . It is suggested that following pathogen control ( or clearance ) , 90–95% of the T effector cells reportedly die ( contraction phase ) , leaving behind a population of pathogen-specific CD8+ T central memory cells ( TCM ) that undergo antigen-independent/cytokine-dependent homeostatic proliferation which supports their long-term maintenance [45] . Moreover , these cells were found to efficiently increase cytokine and chemokine production , acquire cytotoxic ability and proliferate intensely during recall responses [49] . Thus , after an infectious challenge , heightened precursor frequency , expanded anatomical distribution , enhanced proliferative capacity , and rapid recall ability were reported as the hallmark attributes of protective CD8+ TCM cells [50 , 51] . During chronic Leishmania infection , TEF cells were maintained at high frequencies via reactivation of TCM and the TEF themselves and the short lived effector T cells were shown as the critical cells that mediate concomitant immunity [52] . In our studies , we have used a heterologous , prime-boost vaccination regimen consisting of priming with plasmid DNA , followed by a bi with either replication-defective MVA [27 , 28] or recombinant proteins ( this study ) , in all instances using the TcG2 and TcG4 as vaccine candidates . In all these studies , we observed that our immunization protocol induced a stable pool of functional CD8+TEM cells ( CD62L-CD44+ ) , and these cells expanded in response to challenge infection at 14 days pi . In this study , a D/P vaccine ( ± bi ) produced a stable pool of T effector memory ( TEM ) cells that exhibited rapid recall response to challenge infection with a potent increase in antigen-specific inflammatory cytokine production and CTL activity against specific target cells ( Figs 3 , 5 , and 8 ) . We did notice the generation of a strong pool of TCM cells in vaccinated mice after 120 or 180 days post bi , the time period equivalent to chronic infection; however , this pool of TCM was smaller than the TEM cells until 120 days post bi , and became predominant pool only after 180 days post bi ( Figs 3 , 5 , 8 , and 9 ) . Others have shown that TEM cells , induced by ASP2 vaccine , were the main source of the anti-parasitic mediators IFNγ and TNFα , and these cells did not need to proliferate ( rather increase the function ) to provide protective immunity [53] . In patients with severe chagasic disease , a lower frequency of specific CD8+T cells with fewer early memory cells ( CD45RA- , CD27+ , and CD28+ ) and a higher number of late memory cells ( CD45RA- , CD27- , and CD28- ) that lack the capability to respond to parasite stimulus was noted [54] . Further , frequency of CD8+ TEM lymphocytes remains high for long periods of time after infection , likely due to parasite persistence . Indeed , treatment with an anti-parasite drug in the chronic disease phase led to the contraction of CD8+ T cells and a change in their phenotype to CD62Llow [55] . Further studies will be required to establish the implications of these observations on the fate of CD8+T cells in parasitic infections in the development of new vaccines [56] . In summary , we have demonstrated that our DNA-prime/protein-boost vaccine constituted of TcG2 and TcG4 candidate antigens provides long-lived , rapid recall immunity to T . cruzi infection . Mice immunized with a 2-component D/P vaccine elicited long-lived CD4+ and CD8+ effector memory T cells capable of producing type 1 cytokines , and cytotoxic T lymphocyte profile in an antigen-specific manner . The vaccinated mice responded to challenge infection with a rapid and potent expansion of type 1 cytokines producing CD4+ and CD8+ T cells and cytotoxic T lymphocyte activity against infected target cells , resulting in a >2-3-fold control of acute parasitemia and tissue parasite burden . Importantly , vaccine-induced immunity could be enhanced by bi that helped to maintain a high level of T cell population capable of efficiently expanding and providing an up to five-fold control of an invading pathogen . The vaccine-induced immunity waned slightly after 6 months post bi , but was still sufficient to provide 2-fold control of invading pathogens that , according to mathematical modeling , is sufficient to break the parasite transmission cycle [57] and prevent disease progression [8] . Many of the studies discussed highlight the importance of a preventive or therapeutic vaccine to control T . cruzi infection by at least decreasing parasite burden , cardiac tissue inflammation and damage or increasing survival if not providing sterile immunity .
All animal experiments were conducted following NIH guidelines for housing and care of laboratory animals and in accordance with protocols approved by the Institutional Animal Care and Use Committee ( protocol number 08-05-029 ) at The University of Texas Medical Branch at Galveston . The cDNAs for TcG2 and TcG4 ( SylvioX10 isolate , Genbank: AY727915 and AY727917 , respectively; >99% homologous to CL Brenner ( reference strain ) sequences XM_806323 and XM_816508 , respectively ) [10] were cloned in eukaryotic expression plasmid pCDNA3 . 1 [10 , 11] . Plasmids encoding murine IL-12 ( pcDNA3 . msp35 and pcDNA3 . msp40 ) and GM-CSF ( pCMVI . GM-CSF ) were previously described [11] . Recombinant plasmids were transformed into E . coli DH5-alpha-competent cells , grown in L-broth containing 100-μg/ml ampicillin , and purified by anion exchange chromatography by using the Qiagen maxi prep kit ( Qiagen , Chatsworth , CA ) . The cDNAs for TcG2 and TcG4 were cloned in-frame with a C-terminal His-tag into a pET-22b plasmid ( Novagen , Gibbstown , NJ ) . Plasmids were transformed in BL21 ( DE3 ) pLysS-competent cells , and recombinant proteins purified by using the poly-histidine fusion , peptide-metal chelation chromatography system [13] . The pLW44 vector consists of a green fluorescent protein ( GFP ) and multiple cloning site ( MCS ) cassette flanked by a pair of genomic sequences of the Modified Vaccinia Ankara ( MVA ) virus which allows homologous recombination and incorporation of both GFP and the gene of interest into the deletion III locus of the wild-type MVA genome [58 , 59] . The cDNAs for TcG2 and TcG4 were cloned at the Xma1/Sbf1 sites of pLW44 , and recombinant plasmids were transformed and amplified in E . coli , and purified by using the Qiagen maxi prep kit . BHK-21 cells at 70% confluency ( six-well plate ) were infected with wild-type MVA ( multiplicity of infection: 0 . 05 ) for one h , and then transfected with recombinant pLW44 plasmids encoding TcG2 or TcG4 ( 2 μg DNA ) by using Lipofectamine 2000 reagent ( Invitrogen , Grand Island , NY ) . After 48 h of incubation , cells were harvested , and cell lysates added at 10-fold dilutions to new BHK-21 cell monolayers that were then overlaid with 2% methylcellulose ( Sigma , St . Louis , MO ) , and incubated for 48 h . At least three GFP+ fluorescent plaques were picked for each recombinant MVA ( rMVA ) . The plaque purification procedure was repeated 4–6 times to ensure removal of wild-type MVA contamination . Purified plaques of rMVA . TcG2 and rMVA . TcG4 were then amplified in BHK-21 cell monolayers , and viral pellets were stored in 1 mM Tris-HCl ( pH 9 ) at -80°C [27] . T . cruzi trypomastigotes ( Sylvio X10/4 strain ) were maintained and propagated by continuous in vitro passage in C2C12 cells . C57BL/6 female mice ( 6-week-old ) were obtained from Harlan Labs ( Indianapolis , IN ) . Mice ( 7-8-weeks-old ) were immunized with the DNA prime dose consisting of the TcG2- and TcG4-encoding plasmids with or without IL-12- and GM-CSF-expression plasmids ( 25 μg each plasmid DNA/mouse , intramuscularly ) . Twenty-one days later , mice were given recombinant proteins ( TcG2 and TcG4 , 25 μg of each protein emulsified in 5 μg saponin/100 μl PBS/mouse , intra-dermally ) . Mice were harvested at 14 and 120 days after vaccination to determine early and long-term vaccine-induced T cell immunity , respectively . In some experiments , 90 days after the DNA-prime/protein-boost ( D/P ) vaccination , mice were given a booster immunization ( bi ) with recombinant proteins as above , and then sacrificed at 14 , 120 and 180 days later to evaluate if bi provided stable , long-term anti-T . cruzi T cell immunity . To evaluate the recall response to T . cruzi , mice in each group , i . e . , 120 days after D/P vaccination or 14 , 120 , and 180 days after D/P/P vaccination , were challenged with T . cruzi ( 10 , 000 trypomastigotes/mouse , i . p . ) , and sacrificed 10 days post infection ( dpi ) . Some mice were also sacrificed at >90 dpi to evaluate the vaccine efficacy in controlling chronic parasite burden . Single-cell suspensions of spleen cells were prepared and cell number counted ( S1 Table ) . Splenocytes ( 105 cells/100 μl RPMI ) were distributed in 24-well plates , and incubated in the presence of Con A ( 5 μg/ml ) , recombinant proteins ( 10 μg/ml ) , or T . cruzi trypomastigotes lysate ( TcTL , 25 μg protein/ml ) at 37°C , 5% CO2 for 48 h . Un-stimulated or in vitro- stimulated splenocytes were washed in staining buffer ( 2% BSA/0 . 02% sodium azide in PBS ) and incubated for 15 min in Fc Block ( anti-CD16/CD32; BD Pharmingen ) . To evaluate the surface staining of effector and memory markers , we incubated the cells with the fluorescence- conjugated PE-αCD4 , FITC-αCD8 , PerCPCy5 . 5-αCD62L and APC-αCD44 antibodies ( BD Pharmingen ) for 30 min at 4°C in the dark . Then cells were washed twice in PBS and fixed in 2% paraformaldehyde . Fluorescent cells were visualized by using a FACSCalibur Cell Analyzer ( BD Biosciences ) , acquiring >30 , 000 events in a live lymphocyte gate , and further analyzed by using FlowJo software ( version 7 . 6 . 5 , Tree-Star , San Carlo , CA ) [27] . For the measurement of intracellular cytokines , splenocytes were stimulated as above except that brefeldin A ( 10-μg/ml , Sigma ) or monensin ( 5 μg/ml ) ( BD Pharmingen ) was added for the final 6 h of culture to block protein secretion . Cells were labeled with PE-αCD4 and FITC-αCD8 antibodies , fixed with 2% paraformaldehyde , re-suspended in 100 μl permeabilization buffer ( 0 . 1% saponin/1% FBS in PBS ) , and then utilized for intracellular staining with APC-αIL4 , PerCPCy5 . 5-αIL10 , e-Fluor-αIFNγ , Cy5-αTNFα and PerCP-Cy5 . 5-αKi67 antibodies ( 0 . 5-2-μg/100-μl , e-Biosciences ) . In some experiments , splenocytes were also incubated with APC-anti-perforin or Alexa-Fluor 488-anti-CD107 antibodies to determine the cytolytic activity of the activated/proliferating T cell subpopulations . Samples were analyzed by flow cytometry as above . In all experiments , cells stained with isotype-matched IgGs were used as controls [13 , 27] . Splenocytes ( 5×106 cells/2 ml/well in 24-well plates ) were incubated with recombinant TcG2 or TcG4 proteins ( 10-μg/ml ) at 37°C , 5% CO2 for 4–5 days to generate effector T cells . Semi-confluent EL-4 monolayers were exposed to rMVA encoding TcG2 or TcG4 ( 10 pfu/cell ) for 1 h ( controls: WT MVA ) , washed , and incubated in complete medium for 12 h at 37°C . Target EL-4 cells infected with GFP+ rMVA were co-cultured with effector cells ( Effector: target cell ratio: 80:1–20:1 ) in 200 μl of RPMI medium at 37°C , 5% CO2 for 4 h . Cells were mixed with 2 . 5 mM EDTA to reduce the number of cell-cell conjugates , and 5 μl propidium iodide ( PI ) to discriminate viable and nonviable cells , and analyzed by flow cytometry for a fixed period of 60 sec/sample . The forward scatter ( FSC ) acquisition threshold was set to include nonviable events . The CTL activity was calculated using [GFP+ target cells after incubation with effector cells x 100 / Total GFP+ target cells] . Blood DNA was isolated with a QiAamp Blood DNA mini kit ( Qiagen , Chatsworth , CA ) . Skeletal muscle , spleen , and heart tissue ( 50 mg ) were subjected to proteinase K lysis , and total DNA was purified by phenol/chloroform extraction and ethanol precipitation . Total DNA ( 50 ng ) was used as a template , and real-time PCR performed on an iCycler thermal cycler with SYBR Green Supermix ( Bio-Rad ) and Tc18SrDNA-specific oligonucleotides . Data were normalized to murine GAPDH and fold change in parasite burden ( i . e . Tc18SrDNA level ) calculated as 2-ΔCt , where ΔCt represents the Ct ( infected ) —Ct ( control ) [60 , 61] . Data are expressed as mean ± SD ( n = 8/group , triplicate observations per experiment ) . Data were analyzed by the Student’s t test ( comparison of 2 groups ) and 1-way analysis of variance ( ANOVA ) with Tukey’s post-hoc test ( comparison of multiple groups ) by using an SPSS ( version 14 . 0 , SPSS Inc , Chicago , Illinois ) or Graph Pad InStat ver . 3 software . Significance is presented as *p<0 . 05 , **p<0 . 01 , or ***p<0 . 001 ( vaccinated versus non-vaccinated or vaccinated/infected versus non-vaccinated/infected ) .
|
Chagas disease , caused by Trypanosoma cruzi infection , represents the third greatest tropical disease burden in the world . No vaccine or suitable treatment is available for control of this infection . Based upon several studies we have conducted , we believe that TcG2 and TcG4 candidate antigens that are highly conserved in T . cruzi , expressed in clinically relevant forms of the parasite , and recognized by both B and T cell responses in multiple hosts , are an excellent choice for subunit vaccine development . In this study , we demonstrate that the delivery of TcG2 and TcG4 as a DNA-prime/protein-boost vaccine provided long-term protection from challenge infection , and this protection was associated with elicitation of long-lived CD8+ effector T cells . The longevity and efficacy of vaccine could be enhanced by booster immunization . We believe that this is the first report demonstrating a ) a subunit vaccine can be useful in achieving long-term protection against T . cruzi infection and Chagas disease , and b ) the effector T cells can be long-lived and play a role in vaccine elicited protection from parasitic infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
A Two-Component DNA-Prime/Protein-Boost Vaccination Strategy for Eliciting Long-Term, Protective T Cell Immunity against Trypanosoma cruzi
|
Most statistical and mechanistic models used to predict mosquito-borne disease transmission incorporate climate drivers of disease transmission by utilizing environmental data collected at geographic scales that are potentially coarser than what mosquito populations may actually experience . Temperature and relative humidity can vary greatly between indoor and outdoor environments , and can be influenced strongly by variation in landscape features . In the Aedes albopictus system , we conducted a proof-of-concept study in the vicinity of the University of Georgia to explore the effects of fine-scale microclimate variation on mosquito life history and vectorial capacity ( VC ) . We placed Ae . albopictus larvae in artificial pots distributed across three replicate sites within three different land uses–urban , suburban , and rural , which were characterized by high , intermediate , and low proportions of impervious surfaces . Data loggers were placed into each larval environment and in nearby vegetation to record daily variation in water and ambient temperature and relative humidity . The number of adults emerging from each pot and their body size and sex were recorded daily . We found mosquito microclimate to significantly vary across the season as well as with land use . Urban sites were in general warmer and less humid than suburban and rural sites , translating into decreased larval survival , smaller body sizes , and lower per capita growth rates of mosquitoes on urban sites . Dengue transmission potential was predicted to be higher in the summer than the fall . Additionally , the effects of land use on dengue transmission potential varied by season . Warm summers resulted in a higher predicted VC on the cooler , rural sites , while warmer , urban sites had a higher predicted VC during the cooler fall season .
Epidemics of dengue , chikungunya , and Zika are spreading explosively through the Americas creating a public health crisis that places an estimated 3 . 9 billion people living within 120 different countries at risk . This pattern began with the growing distribution of dengue virus ( DENV ) over the past 30 years , infecting an estimated 390 million people per year . More recent invaders , chikungunya ( CHIKV ) and now Zika virus ( ZIKV ) , are rapidly following suit . CHIKV emerged in the Americas in 2013 and has caused 1 . 8 million suspected cases from 44 countries and territories ( www . paho . org ) to date . In 2015 , outbreaks of Zika virus ( ZIKV ) have spread throughout the Americas , resulting in over 360 , 000 suspected cases , with likely many more going unreported ( www . paho . org ) . Temperature is one of the key environmental drivers influencing the dynamics and distribution of these diseases [1–10] . Variation in temperature can profoundly impact mosquito population dynamics [11] , mosquito life history traits [12–18] , mosquito immune responses [19–22] ) , and measures of parasite / pathogen fitness ( prevalence , titers , and the extrinsic incubation period ) [1 , 10 , 23 , 24] . In addition to environmental temperature , variation in precipitation [25–27] and relative humidity [28] also drive vector-borne disease transmission . Most statistical and mechanistic models used to predict mosquito borne disease transmission incorporate climate drivers of disease transmission by utilizing environmental data collected from general circulation weather models [1 , 29–32] , down-scaled weather data [33] , outdoor weather stations [34 , 35] , or remotely sensed land surface temperature data [36–38] . While working with these data is methodologically tractable , mosquitoes do not experience environmental variation at such coarse scales [39 , 40] . Temperature and relative humidity can vary greatly between indoor and outdoor environments [41 , 42] , and can be influenced strongly by variation in landscape features such as density of housing , housing material , vegetation cover , impervious surface cover , waste heat generation , and distance to water [18 , 28 , 43–48] . Thus , the microclimate a mosquito vector experiences will be dependent upon its dispersal ability ( can be < 100 m for some species [49] ) and the habitats it visits throughout its life . In addition , many modeling efforts characterize environmental conditions through measures of mean monthly temperature , relative humidity , and precipitation . Yet , there is a growing body of theoretical and empirical work demonstrating that daily fluctuations in temperature , and likely relative humidity , are important for both mosquito and parasite / pathogen traits that mediate transmission [1 , 2 , 5 , 43] . We conducted a semi-field study examining differences in microclimate and mosquito life history traits across a heterogeneous urban landscape to address the above concerns . Specifically , 1 ) how does mosquito relevant microclimate vary across an urban landscape , 2 ) how does this variation affect mosquito life history traits , and 3 ) what are the implications of microclimate variation for vectorial capacity ? We investigated these questions in Athens-Clarke Country , GA , focusing on the invasive Aedes albopictus ( Asian tiger mosquito ) system due to its widespread distribution throughout the state [50] , as well as its role as an important vector for dengue , chikungunya , and Zika viruses in many parts of the world [51–54] .
We explored microclimate variation across three levels of land use categories characteristic of an urban landscape: urban , suburban and rural . We used an impervious surface map of Georgia generated by the Natural Resources Spatial Analysis Lab at the University of Georgia ( http://narsal . uga . edu/glut/data-stats/georgia-impervious-surface-trends ) for Athens-Clarke County , Georgia , U . S . A . to distinguish sites into these three land use categories . We defined urban , suburban , and rural sites as those with impervious surface scores within the following binned ranges: 55–100% , 5–50% , and 0% , respectively . We then created a new impervious surface map for Athens-Clarke County and selected three replicate sites within each land use category ( Fig 1 ) . Final site selection across Athens-Clarke County was ultimately constrained to sites that we could get permission to access . We did ensure that there was greater than 5 miles between sites , sites were interspersed across the county , and they were of the same area ( 30 m2 , Fig 1 ) . Within each site , we evenly distributed ( 10 m apart ) and staked six black flower pots ( Home Depot 480064–1001 ) in the ground at the base of vegetation ( e . g . grass stands , brush , trees ) in full shade . Within each pot , we placed a wide-mouth glass bell jar ( ~1 L , Walmart , 550797441 ) , and added 300 mL of leaf infusion and 30 first instar Aedes albopictus larvae . Leaf infusion was made a week prior to the start of the experiment . Live oak ( Quercus virginiana ) leaves were collected from the field and dried in an oven ( 50°C ) for 72 hrs to ensure all water had evaporated from the leaf tissue . We then infused 80 grams of dried leaf material and 3 grams of a 1:1 yeast-albumin mixture in 20 L of deionized water for 3 days prior to use . To monitor variation in larval and adult mosquito microclimate across each site , we added a data logger ( Monarch Instruments: RFID Temperature Track-It logger ) to each jar and hung a logger ( Monarch Instruments: RFID Temperature and Relative Humidity Track-It logger ) in vegetation near each jar ( ~ 3 feet above the ground ) . Loggers recorded instantaneous measurements of temperature and relative humidity every 10 min throughout the course of the study . Jars were then screened to prevent escape of emerging adults and a wire cage ( 8 in x 8 in ) with plastic vinyl lining the roof was placed over top and staked into the ground to exclude animals and excess rainfall . Pots were visited daily and emerged adults were removed . We quantified the total number of adults emerging per day , and recorded the sex and wing length of each emerged adult . Wing length was used as a proxy of mosquito body condition due to its associations with female mosquito fecundity , survival , and vector competence for arboviruses [55–57] . One wing was taken from each individual upon emergence , and measurements were taken from the tip of the wing ( excluding fringe ) to the distal end of the alula using a dissecting scope and micrometer eye piece . This experiment was conducted twice , once in early summer ( June 15-July 14 , 2015 ) and once in the fall ( September 7-October 10 , 2015 ) to estimate any effects of season on our response variables . We used the following Eq ( 1 ) to calculate per capita intrinsic population growth rates ( r ) for each experimental pot across all sites [58] , r=ln ( 1No∑Axf ( wx ) ) D+ ( ∑xAxf ( wx ) ∑Axf ( wx ) ) , ( 1 ) in which No represents the initial number of females , Ax the number of adult females emerging per day x , wx the mean wing length of females emerging on day x , D the delay between female emergence and first oviposition , and f ( wx ) predicts the numbers of female offspring produced by females of a given wing size . Because 1st instar mosquito larvae cannot be reliably sexed , and 30 1st instar larvae were deposited in each experimental pot , we assumed No to be 15 females . We also assumed D = 14 . 2 days for Ae . albopictus [58] . We used the following linear function , f ( wx ) = -121 . 240 + 78 . 02wx , to describe the relationship between mean wing size and fecundity [59] . While other relationships between wing length and egg production exist [60–62] , we chose the relationship characterized in [59] for two reasons . First , this study used variation in mean temperatures to generate Ae . albopictus adults of different sizes , which ensured we were predicting egg production from variation in wing lengths generated from an environmental variable we allowed to vary in our study . Second , other environmental manipulations ( e . g . density , food availability ) can alter mosquito body condition or teneral reserves relative to variation in environmental temperature [56] , potentially resulting in different relationships between wing length and egg production . To estimate the effects of microclimate and land use on the larval development and mosquito emergence rates , we used Cox proportional hazard models ( R version 3 . 3 . 0 , package ‘survival’ ) to assess how these predictors influenced probability of mosquito emergence across pots in each season ( summer and fall ) . Each model included land use ( rural , suburban , and urban ) and the following microclimate covariates ( daily temperature mean , minimum , and maximum in each experimental pot and average daily relative humidity mean , minimum , and maximum ) as predictor variables . Additionally , to control for correlated observations , pot was included as a cluster variable in the analysis . We achieved our final models by using a multidirectional stepwise selection method designed to minimize Akaike Information Criterion ( AIC ) [63] . All predictors included in final models were checked by using chi-squared tests to verify the assumption that the hazard functions are proportional over time for each strata was upheld . Influential observations and nonlinearity were investigated by removing one observation for each covariate and observing how much the regression coefficients changed and plotting the Martingale residuals ( the difference between the observed and expected number of events at each time interval ) against each covariate , respectively . We used general linear mixed effects models ( JMP Pro 12 . 1 . 0 ) to investigate the effects of season ( summer and fall ) , land use class ( rural , suburban , and urban ) , and the interaction on metrics of larval microclimate ( average daily mean , minimum , and maximum temperature in each experimental pot and average daily mean , minimum , and maximum relative humidity ) , mosquito body size upon emergence ( wing size ) , the per capita mosquito population growth rate , r , and overall transmission potential . Experimental pot was nested within site as a random factor within each model . Sex , and the interactions with season ( sex x season ) and land use ( sex x land use ) , were also included as predictors in the model with wing size as the response variable . Model fit and assumptions of normality were assessed by plotting model residuals and quantile plots , and Tukey HSD adjusted pairwise comparisons were run to compare differences across land use groups and for any significant interactions . Boxplots of raw data for each of our response variables are included in additional supplementary information files ( S1 , S2 and S3 Figs ) . To estimate how variation in relevant microclimate across different land uses and season might influence the ability of Ae . albopictus to transmit arboviruses , we used a dengue-specific vectorial capacity framework . Vectorial capacity ( VC ) is a mathematical Expression ( 2 ) that describes the daily rate at which future infections arise from one infectious mosquito [10 , 64–66]: VC=ma2bce−μ/EIRμ , ( 2 ) where m represents vector density , a is the daily probability of a female mosquito taking a blood meal , μ is the daily probability of adult mosquito mortality , b is the probability of transmission from an infectious human to a susceptible mosquito , c is the probability of transmission from an infectious mosquito to susceptible human , and EIR is the extrinsic incubation rate of a pathogen . The density of mosquitoes ( m ) was estimated with the following Eq ( 3 ) : m=EFDpEAMDRμ2 , ( 3 ) with m being comprised of the number of eggs laid per female per day ( EFD ) , the egg to adult survival probability ( pEA ) , the development rate of larvae ( MDR ) , and adult daily probability of mortality ( μ ) . We incorporated both parameter estimates derived from observations in our semi-field experiment ( pEA , MDR , and EFD ) with estimates of parameters from the literature ( a , b , c , EIR , μ ) to calculate the effects of season and land use on vectorial capacity . From our survival analyses in our semi-field experiment , we estimated the probability of egg to adult survival ( pEA ) and the mosquito development rate ( MDR ) as the maximum proportion of adult females emerging across each site and the slope of the inflection point of the cumulative emergence curves , respectively . We also estimated the number of eggs laid per female per day ( EFD ) by taking the number of expected eggs laid per gonotrophic cycle based on body size , using the linear relationship between eggs laid and wing length ( y = 78 . 02x-121 . 24 ) [59] . Because there is uncertainty in our estimates of EFD that is introduced from the allometric relationship of wing size and egg production [59] , we used Monte Carlo simulations to incorporate this uncertainty into our estimates of vectorial capacity . We first generated a variance-covariance matrix from the linear regression of wing size and egg production to generate a contour using the mvtnorm package in R . This generated a distribution of wing sizes that we can sample to estimate fecundity . For each pot , we used a random sample of 999 wing lengths to calculate associated egg production values . These were then divided by the expected lifespan ( 1/μ ) for each pot to generate an EFD estimate for each pot . These EFD values were then used to estimate a pot-level vectorial capacity , and the average vectorial capacity for each season and land use . Because the number of samples will be artificially inflated by the Monte Carlo permutations , we used the number of sites per season and land use as the true sample size n in our standard error calculations . To estimate the effects of daily mean temperature ( T ) variation across our sites and with season on parameters in vectorial capacity that we did not measure ( a , b , c , EIR , and μ ) , we used two non-linear functions described in Mordecai et al . [35] . The Briere thermal Eq ( 4 ) is used to explain asymmetric , non-linear relationships of traits with mean temperature , x ( t ) =cT ( T−To ) ( Tm−T , ( 4 ) while the quadratic Eq ( 5 ) is used to explain symmetric relationships , x ( t ) =c ( T−To ) ( T−Tm ) . ( 5 ) In both functions , To is the daily minimum temperature , Tm as the daily maximum temperature , and c is a fit parameter with values for these parameters taken from Mordecai et al . [35] . In order to estimate potential effects of variation in diurnal temperature ranges across our sites with land use and season on these parameters , we used rate summation [43 , 67] Eq ( 6 ) defined as x=∫r ( T ( t ) ) dt , ( 6 ) where a given trait ( x ) is defined as a rate ( r ) that adjusts instantaneously to temperature ( T ) , which in turn is a function of time ( t ) . Thus , for each hourly change in mean temperature , we used the Briere Eq ( 4 ) to estimate the biting rate ( a ) , transmission probabilities associated with vector competence ( b , c ) , and the extrinsic incubation rate ( EIR ) . We used the Quadratic Eq ( 5 ) for mosquito mortality ( μ ) . Selection criteria for using the Briere vs . the Quadratic curves for each parameter are outlined in Mordecai et al . [35] . We then summed the proportional hourly changes in parameter estimates across the course of the experiment to incorporate the effect of diurnal temperature fluctuation on each parameter estimate .
We found that the larval microclimate mosquitoes experienced significantly varied with time of season and with land use ( Table 1 , Fig 2 ) . We did not observe any significant interactions between season and land use , suggesting that the effects of land use on mosquito microclimate were consistent across the summer and fall experiments . Due to larval data logger failure , we were unable to track daily water temperatures across a total of six pots ( n = 48 pots ) in the summer and one pot ( n = 53 ) in the fall; however , as the failure was equally distributed across treatments , we do not believe this significantly affected our results . As expected , summer temperatures were on average higher than fall temperatures , with significantly higher daily mean ( summer: 26 . 0°C ± 0 . 08°C; fall: 20 . 5°C ± 0 . 08°C ) , minimum ( summer: 22 . 4°C ± 0 . 07°C; fall: 15 . 6°C ± 0 . 07°C ) , and maximum water temperatures ( summer: 29 . 6°C ± 0 . 12°C; fall: 24 . 5°C ± 0 . 12°C ) . Additionally , experimental pots in the summer were subject to lower daily mean ( summer: 82 . 8% ± 0 . 30%; fall: 92 . 8% ± 0 . 29% ) and minimum relative humidity ( summer: 55 . 9% ± 0 . 63%; fall: 74 . 8% ± 0 . 60% ) . We did not include maximum relative humidity in our analyses because the daily maximum relative humidity across all sites and seasons was consistently close to 100% ( Fig 2; S2 Fig ) . These seasonal differences in daily temperature and relative humidity resulted in summer mosquitoes experiencing a lower diurnal temperature range ( summer: 7 . 3°C ± 0 . 13°C; fall: 8 . 9°C ± 0 . 12°C ) and higher diurnal relative humidity range ( summer: 43 . 0% ± 0 . 63%; fall: 25 . 0% ± 0 . 61% ) across all sites . Urban sites were on average warmer than rural sites ( Fig 2 ) . Urban sites were characterized by higher daily mean temperatures ( Tukey HSD: urban vs . rural , p = 0 . 0234; urban vs . suburban , N . S . ; suburban vs . rural , N . S . ) and maximum temperatures ( Tukey HSD: urban vs . rural , p = 0 . 0011; suburban vs . urban , N . S . ; suburban vs . rural , N . S . ) . Interestingly , daily minimum temperatures were similar across suburban and urban sites , with larvae on rural sites experiencing significantly lower daily minimum temperatures ( Tukey HSD: rural vs . suburban , p = 0 . 0123; suburban vs . urban , N . S . ; urban vs . rural , N . S . ) . Urban sites were also significantly drier . Urban sites had lower daily mean relative humidity ( Tukey HSD: urban vs . suburban , p < 0 . 0001; urban vs . rural , p < 0 . 0001 , rural vs . suburban , N . S . ) and minimum relative humidity ( Tukey HSD: urban vs . suburban , p = 0 . 0023; urban vs . rural , p = 0 . 0007 ) . Finally , land use significantly affected fluctuations in diurnal temperature ( urban: 8 . 5°C ± 0 . 40°C; suburban: 7 . 9°C ± 0 . 13°C; rural: 8 . 0°C ± 0 . 14°C ) and relative humidity ( urban: 36 . 1% ± 0 . 85%; suburban: 33 . 2% ± 0 . 85%; rural: 32 . 7% ± 0 . 87% ) . Urban sites on average experienced wider fluctuations in diurnal temperature ( Tukey HSD: urban vs . suburban , p = 0 . 0023; urban vs . rural , p = 0 . 0007 , suburban vs . rural , N . S . ) and relative humidity ( Tukey HSD: urban vs . suburban , p = 0 . 0473; urban vs . rural , p = 0 . 0183; and suburban vs . rural , N . S . ) than suburban and rural sites ( but note that the comparison in mean diurnal humidity ranges between urban and suburban sites is only marginally significant ) . While the daily climate data collected by the local weather station do track the daily variation in temperature and relative humidity recorded by data loggers ( Fig 2 ) , the local weather station did not accurately predict daily mean , minimum , maximum , and diurnal ranges of temperature and relative humidity across each land use ( Fig 3 ) . Further , the ability of the local weather station to predict urban , suburban , and rural microclimate varied qualitatively across seasons . For example , in the summer , local weather station data over predicted daily mean ( by 1 . 3°C– 1 . 8°C ) , maximum , ( by 3 . 0°C– 4 . 2°C ) and diurnal temperature ranges ( by 3 . 1°C– 3 . 7°C ) , while under predicting variation in the daily mean ( by 6 . 8% to 13 . 3% ) , minimum ( 5 . 0%–9 . 4% ) , and maximum relative humidity ( 6 . 4%–8 . 2% ) across all land uses ( Fig 3 ) . In contrast , in the fall , the local weather station was better able to characterize daily mean ( a difference of 0 . 3°C– 0 . 7°C ) , maximum ( a difference of 0 . 8°C– 1 . 2°C ) , and the diurnal temperature range ( -0 . 8°C to -0 . 4°C ) across these sites . In the fall , like the summer , the local weather station continued to under predict the daily mean , minimum , and maximum relative humidity across urban , suburban , and rural sites . Interestingly , while the difference in relative humidity recorded by the local weather station and our data loggers was minimal in the summer ( -1 . 3%–1 . 2% ) , the local weather station in the fall marginally over estimates the relative diurnal humidity range ( 3 . 7%–7 . 8% ) in urban , suburban , and rural sites ( Fig 3 ) . Overall , larval survival and the number of adult mosquitoes emerging were much higher in the fall than in the summer ( Fig 4 ) . Of approximately 1 , 620 first instar Ae . albopictus placed into the field during each experiment , we had a total of 318 females and 387 males successfully emerge during the summer replicate and 569 females and 623 males emerge during the fall replicate . Additionally , adults began to emerge at an earlier date in the summer ( day 7 ) than in the fall ( day 11 ) . We found significant effects of land use on the likelihood of mosquito emergence in both the summer and fall , with a 44% and 47% decrease in the likelihood of mosquito emergence on urban sites relative to suburban and rural sites ( which had similar likelihoods of mosquito emergence ) , respectively ( Table 2 ) . There also was an effect of temperature and relative humidity on mosquito emergence in the summer and fall experiments , but interestingly these effects differed . Mosquitoes developing in the summer experienced an 18% decrease in the likelihood of emergence with each 1°C increase in the daily minimum temperature and a 7% decrease with each 1% increase in daily mean relative humidity ( Table 2 ) . In contrast , mosquitoes developing in the fall experienced a 28% increase in the likelihood of emergence with each 1°C increase in daily maximum temperature and a 19% decrease with every 1% increase in daily maximum humidity ( Table 2 ) . Together , these results suggest that higher temperatures on urban sites may decrease the likelihood of mosquito emergence through increased larval mortality , and that temperature variation throughout the day has qualitatively different effects on mosquito development and emergence in the summer than the fall . We found significant effects of sex , season , and land use on the size of emerging adult mosquitoes ( Table 3 , Fig 5 ) . Overall , female mosquitoes were larger than male mosquitoes ( females: 3 . 21 mm ± 0 . 01 mm; males: 2 . 71 mm ± 0 . 01 mm ) . Mosquitoes emerging in the summer were significantly smaller than those emerging in the fall ( summer: 2 . 77 mm ± 0 . 01 mm; fall: 3 . 15 mm ± 0 . 01 mm ) , and mosquitoes developing on urban sites emerged as smaller adults ( urban: 2 . 91 mm ± 0 . 02 mm; suburban: 2 . 96 mm ± 0 . 02 mm; rural: 3 . 01 mm ± 0 . 02 mm ) relative to rural sites ( Tukey HSD: urban vs . rural , p = 0 . 0047; urban vs . suburban , N . S . ; suburban vs . rural , N . S . ) . Interestingly , there were significant interactions between season and mosquito sex ( season x sex ) and land use ( season x land use ) , suggesting the effects of season on mosquito body size differs for males and females and across land use . For example , female mosquitoes were significantly larger than male mosquitoes ( female: 3 . 21 mm ± 0 . 01mm; male: 2 . 71 mm ± 0 . 01 mm ) , however this difference in body size was greater in the fall ( female: 2 . 95 mm ± 0 . 02 mm; male: 2 . 58 mm ± 0 . 01 mm; Tukey HSD: female vs . male , p < 0 . 0001 ) than the summer ( female: 3 . 46 mm ± 0 . 01 mm; male: 2 . 84 mm ± 0 . 01 mm; Tukey HSD: female vs . male , p < 0 . 0001 ) . Further , there were no significant effects of land use on mosquito body size in the summer ( urban: 2 . 73 mm ± 0 . 02 mm; suburban: 2 . 77 mm ± 0 . 02 mm; rural: 2 . 79 mm ± 0 . 02 mm ) , however in the fall , mosquitoes emerging on urban sites were significantly smaller ( urban: 3 . 09 mm; suburban: 3 . 14 mm; rural: 3 . 22 mm ) than those on rural sites ( urban vs . rural , p = 0 . 0003; urban vs . suburban , N . S . ; suburban vs . rural , N . S . ) . Integrating the daily emergence and wing size data into Eq ( 1 ) , we identified significant effects of season ( summer: 0 . 09 ± 0 . 004; fall: 0 . 157 ± 0 . 004 ) and land use ( urban: 0 . 115 ± 0 . 005; suburban: 0 . 134 ± 0 . 005; rural: 0 . 121 ± 0 . 005 ) on mosquito per capita population growth rates ( r , Table 3 , Fig 5 ) . Overall , the mosquito per capita growth rate was approximately two times higher in the fall than the summer . Further , the mosquito per capita growth rate was significantly lower on urban sites ( Tukey HSD: urban vs . suburban; p = 0 . 0269; urban vs . rural , N . S . ; suburban vs . rural , N . S . ) . We found mosquito transmission potential to vector dengue to significantly vary across seasons ( Table 3 , Fig 5 ) . Transmission potential was higher overall in the summer relative to the fall season . Interestingly , the effects of land use on mosquito transmission potential varied depending on time of season ( summer: urban , 152 . 3 ± 32 . 2; suburban , 153 . 6 ± 10 . 1; rural , 207 . 6 ± 16 . 9 and fall: urban , 52 . 8 ± 8 . 0; suburban , 12 . 0 ± 7 . 1; rural , 21 . 4 ± 4 ) . The model predicts that during the hot summer , Ae . albopictus on rural sites have the highest transmission potential relative to suburban and urban sites . In contrast , in the cooler fall , mosquitoes on urban sites were predicted to have the highest transmission potential ( Fig 5 ) . Together these results demonstrate fine-scale variation in transmission potential could potentially occur across an urban landscape , and seasonal shifts in microclimate may result in qualitatively different patterns of arbovirus transmission potential with land use .
To date , the majority of studies investigating the effects of urbanization on mosquito population dynamics and disease transmission have been sampling or modeling studies investigating how the distribution and abundance , feeding preferences , and incidence of diseases vectored by different mosquito species vary across land uses [46 , 68–77] . In contrast , there have been a handful of experimental studies in the field that mechanistically link observed variation in mosquito traits and metrics of disease transmission to sources of microclimate variation that exist across human-modified landscapes ( Anopheles spp . [18 , 47 , 78] , Culex pipiens [45] , Aedes albopictus [79] ) . Our study , in combination with the previous work , demonstrates that relevant microclimate variation in the field ( rather than coarser environmental manipulations in the lab ) can translate into significant heterogeneity in mosquito life history traits , and ultimately disease transmission potential . Across both the summer and fall , we observed urban microclimates to be significantly warmer and less humid than non-urban sites , which is reflective of the urban heat island ( UHI ) effect [80] . This is consistent with other studies showing that urban centers can have different temperature [81–83] and precipitation regimes [84–86] than surrounding areas due to significant modifications to the land-surface structure [44] and increases in the production of waste heat [44] . In other systems , these changes have led to shifts in organism phenology ( plants [87–89] ) , life history ( e . g . insect pests , ants , fruit bats [90–93] ) , and overwintering behavior ( mosquitoes [83] ) , all of which can have significant implications for vector-borne disease transmission [76 , 83] . Further , because our study site ( Athens , Georgia ) is a relatively small city , the observed effects of land use on fine-scale variation in microclimate could be much larger in more expansive cities with greater temperature differentials between urban cores and surrounding areas ( 3°C-10°C differential [79 , 80 , 83] ) . Despite the subtle effects of land use on mosquito microclimate , we still observed noticeable effects on larval survival , larval development rates , and adult mosquito body sizes , which translated into estimated differences in intrinsic population growth rates and overall transmission potential . This reinforces findings from a diversity of laboratory studies on Ae . aegypti and Ae . albopictus demonstrating the effects of relatively large changes in mean temperature [1 , 13 , 15 , 24 , 60 , 94–100] and diurnal temperature range [1 , 7 , 101–103] on a diversity of mosquito life history traits ( e . g . survival , biting rate , fecundity , larval development , vector competence , and viral extrinsic incubation period ) . We found mosquitoes developing on urban sites experienced lower survival in the larval environment , emerged as smaller adults , and experienced lower per capita growth rates than on non-urban sites , which could be due to urban sites being in general warmer than non-urban sites . Other similar studies report increases in mosquito development times [45 , 79] on urban sites and an increase in adult mosquito emergence [79] , which we did not observe . Surprisingly , different components dictating the diurnal range of temperature and relative humidity were important for larval survival . Overall , in the hot summer , the probability of adult mosquito emergence decreased with higher daily thermal minimums . In contrast , in the cooler fall , increases in the daily maximum temperatures corresponded to increases in the number of adults emerging . Despite having higher average daily thermal maximum temperatures relative to non-urban sites , mosquitoes developing on urban sites still experienced higher larval mortality in the fall . This suggests other unmeasured sources of variation with land use might also influence and have larger effects on larval survival ( micro-biotic activity in larval environments , exposure to insecticides , variation in vegetation cover , etc . ) on these sites [104 , 105] . Variation in relative humidity was also a predictor for the probability of adult emergence across these sites , and like temperature , different metrics of relative humidity were important across different seasons . Interestingly , in both the summer and fall , increases in either the daily relative humidity mean or maximum resulted in proportional decreases in the probability of adult emergence . While an increase in relative humidity has been shown to improve adult mosquito longevity and activity [106–108] , it can result in decreases in surface tension of aquatic environments [109] , which in turn can increase pupal mortality and decrease the probability of adult emergence in a diversity of mosquito species [110] . To the best of our knowledge , this is the first report of variation in relative humidity affecting the likelihood of larval survival and adult emergence and demonstrates that microclimate variation can have opposing effects on larval and adult traits that are relevant for fitness and transmission . Variation in daily temperature and relative humidity , as well as the observed variation in mosquito body size with land use and season , could have significant implications for other , unmeasured mosquito traits that are important for arbovirus transmission . For example , variation in both mean temperature and diurnal temperature range in the lab have been shown to impact the daily probability of adult survival ( μ ) , female gonotrophic cycles and biting rates ( a ) , the number of eggs females produce per day ( EFD ) , vector competence ( bc ) and the extrinsic incubation period ( EIP ) for a diversity of mosquito species and pathogens ( e . g . Anopheles [10 , 43] , Culex [23 , 111 , 112] , Aedes [7 , 101] ) . Modeling studies have linked increased precipitation and relative humidity to increased disease incidence ( e . g . dengue and malaria ) [113–117] , likely through the negative effects of low relative humidity ( e . g . < 40% relative humidity ) on mosquito longevity [107] and activity [108] . Finally , the observed variation in mosquito body size across land use and seasons were consistent with mosquito body sizes reported in other studies and could further compound the effects of microclimate variation on traits like the daily probability of adult survival ( μ ) [118–120] , egg production [60–62] , and vector competence [95 , 121] . For example , observed variation in average wing sizes of mosquitoes across our sites from the summer to the fall ( 2 . 7–3 . 2 mm ) could result in individual females producing between 86 and 132 eggs / gonotrophic cycle [59] and result in a 3 fold increase in the probability of dissemination of chikungunya [121] . We used a temperature dependent vectorial capacity equation parameterized for Ae . albopictus [35] to predict how dengue transmission potential varies across urban , suburban , and rural sites and with season . While the vectorial capacity formula ignores some potentially important sources of variation ( e . g . underlying the mosquito-human interaction ) , it provides a framework for estimating the relative importance of key mosquito / pathogen parameters and the effects of environmental variation on these parameters [1 , 43 , 122] . Relative vectorial capacity was predicted to be lower in the fall relative to the summer despite the fact that per capita mosquito population growth rates were predicted to be higher in the fall due to increased mosquito survival and egg production associated with increased body sizes . This is due to the negative effect of cooler temperatures on daily probability of mosquito biting ( a ) , the extrinsic incubation rate of dengue ( EIR ) , and the probabilities of transmission ( b , c ) [35] , which ultimately result in a smaller proportion of the mosquito population that is infectious and biting at this time of season . We also found arbovirus transmission potential to vary with land use , and the effects of land use on vectorial capacity depended on time of season . These results suggest that the environmental suitability for arbovirus transmission will be dependent upon the shape of the non-linear relationships mosquito and pathogen traits share with temperature , the daily average habitat temperatures and their proximity to the thermal optimum of this non-linear response , and how the effects of daily temperature fluctuation integrate with daily mean habitat temperatures to impact trait performance , and ultimately transmission potential . This study captures how mosquito life history , population growth rates , and transmission potential respond to variation in microclimate with land use and season . However , there could be variation in other factors that we did not quantify in this study that could ultimately be more important for transmission . Variation in quantity and quality of larval habitat , adult resting habitat , access to hosts , and insecticide application with land use will also likely influence mosquito population dynamics , densities , and transmission potential [73 , 123–126] . Further , while environmental conditions shape the potential distribution and magnitude of disease vectors , socio-economic and demographic factors ( e . g . variation in human population density , outdoor recreation , housing quality , etc . ) , human behavior and cultural variation , as well as mosquito feeding preferences will determine the level of human exposure and the realized transmission risk [127 , 128] . Thus , even though transmission potential is predicted to be lower in the fall than the summer , seasonal changes in human behavior may result in higher transmission risk in the fall when cooler temperatures encourage more outdoor activity . Likewise , transmission risk may actually be higher in the summer on urban relative to rural sites due to urban sites having higher human population densities . Finally , the replication associated with this study was relatively low , which could introduce uncertainty in our results inherent with small sample sizes . Most studies that consider the role of climate in vector-borne disease transmission use climate data reported from local weather stations . Our proof of concept study demonstrates that the climate conditions captured by local weather station data do not reflect the microclimates mosquitoes experience , and that subtle variation in mean and diurnal ranges of temperature and relative humidity can lead to appreciable variation in key mosquito / pathogen life history traits that are important for transmission . Greater effort is needed to quantify the activity space mosquitoes occupy and the conditions of relevant transmission environments . This will not only be important for predicting variation in transmission potential and risk across seasons , geographic regions , and land uses , but also for building realistic environmental variation in future laboratory work on mosquito-pathogen interactions .
|
Environmental factors influence the dynamics of mosquito-borne disease transmission . Most models used to predict mosquito-borne disease transmission incorporate climate data collected at coarser scales than mosquitoes typically experience . Climate conditions can vary greatly between indoor and outdoor environments , and are influenced by landscape features . We conducted a field experiment with the Asian tiger mosquito to explore how microclimate variation across an urban landscape affects mosquito life history and potential to transmit arboviruses , like dengue . We demonstrate that climate conditions captured by weather stations do not reflect relevant mosquito microclimate , and that subtle variation in mean and diurnal ranges of temperature and relative humidity can lead to appreciable variation in key mosquito / pathogen traits that are important for transmission . Our results have implications for statistical and mechanistic models used to predict variation in transmission across seasons , regions , and land uses , but also for building in realistic environmental variation in laboratory work on mosquito-pathogen interactions . Finally , the variation in microclimate we observed across land use was subtle; likely because our study site is a relatively small city . Nevertheless , these translated into considerable differences in mosquito traits and dengue transmission potential , suggesting these effects could be much larger in more expansive cities .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"autumn",
"weather",
"stations",
"atmospheric",
"science",
"social",
"sciences",
"animals",
"seasons",
"summer",
"land",
"use",
"physiological",
"parameters",
"humidity",
"insect",
"vectors",
"human",
"geography",
"research",
"and",
"analysis",
"methods",
"research",
"facilities",
"infectious",
"diseases",
"geography",
"disease",
"vectors",
"insects",
"arthropoda",
"mosquitoes",
"meteorology",
"earth",
"sciences",
"physiology",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"organisms"
] |
2017
|
Fine-scale variation in microclimate across an urban landscape shapes variation in mosquito population dynamics and the potential of Aedes albopictus to transmit arboviral disease
|
Explaining how interactions between genes and the environment influence social behavior is a fundamental research goal , yet there is limited relevant information for species exhibiting natural variation in social organization . The fire ant Solenopsis invicta is characterized by a remarkable form of social polymorphism , with the presence of one or several queens per colony and the expression of other phenotypic and behavioral differences being completely associated with allelic variation at a single Mendelian factor marked by the gene Gp-9 . Microarray analyses of adult workers revealed that differences in the Gp-9 genotype are associated with the differential expression of an unexpectedly small number of genes , many of which have predicted functions , implying a role in chemical communication relevant to the regulation of colony queen number . Even more surprisingly , worker gene expression profiles are more strongly influenced by indirect effects associated with the Gp-9 genotypic composition within their colony than by the direct effect of their own Gp-9 genotype . This constitutes an unusual example of an “extended phenotype” and suggests a complex genetic architecture with a single Mendelian factor , directly and indirectly influencing the individual behaviors that , in aggregate , produce an emergent colony-level phenotype .
Considerable interest surrounds the genetic architectures underlying fundamental adaptive traits in wild populations [1]–[5] . In social organisms , such interest centers on the numbers and types of genes directly regulating expression of the individual behaviors that , in aggregate , create social organization , as well as genes in interactants that indirectly influence expression of socially relevant behaviors by altering the social environment [6]–[12] . This indirect influence is mediated by interactions of the genotype of a given individual with those of other group members who collectively comprise the social environment . Information on the genetic architecture of social organization is essential to constructing realistic models of social evolution that can answer questions about the numbers and types of genetic changes necessary to change a solitary to a social animal or to convert a simple society to a large and highly complex one [13] . A remarkable case of a fundamental social polymorphism that appears to be under simple genetic control ( single Mendelian factor of large effect ) is variation in colony social organization in the fire ant Solenopsis invicta . In this species a single genomic element marked by the protein-encoding gene Gp-9 is implicated in the production of two distinct types of queens that differ in physiology , fecundity and behavior [14]–[19] . This genetic factor also determines whether workers tolerate a single fertile queen ( monogyne social form ) or multiple queens ( polygyne social form ) in their colony . Colonies containing only homozygous Gp-9 BB workers accept only a single queen , whereas colonies containing both Gp-9 BB and Gp-9 Bb workers invariably accept multiple queens , but only those bearing a Gp-9 b haplotype [20]–[23] . The near complete absence of adult workers and queens with a bb genotype stems from the deleterious effects associated with the genomic region marked by the b allele , inducing homozygous females to die shortly after they eclose from the pupa [22] , [24] . The monogyne and polygyne social forms also differ in a number of important reproductive , behavioral , and life history traits besides colony queen number [25] , [26] , differences that are also completely associated with differences at the genomic region marked by Gp-9 . In contrast , there is a complete lack of differentiation at genes not tightly linked to Gp-9 , presumably because frequent matings between sexuals from sympatric monogyne and polygyne colonies result in extensive gene flow between the forms [22] , [27]–[29] . Colony queen number in S . invicta is regulated by the workers , which collectively decide which and how many queens from within or outside the colony are recruited as new egg-layers [21] , largely on the basis of chemical signals emanating from the queens [20] . Workers in monogyne colonies ( all of which possess the BB genotype ) accept only a single replacement queen that must also bear genotype BB , whereas workers in polygyne colonies accept multiple queens , each of which must possess the b haplotype . Significantly , the presence of as few as 5–10% of workers with the b haplotype induces the entire colony worker force , including BB workers , to become tolerant of multiple Bb queens and thus display the polygyne social phenotype [23] . Thus the genomic region marked by Gp-9 exerts indirect genetic effects [9] , in that the presence of the b variant in a colony induces changes in the social behavior of all colony members , even those lacking the b haplotype . Although the identity of the product of Gp-9 as an odorant binding protein ( OBP ) and other lines of evidence suggest that the gene may play a direct part in regulating social organization via a role in chemical communication , it remains an open question whether other genes tightly linked to Gp-9 ( possibly locked up in an inversion with it ) are also involved [22] . The first aim of this study was to investigate whether variation in the genomic region marked by Gp-9 is associated with differences in patterns of expression of genes other than Gp-9 in workers . The second aim was to study how the social environment ( i . e . , presence or absence of nestmate workers with the b allele ) can alter individual gene expression patterns . To answer these questions and begin to address the issue of how variation at a single Mendelian factor can directly and indirectly affect gene expression to produce a complex colony-level phenotype , we employed a fire ant microarray platform representing some 10 , 000 genes [30] .
To determine the effect of Gp-9 genotype on gene expression in focal individuals , we compared expression profiles between BB and Bb adult workers from 20 polygyne S . invicta colonies . This comparison revealed 39 genes consistently differentially expressed between workers of the two genotypes , of which about two-thirds were more highly expressed in Bb than BB workers ( Figure 1; see also confirmation of microarray data by quantitative RT-PCR [qRT-PCR] in Text S1 and Table S1 ) . Sixteen of these genes did not significantly match any sequence in the public databases ( BLASTX , threshold E-value = 1e-5 ) , ten matched predicted proteins of unknown function , and the remaining 13 matched genes with a known or inferred function ( Table 1 ) . Three gene categories were overrepresented among the genes differently expressed between workers of alternate genotypes ( Table 1 and Table S2; all P<0 . 05 ) . The first two are the allergen and odorant binding protein categories , which collectively include five genes likely to contribute to chemical signaling and response , the essential components of regulation of colony queen number and social organization . In ants , venom allergens are proteins released from the venom sac , an organ that , in queens , appears to store and release chemical signals allowing recognition by workers [25] . Similarly , the two odorant binding genes that encode members of the insect OBP protein family ( as does Gp-9 ) and the antennal chemosensory protein may be involved in pheromone transduction , thereby potentially influencing the abilities of workers of the two Gp-9 genotypes to recognize and discriminate among queens . Experiments from other systems are suggestive that changes in the expression levels of OBPs could influence discriminatory behavior by modulating the threshold for a particular response; differential regulation of OBPs has been observed in Drosophila following mating [31] , exposure to starvation stress [32] , and alcohol tolerance development after exposure to alcohol [33] . Additionally , genetic and biochemical evidence suggests that OBPs may interact combinatorially in odor discrimination [34] , [35] . The third overrepresented category comprises two transposons , which are of special interest with respect to properties that may be shared between the genomic region including Gp-9 and regions containing the sex-determining genes in species with sex chromosomes [36] . The b haplotype is found only in the polygyne social form , just as the Y chromosome is found only in males in species with male heterogamety . By analogy with the Y chromosome , theoretical predictions and empirical observations suggest that the Gp-9 b region should ( i ) accumulate genes beneficial in the polygyne social environment ( as the Y chromosome accumulates genes beneficial to male function [37] ) , ( ii ) evolve reduced recombination to preserve associations of genes advantageous for polygyny ( as occurs for genes advantageous to males on the Y chromosome [38] , [39] ) , and ( iii ) accumulate deleterious alleles and transposable elements ( because of reduced recombination [38] , [39] ) . Consistent with these expectations , the genomic region marked by Gp-9 is characterized by low recombination [40] , [41] , the b haplotype is a homozygous lethal [15] , [24] , [40] , and the piggyBac-like transposon , which is differentially expressed between workers of alternate genotypes , appears to occur almost exclusively in individuals possessing haplotype b ( data not shown ) . Thus the strong expression of at least this transposon in b-bearing workers , which constitutes the most extreme expression difference among the 13 genes with annotated matches ( Table 1 ) , likely reflects its unique insertion in b haplotypes . While this distribution could signify that the piggyBac-like transposon directly affects the differential expression of other candidate genes in BB and Bb workers , we note that , consistent with earlier protein electrophoresis data [40] , no significant difference in the expression levels of Gp-9 was detected between workers of the two genotypes; therefore , whatever elements control the differential expression of genes in parallel with Gp-9 genotype appear not to regulate the expression of Gp-9 itself . To determine the indirect effects of colony Gp-9 genotype composition as well as other aspects of the social environment on worker gene expression , while controlling for individual Gp-9 genotype , we compared profiles of adult workers bearing genotype BB between 20 polygyne and 20 monogyne colonies . This comparison revealed 91 genes consistently differentially expressed between workers of the alternate forms , of which over three-quarters were more highly expressed in polygyne than monogyne workers ( Figure 2; see also confirmation of microarray data by qRT-PCR in Text S1 and Table S1 ) . Forty-five of these genes did not significantly match any sequence in the public databases ( BLASTX , threshold E-value = 1e-5 ) , 13 matched predicted proteins of unknown function , and the remaining 33 matched previously annotated genes ( Table 2 ) . Three gene categories ( mitochondrial , prefoldin complex , and viral genes ) were overrepresented among the genes that were differentially expressed between BB workers from monogyne and polygyne colonies ( Table 2 and Table S2; all P<0 . 05 ) . The 11 genes encoding mitochondrial and prefoldin complex ( molecular chaperone ) proteins were all up-regulated in polygyne compared to monogyne workers . Increased mitochondrial gene expression may reflect increased oxidative metabolism , while increased prefoldin expression may indicate higher protein synthesis rates , possibly in relation to the relatively smaller size and higher metabolic rates of polygyne workers [42] . The pattern of expression of the six genes in the viral gene category is consistent with the expectation that differences in social organization affect susceptibility to pathogens and parasites . In the monogyne form , there is intense selection against susceptible infected individuals during independent colony founding , a stage that colonies of the polygyne form never display [26] . Accordingly , we found that workers in the polygyne form express more sequences corresponding to viral genes than their counterparts in the monogyne form , presumably because of relaxed selection and generally greater susceptibility in the former ( see also [43] ) . Based on sequence similarity and correlated expression across our experiments , we identified six gene products that likely represent three different viruses , a ssRNA negative-strand ( − ) virus and two ssRNA positive-strand ( + ) viruses , one of which is the SINV-2 virus [44] . All 20 polygyne study colonies showed evidence of infection with at least one virus ( mean number of viral types per colony , 2 . 5±0 . 67 ) , whereas only three of the 20 monogyne colonies showed evidence of infection , in all cases by a single viral type . Finally , the pattern of expression of another socially-regulated gene , this one encoding a defensin ( a class of small protein antibiotics active against viruses , bacteria , and fungi [45] ) , also is consistent with greater selection for resistance in the monogyne form , as this gene was more highly expressed in monogyne than polygyne workers . The numbers of genes differentially expressed in the genotype ( 39 ) and social form ( 91 ) comparisons are relatively low compared to the numbers expected based on other published microarray experiments . There are several possible explanations for this . First , the use of whole worker bodies as the source of RNA may decrease the probability of detecting genes whose level of expression varies among cells or tissues . Second , our comparisons were performed on groups of workers originating from different colonies , thus adding a colony-level effect to , and thus increasing the total variance in , gene expression . Finally , workers of alternate genotype or social form apparently exhibit fewer phenotypic differences than queens [20] , [21] , [46] , [47] , possibly reflecting the involvement of fewer differentially expressed genes in the former caste . Remarkably , there was almost no overlap between genes whose level of expression was influenced by the focal workers' Gp-9 genotypes and genes whose expression was influenced by the social environment , with only one of the 129 differentially expressed genes appearing in both categories . This demonstrates an almost complete decoupling of the direct effects of the genomic region marked by Gp-9 and the indirect effects mediated by social interactions within colonies . Moreover , there is little indication of an interaction between these direct and indirect effects; genes whose levels of expression depend on the Gp-9 genotype of focal individuals generally are expressed at similar levels in the two social forms when genotype is held constant ( Figure 1 ) , whereas genes whose levels of expression depend on the indirect influence of the social environment almost always are expressed at similar levels in polygyne workers of different Gp-9 genotypes ( Figure 2; see also Figure S1 ) . This study reveals that variation at the S . invicta genomic region marked by Gp-9 is associated with the differential expression in workers of a relatively small number of genes that , with the exception of the piggyBac-like transposon , presumably are unlinked to Gp-9 . A high proportion of these differentially expressed genes have putative functions implying a role in chemical signaling and behavior relevant to the regulation of colony queen number and , therefore , these genes may have a primary function in determining social organization . The number of such genes is unexpectedly low given the profound behavioral , physiological , and life-history differences between the two social forms and the fact that widespread changes in gene expression patterns can be observed after just a few generations of selection [48] , [49] . A perhaps more surprising finding is that worker gene expression profiles are significantly more strongly influenced by indirect effects associated with the Gp-9 genotypic composition within their colony than by the direct effect of their own Gp-9 genotype ( chi-squared test , P<0 . 001 ) , with the indirect-effect genes largely implicated in the secondary differences in colony social characteristics expected between the forms . While several studies have demonstrated that the social environment can modulate gene expression [50]–[53] , and others have revealed indirect genetic effects on phenotypes or levels of gene expression [10] , [54]–[60] , this is the first example of a naturally occurring polymorphic Mendelian element that affects gene expression in other group members . The finding of a complex genetic architecture directly and indirectly influencing the individual behaviors that , in aggregate , generate a fundamental colony-level social phenotype represents an unusual example of an “extended phenotype” [61] .
Colonies of S . invicta were collected near Athens , Georgia ( eight polygyne and eight monogyne colonies in 2004; eight polygyne and eight monogyne colonies in 2006 ) and near Hammond , Louisiana ( four polygyne and four monogyne colonies in 2006 ) , USA . All colonies were returned to the laboratory and reared for one month under standard conditions [62] . We determined the social form of each study colony using several lines of evidence . Nest density , worker size distribution , and nest brood composition were used to make initial identifications of social form in the field ( see [63] ) . Subsequently , polygyny was confirmed for each suspected polygyne colony by discovering two or more wingless inseminated ( reproductive ) queens , while monogyny was confirmed in each suspected monogyne colony by discovering a single , highly physogastric , wingless inseminated queen . The social form of each colony was further substantiated by electrophoretically detecting the b allele of Gp-9 in pooled samples of 20 female inhabitants of each polygyne colony and failing to detect the allele in such samples from each monogyne colony ( the b allele is completely diagnostic for polygyny in S . invicta in the USA [14] , [40] , [63] ) . From each polygyne colony , 24–40 medium-sized adult workers were haphazardly collected from the foraging area of each colony and individually flash-frozen with liquid nitrogen in tubes containing 1 g of 1 . 4 mm ceramic beads ( Quackenbush ) . From each monogyne colony , 10 medium-sized adult workers were collected in an identical fashion . For the 2004 samples , individual ants were homogenized in Trizol ( Invitrogen ) using a Fastprep bead shaker , and DNA and RNA were extracted using the manufacturer's recommended protocol . For the 2006 samples , individual ants were homogenized in RLT+ buffer , and DNA and RNA were extracted using the AllPrep RNA/DNA Mini Kit ( Qiagen ) . An RFLP analysis was used to determine the Gp-9 genotype of each individual in polygyne colonies [14] . We pooled the RNA from 7–10 BB workers and 7–10 Bb workers from each polygyne colony . Although bb workers generally are rare due to deleterious effects associated with the genotype [24] , [40] , we found 13 such workers from eight colonies . Two pooled bb samples were created , one for 2004 ( nine workers from four colonies ) and one for 2006 ( four workers from four colonies ) . These samples were hybridized ( after amplification ) but not included in the statistical analysis due to the small number and pooling of individuals across colonies . RNA from ten workers from each monogyne colony ( all with genotype BB ) was pooled by colony . For both the 2004 and 2006 samples , pooled total RNA was linearly amplified once ( Ambion MessageAmp II kit ) , then labeled using a modified version of the aminoallyl-labeling method in which reverse transcription is performed in the presence of aminoallyl-dUTP and the resulting cDNA is coupled to Cy3 or Cy5 fluorescent monomers [64] , [65] . Briefly , amplified RNA ( ∼5 µg ) was mixed with random 9mers ( 2 µg/ul ) , 0 . 5 µl of Alien mRNA Spike mix ( Stratagene ) , and water for a final volume of 17 . 5 µL . This RNA/primer mix was incubated for 10 min at 70°C , then held for 5 min on ice . Reverse transcription was performed for 2 h at 50°C after adding 6 µL of 5× first-strand buffer , 3 µL of 0 . 1 M DTT , 0 . 6 µL of 50× aminoallyl-dNTP mix ( 25 mM dATP , 25 mM dCTP , 25 mM dGTP , 15 mM dTTP , 10 mM aminoallyl-dUTP ) , 1 µL of RNAse inhibitor ( 15 U/µL , Invitrogen ) , and 2 µL of SuperScript III reverse transcriptase ( 200 U/µL , Invitrogen ) . The RNA was then hydrolyzed by adding 15 µL of 0 . 1 M NaOH and incubating for 10 min at 70°C . The pH was neutralized by adding 15 µL of 0 . 1 M HCl . The aminoallyl-labeled cDNA was purified with a modified Qiaquick PCR Purification Kit ( QIAGEN ) and coupled to Cy3 or Cy5 dyes [64] . The combined Cy3- and Cy5-labeled probes were purified using the Qiaquick PCR Purification Kit ( QIAGEN ) and eluted in 72 µL of elution buffer . After adding 13 . 5 µL of 20× SSC , 2 . 7 µL of yeast tRNA ( 2 µg/µL ) , 2 . 7 µL of polyA DNA ( 2µg/µL; Sigma ) , and 1 . 62 µL of 10% SDS , the probe was denatured at 100°C for 45 sec and hybridized to the microarray slides at 64°C overnight . Excess probe was removed by washing for 2×5 min in 2× SSC , 0 . 1% SDS; 2×1 min in 0 . 2× SSC; 1×1 min in 0 . 1× SSC; and 1×5 min in 0 . 1× SSC , 0 . 1% Triton at room temperature . Experimental samples were labeled with Cy3 and were hybridized against Cy5-labeled “common reference” RNA on our custom-made spotted cDNA microarrays . We employed a common reference design because not all samples provided enough amplified RNA for multiple hybridizations ( e . g . , for loop designs ) and because this allowed within-form comparisons of polygyne genotypes and between-form comparisons of BB workers . All experimental samples were labeled in Cy3 , allowing for unbiased comparisons . We used two different batches of reference RNA . For the 2004 samples , we pooled 25% of the amplified RNA from each experimental sample . For the 2006 samples , we amplified total RNA isolated en masse from hundreds of adult workers collected from the foraging area of 30 colonies ( eight and seven of each social form from Georgia and Louisiana , respectively ) . The microarrays were made from 22 , 560 independent cDNAs generated from a fire ant expressed sequence tag project and are estimated to represent 11 , 864 different genes [30] . Two different batches of microarrays were used , one set printed in 2004 and the other in 2006 . For both batches , only the 18 , 438 spots yielding a single PCR product ( representing 9 , 722 putative genes ) were considered in the analyses . Images of the competitive hybridization were obtained with an Agilent Technologies Scanner . The signal intensities for each spot were extracted from the images using GenePix software . After scanning , bad spots were flagged and the background-subtracted median foreground values were used as the intensity levels in the subsequent analysis . All spots with a positive intensity were considered for the subsequent analyses ( i . e . , no threshold filtering was used ) . Raw intensity data were converted to normalized log2 ratios using “print-tip specific” loess normalization ( within arrays; marray Bioconductor package , R [66] ) . Selected gene expression results were confirmed using qRT-PCR ( see Text S1 and Table S1 ) . Primers used for qRT-PCR are listed in Table S5 . For the genotypic analysis , we tested for differential expression of each gene between samples of BB and Bb workers in the 20 polygyne colonies using a 2-factor mixed-model ANOVA of the form:where Y , representing the reference/sample log-transformed ratio for a spot , is the sum of effects . The symbol μ represents the overall average log-transformed ratio for a given spot over all experiments . BATCH is a random effect ( denoted by ∼ ) with two levels , batch_2004 and batch_2006 , that accounts for the variation between hybridizations performed in the two different years ( this “year” effect also encapsulates the effects of two different batches of microarrays and of distinct reference RNAs ) . The term GENOTYPE captures the gene expression changes that are attributable to the BB and Bb genotypes . Finally , ε represents the measurement error . We did not include data for the bb workers in the statistical analysis due to the small number of samples . However , these samples yielded expression profiles that appeared similar to those of Bb workers ( but with even more marked differences from the profiles of BB workers , data not shown ) . For the social form analysis , we tested for differential expression of each gene between BB samples from 20 monogyne and 20 polygyne colonies by using the same 2-factor mixed-model ANOVA , but with the variable SOCIAL FORM ( monogyne or polygyne ) replacing the variable GENOTYPE . Analysis of variance calculations were performed in R . For the genotype comparison , 4 , 005 clones were removed from the ANOVA analysis because there were not enough data points for the F-statistic calculations ( for example , for a given clone all the batch_2004 samples for GENOTYPE BB had negative intensities and/or were flagged ) . However , because many genes are represented by multiple independent clones , 95 . 5% ( 9 , 288/9 , 722 ) of the putative genes on the microarray were present in the 14 , 433 clones used in the analyses . Similarly , for the social form analysis 3 , 791 clones were removed from the ANOVA analysis , with the remaining 14 , 647 clones representing 96 . 9% ( 9 , 419/9 , 722 ) of the putative genes . We restricted our analyses to the 73 and 139 cDNA clones that satisfied a false discovery rate ( FDR ) of 10% for the genotype and social form comparisons , respectively [67] . Duplicated clones on the microarray , independent cDNA clones representing the same gene , and sequences with greater than 95% sequence identity were merged and averaged , resulting in 39 genes with significantly different expression in the genotype comparison and 91 in the social form comparison . Expression levels presented in the figures are modified from the loess-normalized log2 expression ratios . For each gene , the batch effect ( derived from the ANOVA calculations ) was first subtracted from the loess-normalized log2 expression ratio . Then , the batch-adjusted expression ratios were normalized to the average across all experiments ( including the two bb hybridizations ) . Statistical significance of the expression differences detected by the ANOVA calculations was additionally evaluated by means of non-parametric Mann-Whitney tests conducted on the normalized , batch-adjusted data ( Figure S1 ) . Expression differences between polygyne workers of different Gp-9 genotype ( BB , Bb ) were confirmed to be highly significant for all 39 genes identified by the ANOVA ( all P<0 . 002 ) , as were expression differences between BB workers of different social form for all 91 genes identified by the ANOVA ( all P<0 . 002 ) . In contrast , among the 39 genes influenced by Gp-9 genotype , only seven ( 18% ) showed significantly different expression between monogyne and polygyne workers with the BB genotype ( 0 . 001<P<0 . 041 ) , while among the 91 genes influenced by social form , only three ( 3 . 3% ) showed significantly different expression between BB and Bb workers of the polygyne form ( 0 . 0001<P<0 . 047 ) ( see Figure S1 ) . Given the large number of these tests performed , some 5% of the significant results are presumed to represent Type I errors . Expression data were hierarchically clustered and examined using Cluster and Treeview [68] . We also performed SOM ( self-organizing map ) clustering of the experimental samples ( by array ) for both the genotype and social form comparisons ( data not shown ) . For the genotype comparisons , the samples clustered into two distinct groups according to genotype ( BB and Bb ) and no additional group was uncovered . Similarly , for the social form comparison , all the monogyne samples clustered together while the polygyne samples separated into two groups , those with high and those with low levels of viral gene expression . Because this analysis did not reveal any striking new patterns , the results are not presented in detail . Because previous annotations of the genes represented on the fire ant microarray [30] may be outdated , we performed new similarity searches against the non-redundant protein sequence database using the BLASTX algorithm [69] , [70] . All comparisons were performed on the Blast Network Service provided by the Swiss Institute for Bioinformatics ( release July 17 , 2007 ) . The default settings were used with an E-value threshold of 1e-5 , except where otherwise indicated . The accession number of the best match for each gene is reported in Tables 1 and 2 , except when it was an Apis mellifera gene derived from the Ensembl automatic annotation . In this case , we chose the next best hit , because little is known about gene function in A . mellifera , and the genome of this species has been removed from the current Ensembl releases . All BLASTX matches with E≤1 ( but limited to the top 20 ) are listed in Tables S3 and S4 . Each fire ant gene was also manually assigned to a descriptive category ( Tables 1 and 2 and Text S1 ) . The category putatively encapsulates the general function of each gene and is subjectively derived from examining the SwissProt or Ensembl database entries of the five best hits ( all E<1e-5 ) , with an emphasis on Gene Ontology , Interpro , and PANTHER annotations . To determine which categories were overrepresented in each set of differentially expressed genes , we used a one-tailed hypergeometric test implemented in R [71] , [72] . Gene expression data meet Minimum Information About a Microarray Experiment ( MIAME ) standards and have been deposited at Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo/ ) with accession number GSE11694 .
|
Fundamental research goals for scientists interested in social evolution are to determine the numbers and types of genes that directly regulate individual social behaviors as well as to understand how the social environment indirectly influences the expression of socially relevant traits . The fire ant Solenopsis invicta features a remarkable form of social variation in which the occurrence of two distinct social types that differ in colony queen number is associated with genetic differences at a genomic region marked by the gene Gp-9 . Our analyses of gene expression profiles in fire ant workers revealed that differences in Gp-9 genotype are associated with the differential expression of an unexpectedly small number of genes , many of which are predicted to function in chemical communication relevant to the regulation of colony queen number . Surprisingly , worker gene expression profiles are more strongly influenced by indirect effects associated with the social environment within their colony than by the direct effect of their own Gp-9 genotype . These results suggest a complex genetic architecture underlying the control of colony queen number in fire ants , with a single Mendelian factor directly regulating , and the social environment indirectly influencing , the expression of the individual behaviors that , in aggregate , yield an emergent colony social organization .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"evolutionary",
"biology",
"genetics",
"and",
"genomics/complex",
"traits",
"ecology/behavioral",
"ecology",
"genetics",
"and",
"genomics/gene",
"expression"
] |
2008
|
Genome-Wide Expression Patterns and the Genetic Architecture of a Fundamental Social Trait
|
Sepsis , a manifestation of the body’s inflammatory response to injury and infection , has a mortality rate of between 28%-50% and affects approximately 1 million patients annually in the United States . Currently , there are no therapies targeting the cellular/molecular processes driving sepsis that have demonstrated the ability to control this disease process in the clinical setting . We propose that this is in great part due to the considerable heterogeneity of the clinical trajectories that constitute clinical “sepsis , ” and that determining how this system can be controlled back into a state of health requires the application of concepts drawn from the field of dynamical systems . In this work , we consider the human immune system to be a random dynamical system , and investigate its potential controllability using an agent-based model of the innate immune response ( the Innate Immune Response ABM or IIRABM ) as a surrogate , proxy system . Simulation experiments with the IIRABM provide an explanation as to why single/limited cytokine perturbations at a single , or small number of , time points is unlikely to significantly improve the mortality rate of sepsis . We then use genetic algorithms ( GA ) to explore and characterize multi-targeted control strategies for the random dynamical immune system that guide it from a persistent , non-recovering inflammatory state ( functionally equivalent to the clinical states of systemic inflammatory response syndrome ( SIRS ) or sepsis ) to a state of health . We train the GA on a single parameter set with multiple stochastic replicates , and show that while the calculated results show good generalizability , more advanced strategies are needed to achieve the goal of adaptive personalized medicine . This work evaluating the extent of interventions needed to control a simplified surrogate model of sepsis provides insight into the scope of the clinical challenge , and can serve as a guide on the path towards true “precision control” of sepsis .
Approximately 1 million people will be diagnosed with sepsis , a condition with a mortality rate ranging from 28%-50% , each year [1] . Attempts to discover biologically-targeted therapies for sepsis have thus far been focused on manipulating a single mediator/cytokine , generally administered with either a single dose or over a very short course ( <72 hours ) [2 , 3] . Unfortunately , all these attempts have been unsuccessful [2 , 3] , likely due to both the nonlinear nature of the human inflammatory signaling network and the paucity of clinical time-course data to place network relationships in context , . In fact , we would propose that the universal failure to find effective cellular/molecular control strategies effective at the clinical level raises the question as to whether the system can be effectively controlled at all . The rationale for the current investigation is an attempt to address this fundamental question: can the trajectory of clinical sepsis be controlled , and if so , what is the scale and scope of the therapeutic interventions required to do so ? It is well known in biology that the systemic response to identical perturbations in genetically identical individuals ( i . e . , mice ) is governed according to some probability distribution . This small stochastic variability in response can propagate over time such that it ultimately leads to divergent phenotypes . It logically follows then that a single time point/single cytokine intervention is unlikely to be successful on a broad range of patients with a broad range of conditions that have led to the state of sepsis . The challenge , however , is that the range of possible interventions , which is a function of the number of potential molecular targets , the extent to which they are modified , the time at which such modification can occur and the combinations thereof , is staggering , and cannot be tractably investigated given the logistical and practical limitations of both experimental and clinical research . We propose to address this challenge by the use of evolutionary computing ( in the form of genetic algorithms ) applied to a sufficiently complex , albeit abstracted , computational model of sepsis . We have previously proposed that dynamic computational modeling , and specifically agent based modeling , can be used to represent mechanistic biological knowledge in a way that reproduces the non-linear dynamics of the real world system [4 , 5] . Agent-based models ( ABM ) have been used to study and model a wide variety of biological systems [6] , from general purpose anatomic/cell-for-cell representations of organ systems capable of reproducing multiple independent phenomena [5] to platforms for drug development [7] , and are frequently used to model non-linear dynamical systems such as the human immune system [8 , 9] . Specifically , was have previously developed an ABM of systemic inflammation , the Innate Immune Response agent-based model ( IIRABM ) . We propose to use the existing IIRABM as a surrogate proxy system [10] for the investigation of potential control strategies for sepsis . We note that while the model does not contain a comprehensive list of all signaling mediators present in the human body , all relevant cellular behaviors are represented . The named cytokines in this model are those that are most often associated with the available behavior rules in the current literature[11] . The IIRABM is a two-dimensional abstract representation of the human endothelial-blood interface . This abstraction is designed to model the endothelial-blood interface for a traumatic ( in the medical sense ) injury , and does so by representing this interface as the unwrapped internal vascular surface of a 2D projection of the terminus for a branch of the arterial vascular network . The closed circulatory surface can be represented as a torus , and this two-dimensional area makes up the space that is simulated by the model . The spatial scale is not directly mapped using this scheme . This abstraction serves two primary purposes: to allow circumferential access to the traumatic injury by the innate immune system , and to incorporate multiple levels of interaction between leukocytes and tissue . The IIRABM utilizes this abstraction to simulate the human inflammatory signaling network response to injury; the model has been calibrated such that it reproduces the general clinical trajectories of sepsis ( see [13] for details ) . The IIRABM operates by simulating multiple cell types and their interactions , including endothelial cells , macrophages , neutrophils , TH0 , TH1 , and TH2 cells as well as their associated precursor cells . The simulated system dies when total damage ( defined as aggregate endothelial cell damage ) exceeds 80%; this threshold represents the ability of current medical technologies to keep patients alive ( i . e . , through organ support machines ) in conditions that previously would have been lethal . The IIRABM is initiated using 5 parameters representing the size and nature of the injury/infection as well as a metric of the host’s resilience–initial injury size , microbial invasiveness , microbial toxigenesis , environmental toxicity , and host resilience . In previous work [12] , we have performed a sweep over these parameters to determine the plausible boundaries for conditions that could be considered clinically relevant . These are parameter sets that lead to all possible final outcomes when stochastically replicated–complete healing , death by infection , or death from immune dysregulation/sepsis . Additionally , the IIRABM has been used to perform in silico clinical trials of mediator-directed therapies via the inhibition or augmentation of single and specific cytokine synthesis pathways [11] . Those studies accurately reproduced unsuccessful clinical trials , as well as the non-efficacy of hypothetical interventions; however , to date no effective putative interventions have been discovered . The human innate immune response , and specifically in terms of sepsis , can be characterized through measurement of various biomarkers , including the pro-inflammatory and anti-inflammatory cytokines included in the IIRABM [13] . This implies that the system can be characterized by its state at some specific time , and therefore we apply that same perspective to our investigations with the IIRABM . At each time step , the IIRABM measures the total amount of cytokine present for all mediators in the model across the entire simulation . The ordered set of these cytokine measurement creates a high-dimensional trajectory through cytokine space that lasts throughout the duration of the simulation ( until the in silico patient heals completely or dies . Prior analysis of these trajectories has shown that the aggregate output of the IIRABM behaves as a Random Dynamical System ( RDS ) with chaotic features [12] ( in the sense that future simulation state can be sensitive to initial conditions ) . A Random Dynamical System [14] can be described by the triplet ( S , Γ , Q ) where S is the state space , Γ is a family of maps which maps S back onto itself ( often referred to as the “equations of motion” ) , and Q is some probability distribution on Γ . Simply put , an RDS is a system in which the equations of motion ( in this case , the equations which give the aggregate cytokine value for the system at a specific instance in time ) contain elements of randomness . A detailed discussion of this , and more formal definition , can be found in [15] . System state in the IIRABM can be defined by a vector of cytokines , C→ ( t ) , in which each element of C→ ( t ) , C→i ( t ) , is the total amount of an individual cytokine present in the simulation at that instance in time . At each time step , C→i ( t ) is given by: C→i ( t ) =∑n=1Nci , n ( t ) where ci , n ( t ) is the individual cytokine concentration seen by the endothelial cell at a specific grid point and N is the total number of endothelial cells which make up the endothelial surface . The random element comes from the calculation: ci , n ( t+1 ) =fn , i , EC ( c→ ( t ) ) +∑n , Apn , Afn , i , A ( c→ ( t ) ) +Di ( ∑n′ci , n′ ( t ) Di−ci , n ( t ) ) −λici , n ( t ) . We note that this is a descriptive equation and is not explicitly solved in the IIRABM . It should also be mentioned that the exact output generated by an individual cell at a given time-step is dependent on the cellular action-execution ordering . We utilize a shuffled ordering system to achieve emulated concurrency and avoid artifacts associated with fixed agent ordering . This equation contains four primary terms: fi , n , EC ( c→ ( t ) ) is the amount of cytokine i produced by endothelial cell n in response to current cytokine concentrations; ∑n , Apn , Afn , i , A ( c→ ( t ) ) is the sum of the responses of all other cell types , indexed with A , at the location of endothelial cell n , where pn , A is the population of a specific cell type at that location;Di ( ∑n′ci , n′ ( t ) Di−ci , n ( t ) ) represents the amount of cytokine i which diffuses into location n ( according to the associated diffusion constant , Di ) from the neighboring cell locations , denoted by n′ , and out of the current cell location into the space occupied by neighboring cells; and λici , n ( t ) is the amount of cytokine at cell location n that degrades at each time step according to degradation constant λi . The randomness in this model comes from the term pn , A . This population is random in two ways: 1 ) in the absence of cytokine markers , inflammatory cell movement follows a random walk , and 2 ) cellular differentiation of inflammatory cells proceeds according to a probability distribution parameterized by current cytokine concentrations . We note that this explicit randomness does not comprise the entire uncertainty present in the model . The aggregation of individual cytokine concentrations into a single measure conceals any spatial information present in the simulation . Due to this , future behaviors can appear to be random when , in reality , an epistemic uncertainty ( as opposed to intrinsic or “real” stochasticity ) prevents an accurate prognosis . A high-level overview of this system and visual depiction of the functional rules governing cytokine production can be seen in Fig 1 . Discovery of an effective or optimal intervention can then be viewed as a nonlinear optimization of a control problem [16–18] . Genetic Algorithms ( GA ) [19] and Evolutionary Computing have been used to optimize a wide variety [20] of nonlinear systems . Medical applications of GA include vaccine dosing strategies and protein binding site prediction [21 , 22] . Given a sufficiently validated model , and interpretation of the development of an intervention/control strategy to be a nonlinear optimization problem , GA’s can also be utilized to develop complex treatment and control strategies . We acknowledge that the intrinsic limitation to this approach is that the use of machine-learning approaches to develop therapeutic strategies requires the use of a computational model to generate a sufficient amount of data to inform the algorithm . Hence , the strategy computed by a GA is only as valid as the model upon which it was trained . While the behaviors and rules in the IIRABM have been validated [12 , 13] , we recognize that it is incomplete and less complex than its referent , the actual human immune cell/gene regulatory network; thus , we assert that if one is unable to reliably control this model , one would also be unable to reliably control its referent . While control of the IIRABM does not necessarily guarantee one could control the human immune system clinically , it does set a baseline for the types of control strategies one could expect to be successful as well as the data required to inform those control strategies .
The parameter set ( invasiveness = 2 , toxigenesis = 5 , host resilience = 0 . 1 , environmental toxicity = 2 , injury radius = 33 cells ) upon which the GA was trained on a specific patient drawn from a parameter set that led to a final outcome of death by sepsis with a probability of 82% . We chose this parameterization because it represented a challenging case to digitally “cure . ” Table 1 displays results from a variety of interventions that minimized their associated fitness function functions . The first column displays the number of sequential interventions ( spaced 100 time steps apart ) ; the second column displays the length of the intervention in simulation time steps ( where 100 time steps represents 12 hours ) ; the third column displays the general probability of death for this specific condition ( parameter set ) after intervention; the fourth column displays the probability of death for the specific in silico patient upon which the GA was trained . The general probability of death was calculated by starting the simulation with 100 distinct RNG seeds , applying the selected intervention , and running the simulation to completion . The probability of death for the specific patient ( meaning specific parameter set and specific RNG seed ) was calculated by starting the simulation with a specific RNG seed , reseeding the RNG 100 times at the start of the intervention sequence , applying the intervention , and running the simulation to completion ( either complete healing or death ) . We should note that the probability of death for a specific patient utilizes two distinct pieces of information: the random number seed and the time at which the probability is calculated . A specific patient’s probability of death evolves over time as their total level of health evolves due to simulation rules . The initial patient-specific mortality rate of 68% then is only valid for that particular random number seed at the particular instance in time 12 hours post-injury . ” We have previously characterized the stochastic properties of the trajectory space of the IIRABM in [12] , and note the existence of 3 distinct regions of state space: 1 ) the region of space under the influence of the Life Attractor–trajectories in this region will always lead to a state of complete healing; 2 ) the region of space under the influence of the Death attractor–trajectories in this region will always lead to complete system failure and death; and 3 ) the “stochastic zone”–trajectories in this region are influenced more strongly by random effects ( stochastic randomness and epistemic uncertainty ) than by the effects of either of the aforementioned attractors . At the time point when the intervention is started , the system is still in the “stochastic zone , ” and thus a given intervention will not have a guaranteed effect; rather , it changes the probabilistic future outcome distribution to favor a state corresponding to greater health . Thus , a subtle effect sustained over time causes a significant improvement in mortality rates . Throughout the course of this work , we initiate our interventions 12 hours post-simulated injury . The GA has found an effective intervention which begins at this point , however , if stochasticity leads an in silico patient to a sufficiently different location in cytokine state-space , the stochastic failure of the first stage of the intervention will lead to further failure at subsequent stages . The nature of these dynamics explains the phenomenon of “non-responders” to the putative interventions as defined at various time points during an individual trajectory; ultimately “non-responders” represent trajectories that are either not driven out of the stochastic zone by our proscribed duration of therapy or those whose stochastic response lead them to a region of parameter space in which the intervention no longer alters the future outcome probability distribution to favor a state of increased health . The best solution ( that which minimized the probability of death for both the specific patient case and the general case ) was found by using 8 sequential interventions . For the specific patient upon which the GA was trained , the probability of death was reduced from 68% to 12% through application in the intervention shown in Fig 2; for the general case , the probability of death with this intervention was reduced from 82% to 16% . This intervention is represented as a series of bar graphs in Fig 2 . The height of the bars along the y-axis represents the log2 of the intervention multiplier; the x-axis shows which cytokine has its protein synthesis augmented or inhibited according to its associated bar . In order to explore the generalizability of this intervention , we tested it against all clinically relevant parameter sets ( 8291 parameter sets ) that generated a mortality rate between 1 and 99%—this is the population of parameter sets that have been defined as “clinically relevant” [12] . The results of this test are shown in Fig 3 . In Panel A , we show the untreated mortality rate distribution for all parameter sets that generate at least two outcomes ( life and death , either from infection or from sepsis ) . This distribution is skewed towards the edge cases as the hyper-surface-area of the clinically relevant region is maximized at the boundaries of the space; this represents the total number of simulations that are at the “edge” of clinical relevance . This is discussed more extensively in previous work [12] . Panel 3B shows the overall shift in the mortality rate distribution for this population of parameter sets . Fig 3 , Panels C-F show the shifts in mortality rate distribution for more narrow ranges of mortality . We note that Fig 3 , panels E and F appear to have a bimodal distribution ( also hinted at in Panel D ) ; this bimodal distribution occurs because as the severity of the simulated condition increases , the likelihood of a given stochastic replicate being a “non-responder” to a standardized treatment also increases . The phenomenon of “non-responders” provides an excellent argument for adaptive personalized medicine , that is , the need to adapt a therapeutic strategy based on an individual’s response to therapy ( “in silico clinical trials of 1” ) , with the goal of returning a system to a state of health . To further explore this , we investigated the cases that were unable to be cured using the derived intervention . Fig 4A shows the total oxygen deficit trajectories for the average of all the simulations that healed when using the optimal intervention and for a single instance of the simulation that does not heal; Fig 4B shows the neutrophil population and Fig 4C shows the total systemic GCSF . After three sequential interventions , it is apparent that this patient has a stronger response to GCSF stimulation , and thus a higher neutrophil population . As this difference continues to compound over time , interventions later in the sequence lose their efficacy ( as they are optimized for a system in a different state ) . We pause the simulation at this point and use the GA to find a sequence of 5 more interventions that could heal the system . The newly derived final 5 interventions are compared with the old intervention in Fig 5 . This shifted the probability of death for that patient ( at the time in which the intervention is changed ) from 75% to 8% . This success suggests that an algorithm that adapts interventions based on system response could be more successful than a GA without the ability to adapt based on system response .
The human inflammatory signaling network as represented by the IIRABM is a nonlinear and stochastic system that lacks a unique set of analytical equations that can adequately describe it . Designing an effective policy to control such a system requires the exploration of an astronomically large search space–in the case of 8 sequential interventions with 9 defined augmentation/inhibition strengths , there are approximately 1091 combinations that can be applied to the system . Given the size of this space , we make no claim that we have found the globally optimal intervention for our model , given a fixed set of parameters; rather , we have shown that GA can be used to find a “good-enough” solution that shows success ( though not perfection ) at treating a range of conditions leading to death by sepsis . Longer duration therapies were more successful than shorter therapies due to the stochastic nature of the IIRABM and its response to perturbations ( S1 Fig ) . All interventions are performed on in silico patients whose system state is either in the stochastic zone or under the influence of the death attractor ( in which case their trajectories would have to pass through the stochastic zone on the way to health ) . When the system state is in the stochastic zone , the effects of randomness can be stronger the effects of either of the attractors which influence the system; subsequent states fall will within some probability distribution ( which is discoverable via pausing the simulation and reseeding the random number generator several times ) based on the current state . Just as the system evolves in a stochastic nature when left unattended , its responses to perturbations will also be stochastic , however the possible future distribution of states is now based on the components of the perturbation as well as the current system state . Successful interventions will then shift the future state probability distribution towards health; more successful interventions will generate a larger move towards health more often , though there can still exist a non-zero probability that the system state evolves negatively rather than beneficially . While GA is quite successful at healing the IIRABM under a wide range of conditions , it has a few drawbacks which preclude it from being the ideal solution: 1 ) more extreme conditions ( either very large injuries or extremely virulent bacteria ) require either a finer degree of control than is computationally tractable using GA , as each sequential intervention multiplies the size of the search space by a factor of 912 ( approximately 5 billion ) , or more aggressive interventions; intervention multipliers were limited to a small set of values we considered clinically tractable–removing this constraint would lead to an unconstrained search , increasing computational cost and potentially generating implausible interventions; 2 ) adjusting the temporal density of interventions requires a new run of the GA , which can be computationally expensive; 3 ) the GA does not have the ability to react to non-responders and adjust the intervention accordingly–rather , it finds the single sequence of interventions which ( locally ) maximizes the survival probability for a given patient population . As such we note that we do not claim to find the absolute optimum sequence of interventions for a given parameter set due to the lack of general formal analytical convergence proofs for genetic algorithms [23 , 24] and the fact that it is computationally intractable to explore the entire intervention space , especially for multiple sequential interventions . Additionally , many interventions can have opposing effects with the possibility of cancelling each other out ( i . e . , GCSF augmentation leading to an increasing neutrophil population . Future work will incorporate alternative machine learning algorithms , including deep reinforcement learning [25] and Long Short Term Memory neural networks [26] for time-series prediction of aggregate cytokine values . Both of these techniques would base putative interventions on the sequence of events that led up to the intervention . In this sense , the learning algorithm would adapt the putative intervention to an individual system state rather than attempting to develop a single broad policy that would cover a certain class of injury . As these approaches will incorporate more “adaption” to where the system happens to be in trajectory space ( overcoming a recognized limitation of GAs ) they will have more direct clinical relevance as they will explore several different sampling strategies , including those that are currently clinically infeasible , to determine the types and frequencies of sampling necessary to characterize the trajectory of a random dynamical system . These results strengthen the case for adaptive personalized medicine ( as defined above ) . Rather than searching an astronomically large space for a utilitarian intervention , personalized medicine techniques would respond to cytokine dynamics with an individualized intervention for each patient at varying temporal scales . These things are theoretically possible using GA , but at the present time , the computational expense limits the practicality of using this technique to personalize treatments .
The current investigation involves providing a proof-of-concept example of identifying whether effective controls can be found for the IIRABM , and , by extension , for the treatment of sepsis . This initial proof-of-concept constrains the problem by focusing a particular parameter configuration of the IIRABM with a mortality rate of 68% . We first employed a genetic algorithm ( GA ) to search for possible therapeutic interventions , and then examined the generalizability of the optimal solutions across a wider range of stochastic replicates and additional parameter sets with similar overall baseline mortality rates . The IIRABM was implemented in C++ and simulations were performed on the Edison Cray XC30 Supercomputer at the National Energy Research Scientific Computing Center and on the Beagle Cray XE6 Supercomputer at the University of Chicago . We chose to train the GA on a single parameter set for 2 primary reasons: 1 ) A large number of parameter sets can generate a realistic sepsis condition . Parameter sets that are relatively similar tend to have similar mortality rates and similar simulated length-of-stays in the ICU . In this work , we explored the generalizability of intervention solutions derived using the GA to help assess the future potential of utilizing GA as an in silico drug-development technique . 2 ) The computational cost of running a genetic algorithm is substantial . A single instance of this model simulating 90 days in the ICU will take approximately 4 minutes ( depending on the speed of the processor running the simulation ) . Each iteration of the GA gathers data from up to 2000 independent simulation runs ( where each run repeats the simulation 10 times using 10 stochastic replicates ) , and the GA can run for 1000 or more generations before convergence when considering multi-stage interventions; each GA experiment can utilize up to 1 , 000 , 000 cpu-hrs ( equivalent to $100 , 000 ) [27] . We have selected a parameter set ( invasiveness = 2 , toxigenesis = 5 , host resilience = 0 . 1 , environmental toxicity = 2 ) which leads to simulated death approximately 80% of the time in the general case . This is illustrated in Fig 6A; here we show total health trajectories for 100 stochastic replicates of the above parameter set . The systemic oxygen deficit ( a proxy for total system damage ) is plotted against the time for which the simulation ran . Trajectories shown in red are simulations that die and trajectories shown in blue are simulations that heal completely . From this set of outcome trajectories , we have chosen a specific trajectory/in silico patient with which to train the GA . Given this specific patient’s location in state space 12 hours post injury , we estimate that they have a 68% chance of death . Note that this mortality rate is calculated in a different manner than the rate referenced above . In the general case , this parameter set generates a mortality rate of 80% , meaning that when the simulation’s random number generator ( RNG ) is given 100 unique starting seeds , 80 of those simulations will die ( a population level outcome distribution ) . For the individual patient mortality rate , the simulation is paused at the time step before the intervention would be applied ( in this case , 12 hrs post injury ) ; the simulation’s RNG is then reseeded a number of times to discover the mortality rate for a specific in silico patient at a specific moment in time . Thus , the general case mortality rate represents the mortality rate for a specific parameter set , while the individual patient mortality rate represents the mortality rate for a specific parameter set at a specific instance in time . As discussed above , an individual patient does not have a fixed fate at any given point in time ( as long as their trajectory remains in the stochastic zone ) and their probabilistic future outcome distribution is time dependent and will evolve with the system . This is shown in Fig 6B; this image displays 100 random number generator ( RNG ) re-seedings , reseeded 12 hours post injury , of the specific trajectory we have chosen . The utilization of stochastic replicates on a single patient allows the GA to sample the full range of responses possible for a given intervention . We have selected a set of cytokines and associated targets ( Platelet-activating factor ( PAF ) , Tumor necrosis factor alpha ( TNFα ) , Soluble tumor necrosis factor receptors ( sTNFr ) , Interleukin-1 ( IL1 ) , soluble interleukin-1 receptors ( sIL1r ) , Interleukin-1 receptor antagonist ( IL1ra ) , Interferon-gamma IFNγ , Interleukin-4 ( IL4 ) , Interleukin-8 ( IL8 ) , Interleukin-10 ( IL10 ) , Interleukin-12 ( IL12 ) , and Granulocyte colony-stimulating factor ( GCSF ) ) , which are the principal drivers of the inflammatory/immune dynamics expressed by the model . In order to search for an optimal intervention strategy , we allow production of each of these targets to be augmented or inhibited alone or as a group . For the purposes of this study , we consider an intervention to be a set of signaling augmentations/inhibitions . An individual’s chromosome is then a 1x12n vector , where n is the number of independent sequential interventions , containing this information . The augmentation and inhibition values are selected from the set: {0 . 05 , 0 . 25 , 0 . 5 , 0 . 75 , 1 , 2 , 5 , 10 , 20}; inhibitory values are multiplicative ( i . e . , Si , new = Si , old * Ii ) where Si , new represents the modified signal value , Si , old represents the signal value being modified , and Ii represents the vector element of the intervention vector which applies to the signal . Augmentations are additive rather than multiplicative ( i . e . , Si , new = Si , old + ( Ii – 1 ) ) . This allows for a sustained application of the intervention; were the augmentations to be multiplicative , then the simulation would quickly produce results outside the realm of plausibility due to an exponential explosion in concentrations of cytokines whose values have been multiplied . The descriptive local cytokine update function then becomes: ci , n ( t+1 ) ={Fi , n ( c→ ( t ) ) *TiforTi≤1Fi , n ( c→ ( t ) ) + ( Ti−1 ) forTi>1 where Fi , n ( c→ ( t ) ) = ( fn , i , EC ( c→ ( t ) ) +∑n , Apn , Afn , i , A ( c→ ( t ) ) +Di ( ∑n′ci , n′ ( t ) Di−ci , n ( t ) ) −λici , n ( t ) ) and Ti represents the cytokine augmentation/inhibition value for that state of the treatment . All of these cytokine augmentations/inhibitions represent theoretical drug targets rather than actual drugs . While there are drugs that target some of these specific pathways , i . e . Filgrastim for GCSF , incorporating specific drugs , their dosage-dependent effects , and their change in efficacy over time , presumably based on the drug’s half-life , is beyond the scope of this work . We define the fitness function as F=∑i=1n ( ODi+Xi ) , where OD represents the Oxygen Deficit , and is an inverse measure of the patient’s total health at the time when the fitness is evaluated and X is a measure of the total infectious load in the system at the time when the fitness is evaluated , and n is the number of stochastic replicates used . An optimal solution will minimize this fitness function ( and thus maximize the patient’s total health ) . We train the GA on 10 stochastic replicates of an individual trajectory . These replicates are generated by re-seeding the RNG at the time point when the intervention begins . This allows the GA to learn from a possible range of responses an individual can have to a given intervention . The fitness is evaluated 12 hours after the application of the final intervention ( which has a maximum duration of 12 hours ) . After the fitness is evaluated , we use the tournament selection method [28 , 29] , with a tournament size of 2 , to generate the breeding population . When breeding , each pair of progenitors produces two progeny using a uniform crossover operator [30] . We continue this process until the algorithm has converged to a small set of possible solutions , and select the solution that leads to the minimal fitness value as the intervention to be tested . We evaluate the intervention according to three criteria: 1 ) outcome distribution in specific patient upon which the GA was trained; 2 ) outcome distribution in a population of patients exposed to the same injury and microbial infection; and 3 ) outcome distribution in a population of patients exposed to a range of injuries and microbial infections . Due to the enormous size of the intervention-possibility space , we cannot definitely state that our intervention strategy is the globally optimal intervention strategy , and in fact is likely not; rather , it is a near-optimal strategy stuck in some local minima of the fitness function . We recognize that the utilization of alternate crossover or mutation strategies would lead to alternative intervention strategies , possibly even strategies would be rated as more “fit” than the strategy that we found; however , alternative strategies would still suffer from the same primary weakness identified with this approach–the inability to adapt therapeutic strategy based on system response . Rather than attempt to make small improvements in the fitness of the strategy , future work will focus on the development of a GA-derived intervention strategy with this adaptive ability . Source code for the IIRABM with GA capability is available at https://bitbucket . org/cockrell/iirabm_public . Certain GA experiments used the EMEWS [31] framework with DEAP [32] to automate the GA process .
|
Sepsis , characterized by the body’s inflammatory response to injury and infection , has a mortality rate of between 28%-50% and affects approximately 1 million patients annually in the United States . Currently , there are no therapies targeting the cellular/molecular processes driving sepsis that have demonstrated the ability to control this disease process . In this work , we utilize a computational model of the human immune response to infectious injury to offer an explanation as to why previously attempted treatment strategies are inadequate and why the current approach to drug/therapy-development is inadequate . We then use evolutionary computation algorithms to explore drug-intervention space using this same computational model as a surrogate system for human sepsis to identify the scale and scope of interventions to successfully control sepsis , as well as the types of data needed to derive these interventions . We demonstrate that multi-point and time-dependent varying controls are necessary and able to control the cytokine network dynamics of the immune system .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"death",
"rates",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"applied",
"mathematics",
"endothelial",
"cells",
"immunology",
"genetic",
"algorithms",
"epithelial",
"cells",
"simulation",
"and",
"modeling",
"algorithms",
"developmental",
"biology",
"sepsis",
"systems",
"science",
"mathematics",
"signs",
"and",
"symptoms",
"probability",
"distribution",
"molecular",
"development",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"inflammation",
"animal",
"cells",
"biological",
"tissue",
"dynamical",
"systems",
"immune",
"response",
"probability",
"theory",
"immune",
"system",
"population",
"metrics",
"diagnostic",
"medicine",
"cell",
"biology",
"anatomy",
"physiology",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences"
] |
2018
|
Examining the controllability of sepsis using genetic algorithms on an agent-based model of systemic inflammation
|
Cell-mediated immunity is essential in protection against rickettsial illnesses , but the role of neutrophils in these intracellular vasculotropic infections remains unclear . This study analyzed the plasma levels of nucleosomes , FSAP-activation ( nucleosome-releasing factor ) , and neutrophil activation , as evidenced by neutrophil-elastase ( ELA ) complexes , in sympatric Lao patients with scrub typhus and murine typhus . In acute scrub typhus elevated nucleosome levels correlated with lower GCS scores , raised respiratory rate , jaundice and impaired liver function , whereas neutrophil activation correlated with fibrinolysis and high IL-8 plasma levels , a recently identified predictor of severe disease and mortality . Nucleosome and ELA complex levels were associated with a 4 . 8-fold and 4-fold increased risk of developing severe scrub typhus , beyond cut off values of 1 , 040 U/ml for nucleosomes and 275 U/ml for ELA complexes respectively . In murine typhus , nucleosome levels associated with pro-inflammatory cytokines and the duration of illness , while ELA complexes correlated strongly with inflammation markers , jaundice and increased respiratory rates . This study found strong correlations between circulating nucleosomes and neutrophil activation in patients with scrub typhus , but not murine typhus , providing indirect evidence that nucleosomes could originate from neutrophil extracellular trap ( NET ) degradation . High circulating plasma nucleosomes and ELA complexes represent independent risk factors for developing severe complications in scrub typhus . As nucleosomes and histones exposed on NETs are highly cytotoxic to endothelial cells and are strongly pro-coagulant , neutrophil-derived nucleosomes could contribute to vascular damage , the pro-coagulant state and exacerbation of disease in scrub typhus , thus indicating a detrimental role of neutrophil activation . The data suggest that increased neutrophil activation relates to disease progression and severe complications , and increased plasma levels of nucleosomes and ELA complexes represent independent risk factors for developing severe scrub typhus .
Typhus-like illnesses , represented by rickettsioses , leptospirosis , dengue and typhoid , pose a significant challenge to tropical infectious disease clinicians due to their non-specific clinical presentations and difficulties in laboratory diagnosis . Recent studies have revealed tropical rickettsioses as leading causes of treatable fevers in Southeast Asia [1–4] . Scrub typhus , caused by infection with Orientia tsutsugamushi , and murine typhus , caused by Rickettsia typhi , are the predominant species infecting humans in Asia , and probably represent the most frequent , neglected , severe but easily treatable diseases in the world [5 , 6] . The pathophysiology of scrub typhus and murine typhus in humans remains poorly understood . Mouse models and human post-mortem samples point towards the endothelium as the target of late-stage infection [7–9] . In murine typhus the pathobiology is vasculitic , with R . typhi targeting the endothelium [9 , 10] . Recent human ex vivo data have revealed differences in endothelial host-pathogen interactions of scrub typhus and murine typhus , based on soluble adhesion molecules , coagulation and inflammation profiles [11 , 12] . Histopathological studies of eschar skin biopsies have shown an early polymorphonuclear neutrophil ( PMN ) response in the upper dermis of the eschar , whereas deeper in the dermis the PMNs mix with the predominant mononuclear cell infiltrates , where O . tsutsugamushi localize mainly within antigen-presenting cells ( APCs ) [13] . However , O . tsutsugamushi when phagocytosed by PMNs in vitro can escape phagolysosomal fusion and localize freely in the cytoplasm [14] . Neutrophilia is a common finding in patients with scrub typhus and PMNs can be found in perivascular infiltrates of affected organs and cerebrospinal fluid in severe disease [11 , 15] . Recently , IL-8 , which promotes migration of neutrophils to infection sites and a major neutrophil-activating factor , was associated with scrub typhus disease severity [16 , 17] . The evidence of PMNs playing a role in the host defense against murine typhus is very limited . Clinical reports rarely describe neutrophilia and profiles of soluble adhesion molecules in patients with acute murine typhus suggest endothelial rather than leucocyte activation [11 , 18] . However , PMNs have been described in the perivascular infiltrates and portal triads of endothelial-tropic spotted fever rickettsiosis , with similar pathogenic mechanisms to R . typhi infections [19] . Tissue anoxia , metabolic disturbances secondary to vasculitic changes and the host immune response contribute to the pathology of typhus , raising questions about the role of PMNs in the pathogenesis of severe typhus . Basically , PMNs eliminate pathogens by phagocytosis followed by degradation of the pathogens in phagosomes by the NADPH-oxygenase machinery as well as by antibacterial neutrophilic proteins and proteases . Neutrophil degranulation with release of contents of neutrophilic granules to the extracellular milieu can eliminate extracellular pathogens , but can also result in collateral damage of endogenous structures , e . g . endothelial or parenchymal cells . Recent publications have demonstrated PMNs can also form neutrophil extracellular traps ( NETs ) [20] . NETs are regarded as a form of innate immune response that bind and kill microorganisms , and prevent their dissemination . During NET formation , DNA and DNA-binding proteins are extruded from neutrophils exposing a mesh consisting of nucleosomes , histones and neutrophil proteases such as elastase . Nucleosomes consist of a core octamer with 2 copies of histones H2A , H2B , H3 , and H4 , around which a helical DNA chain of 146 base pairs is wrapped [21] . They can be actively released by factor VII–activating protease ( FSAP ) from apoptotic cells and in concert with DNAse 1 from necrotic cells [22 , 23] . Nucleosomes and histones exposed on these NETs have been reported to be highly cytotoxic to endothelial cells in vitro and are strongly pro-coagulant [24–26] . Concentrations of circulating nucleosomes have been associated with mortality and correlated with markers of coagulation , inflammation and neutrophil activation in sepsis [21 , 27] . Moreover , circulating nucleosomes together with circulating markers for neutrophil activation have been reported to be robust markers to measure NET formation in the circulation [25 , 28] . PMNs are more important in the host defense of some typhus-like illnesses , i . e . leptospirosis and typhoid than others i . e . dengue . To date , there are no data on PMN activation in scrub typhus and murine typhus . Therefore , we measured levels of circulating nucleosomes and systemic neutrophil activation as evidenced by released neutrophil elastase complexed with its natural inhibitor α1-antitrypsin ( ELA complexes ) as well as FSAP activation in form of FSAP- α2-antiplasmin ( FSAP-AP ) complexes in plasma of patients with confirmed scrub typhus and murine typhus , respectively
A total of 248 non-pregnant patients with clinical suspicion of scrub typhus ( ST ) or murine typhus ( MT ) were prospectively recruited between Feb 2005 and Dec 2007 at Mahosot Hospital , Vientiane , Lao PDR . Of these , patients with paired positive dynamic serology were randomly selected ( ST n = 65 , MT n = 64 ) for measuring plasma levels of nucleosomes , human neutrophil elastase- α1-antitrypsin complexes ( ELA ) and FSAP- α2-antiplasmin ( FSAP-AP ) complexes . Healthy controls ( total n = 47 ) consisted of Dutch ( n = 7 ) and recent Lao blood donors ( n = 40 ) . The study was approved by the National Ethics Committee For Health Research , Ministry of Public Health , Lao PDR , and the Oxford Tropical Research Ethics Committee , UK . All patients gave written informed consent prior to sample collection . Minors were participants in this study , and a parent or guardian of any child participant provided written informed consent on their behalf . The definitive diagnoses of scrub typhus and murine typhus were based on ≥ 4 fold dynamic rise in IgM and IgG IFA titers for paired serum samples , which represents the current serological gold standard [29] . Slides prepared and standardized by the Australian Rickettsial Reference Laboratory served for anti-O . tsutsugamushi antibody detection ( using pooled Karp , Kato , Gilliam antigens ) and anti-R . typhi antibody detection ( R . typhi Wilmington strain antigens ) . Bacteremic patients on admission , were identified by realtime PCR , targeting the groEL gene for scrub typhus [30] and the ompB gene for murine typhus [31] , as previously described with modification of the endpoint visualisation by intercalating SYBR green [32] . DNA templates were extracted from 200 μL of Buffy coat collected from EDTA-treated full blood samples ( Qiagen Mini Blood kit , Qiagen , Germantown , MD , USA ) . Nucleosome plasma levels were measured with a quantitative ELISA assay as described [21 , 33] . Briefly , a monoclonal antibody ( CLB-ANA/60 ) reacting with histone 3 captured the plasma nucleosomes , which were then detected by monoclonal antibody ( CLB-ANA/58 F ( ab ) 2 ) after binding to complexes of histone 2A , histone 2B and dsDNA ( CLB , Amsterdam , the Netherlands ) . FSAP-AP complexes in plasma were quantitated as described [22] . Briefly , a mouse monoclonal antibody against AP ( AAP-20 ) was used for catching the FSAP-AP complexes and an antibody targeting the light chain of FSAP ( anti-FSAP-4 ) was used for detection . Human neutrophil elastase-α1-antitrypsin ( ELA ) complexes were measured by an ELISA as described earlier [34 , 35] . Briefly , as a catching and detecting antibody a polyclonal rabbit anti-human neutrophil elastase antibody ( 1 . 5 μg/ml; Sanquin , Amsterdam , The Netherlands ) and a biotinylated monoclonal anti-α1-antitrypsin antibody ( 1 μg/ml ) has been used , respectively . Plasma concentrations of coagulation parameters: Thrombin-antithrombin ( TAT ) complexes , soluble tissue factor ( sTF ) , tissue- type plasminogen activator ( tPA ) , soluble thrombomodulin ( sTM ) and von Willebrand factor ( VWF ) were measured by commercially available ELISAs; Antithrombin ( AT ) , plasminogen activator ( PA ) activity and plasminogen activator inhibitor-1 ( PAI-1 ) activity were measured with automated amidolytic techniques; Protein C ( PC ) activity was determined with an amidolytic assay using chromogenic substrate S2366 ( Chromogenix , Milan , Italy ) as described [12] . Plasma concentrations of TNFα , interleukin ( IL ) -1β , IL-6 , IL-8 , IL-10 and IL-12 were measured as previously described [12] . Skin biopsies of eschars were performed with sterile disposable 3 mm circular punch biopsies ( Stiefel Laboratories Inc . , Offenbach , Germany ) to sample the necrotic edge with perifocal inflamed skin , after local anesthesia with 1% lidocaine . Biopsies in this patient cohort were performed for co-localisation and pathophysiology studies . For this study , consenting patients ( n = 2 ) had a biopsy taken , which was fixed in 10% formalin and processed later into paraffin blocks . Results are reported as medians and interquartile ranges ( IQR ) unless noted otherwise . Patient data were compared between groups using the Kruskal-Wallis test or Mann-Whitney U tests . Correlations between variables were assessed using Spearman’s rank correlation and Pearson’s product correlation where appropriate , corrected for multiple testing with the Bonferroni method . Receiver operating characteristic ( ROC ) analysis provided the selection of optimal cut off points for circulating nucleosomes and EA complexes between controls and cases , as well as between severe and non-severe disease . The associations between circulating nucleosomes and EA complexes and the association of developing complicated disease were explored by means of logistic regression analysis and are expressed as odds ratios ( OR ) with corresponding 95%CIs . Statistical significance was set at p<0 . 05 . All statistical analyses were calculated using Stata/MP 11 . 0 ( Stata Corp . , College Station , Texas , USA ) .
Patients in both disease groups did not differ significantly by age , gender , days of fever and symptoms prior to admission ( Table 1 ) . The presence of an eschar , lymphadenopathy and muco-cutaneous hemorrhages were significantly associated with scrub typhus patients ( all p-values <0 . 0001 ) . All typhus patients survived to discharge . Laboratory parameters outside the normal range that differed between the two forms of typhus were plasma albumin , C-reactive protein ( CRP ) and lactate dehydrogenase ( Table 1 ) . The median ( IQR ) bacterial loads differed significantly between bacteremic patient subgroups with 2 , 100 copies/mL ( 800–4 , 500 ) and 700 copies/mL ( 400–1100 ) in full blood , for scrub typhus and murine typhus patients respectively ( p = 0 . 0001 ) . The plasma levels of nucleosomes , ELA and FSAP-AP complexes in both scrub typhus and murine typhus were significantly higher than in Dutch and Lao controls ( all p-values <0 . 0001 ) . The plasma levels of nucleosomes and ELA complexes were significantly higher in scrub typhus than in murine typhus ( p = 0 . 001 and p<0 . 001 respectively , Fig 1 ) . The median ( interquartile range ) plasma levels for nucleosomes , ELA and FSAP complexes were 219 [72–471] , 977 [511–1647] , 6 . 8 [2–24] U/ml for scrub typhus and 101 [49–201] , 440 [260–1046] , 3 . 6 [2–24] U/ml for murine typhus and 2 [0–4] , 46 [31–65] , 0 . 2 [0 . 2–0 . 4] U/ml for Lao controls and 4 [3–7] , 38 [28–67] , 0 . 2 [0 . 2–0 . 4] U/ml for Dutch controls , respectively . Plasma levels of circulating nucleosomes and ELA complexes correlated strongly with each other in the pooled typhus patient groups ( Spearman correlation coefficient Rho 0 . 452; p<0 . 001 ) , but not within pooled or between control groups . This study included two sets of controls to ensure that the background levels of all markers measured did not differ between the local Lao healthy population and European Caucasians . Healthy control values were previously characterized for these assays , with no statistical differences of results in and between the control groups [27 , 34] . Stratification by disease into scrub typhus versus murine typhus , demonstrated differential profiles: in scrub typhus nucleosomes correlated strongly with ELA complexes ( Rho 0 . 495; p = 0 . 005 ) , whereas in patients with murine typhus nucleosomes correlated strongly with FSAP complexes ( Rho 0 . 492; p = 0 . 01 ) ( Table 2 ) . Neutrophil counts were significantly higher in scrub typhus patients than in murine typhus patients with a median ( IQR ) of 6 , 392/mL ( 4 , 240–9 , 005 ) and 4 , 674/mL ( 3 , 822–6 , 831 ) , respectively ( p = 0 . 004 ) ( Table 1 ) . When stratified for disease severity , the non-severe scrub typhus patients had a significantly higher neutrophil count than non-severe murine typhus patients ( p = 0 . 001 ) , while neutrophil counts were similar for both diseases in severe disease ( p = 0 . 87 ) . In scrub typhus , PMN counts correlated ( spearman’s Rho , p-value ) with plasma levels of IL-6 ( Rho = 0 . 257 , p = 0 . 01 ) , bleeding ( Rho = 0 . 26 , p = 0 . 041 ) seizure ( Rho = 0 . 371 , p = 0 . 002 ) , confusion ( Rho = 0 . 304 , p = 0 . 015 ) and GCS ( Rho = 0 . 22 , p = 0 . 023 ) , but not with ELA complex levels ( p = 0 . 314 ) . In murine typhus , PMN counts correlated well ( Spearman’s Rho , p-value ) with plasma levels of protein C ( Rho = 0 . 37 , p = 0 . 0089 ) , CRP ( Rho = 0 . 31 , p = 0 . 0026 ) , and the presence of skin rash ( Rho = 0 . 23 , p = 0 . 031 ) , but not with ELA complex levels ( p = 0 . 074 ) . O . tsutsugamushi co-localized with neutrophils in high-density neutrophil infiltrates predominantly located adjacent to the central necrotic zone of the eschar lesion , at the dermal-epidermal junction along the necrotic margin and in superficial dermal infiltrates ( Fig 2 ) . O . tsutsugamushi and partially karyorrhectic neutrophils were also found in high densities embedded in the necrotic crust , providing descriptive evidence that the easily accessible and painlessly detachable necrotic crust represents a useful specimen for molecular diagnostics . To date no validated algorithm for assessing the severity of rickettsial infections is available . In this study severe typhus was defined when the following admission clinical criteria were apparent: a Glasgow Comma Score ( GCS ) of <15 , clinical evidence of meningism or central nervous system involvement and/or a respiratory rate ( RR ) of >20/min . In total , 27/128 ( 21% ) of scrub typhus and 14/102 ( 14% ) of murine typhus patients had evidence for CNS and/or respiratory tract involvement and/or reduced vigilance at presentation . Median ( IQR ) plasma levels of nucleosomes and ELA complexes were significantly higher in patients with severe scrub typhus than in non-severe patients ( p = 0 . 02 and p = 0 . 01 , respectively ) , with 372 U/ml ( 187–952 ) versus 136 U/ml ( 68–414 ) and 1471 U/ml ( 893–1876 ) versus 713 U/ml ( 367–1515 ) , but FSAP complexes were not ( Fig 3 ) . Increased disease severity was not reflected by higher levels of nucleosomes or ELA complexes in murine typhus patients , although the p-values of p = 0 . 06 and p = 0 . 08 , respectively , are suggestive of a trend ( Fig 3 ) . We performed ROC analysis to calculate the optimal cut off associated with the highest percentages of correct predictions into severe or non-severe groups . The cut off values corresponded to 1 , 040 U/ml for circulating nucleosomes and 275 U/ml for ELA complexes , respectively . For patients with scrub typhus there was a statistically significant increased risk of developing severe disease , if nucleosomes or ELA complex plasma levels were greater than these cut offs with an OR of 4 . 0 ( 95% CI 1 . 2–13 . 8; p = 0 . 028 ) and 4 . 8 ( 95%CI 1 . 3–17 . 6; p = 0 . 019 ) respectively . In patients with murine typhus , the cut off points corresponded to 129 U/ml and 14 U/ml for ELA complexes and for nucleosomes , respectively; no significant association was found for these markers and disease severity . A proportion of enrolled patients had coagulation and inflammation data available for subgroup analyses; for scrub typhus 44/65 ( 68% ) and for murine typhus 44/64 ( 67% ) ( Table 2 ) . Among patients with scrub typhus , elevated nucleosomes correlated with involvement of the CNS , lung and liver , as evidenced by correlations to GCS , respiratory rate and liver function tests . In murine typhus , nucleosomes correlated with pro-inflammatory cytokines , liver transaminases and days of illness , whereas ELA complexes correlated with lung and liver involvement and inflammatory markers ( Table 2 ) .
Although the importance of cell-mediated adaptive immune response in host protection against scrub typhus and murine typhus is well established , the role of neutrophils in these intracellular infections remains unclear . Pathophysiological studies of surrogate markers for cell adhesion molecules , coagulation and inflammation including histopathological observations in patients with scrub typhus and murine typhus , have highlighted differential profiles for the two diseases at the hospital admission time point [11–13 , 17] . In this study we hypothesized that neutrophils could provide a differential contribution to disease severity in these diseases , and measured the plasma levels of nucleosomes , FSAP activation and neutrophil activation in sympatric patients with scrub typhus and murine typhus in Laos . Although raised levels of circulating nucleosomes and systemic neutrophil activation were found in patients with both forms of typhus compared to controls , the increase of both markers was significantly higher in scrub typhus than in murine typhus ( Fig 1 ) . In scrub typhus , neutrophilia was recently described in the acute phase of infection [15] , but if neutrophils relate to a beneficial or detrimental role in the human pathogenesis of typhus has not been determined . In this study , neutrophil counts correlated with clinical signs of severe disease in patients with scrub typhus , and were significantly higher than in murine typhus ( Table 1 ) . That blood neutrophil counts did not correlate well with ELA complexes in both diseases could suggest that peripheral PMNs reflect the circulating pool , but not the marginal pool and PMNs located in tissue infiltrates , which in these vascular diseases could be substantially larger [36–38] . The immune histological investigation of skin biopsies of inoculation eschars ( Fig 2 ) , revealed the presence of prominent neutrophil-rich infiltrates within and adjacent to the necrotic center , as well as in perivascular infiltrates , suggesting that PMNs play a prominent role in the first-line defense against O . tsutsugamushi [7 , 13] . Although blood neutrophilia does not reflect the extent of systemic neutrophil activation , the data suggest that neutrophils could contribute more to disease severity in scrub typhus than previously known and that measuring neutrophil activation and nucleosomes may be useful for assessing disease severity in analogy to findings in sepsis patients . This study did not determine if neutrophils contribute to host defense against O . tsutsugamushi , but as disease severity and disease resolution can be independent of each other , the evidence tends towards a detrimental effect of neutrophil activation , associated with disease progression and complications . Neutrophil activation was raised in murine typhus , but significantly less prominent than in scrub typhus . In murine typhus patients , neutrophil activation did not associate with nucleosomes or disease severity , but an association to disease severity was seen ( Table 2 ) . Nucleosomes are histones and cell-free DNA released upon cell death . The high nucleosome levels in scrub typhus patients showed a strong association with disease severity , which was reflected in lower GCS scores , raised respiratory rates , jaundice and impaired liver function ( Table 2 ) . This association was not found in patients of the murine typhus group , where raised nucleosome levels associated with inflammatory cytokines , systemic inflammation and the duration of illness , while ELA complexes correlated strongly with inflammation , jaundice and increased respiratory rates . Even though these markers were not significantly higher in the severe murine typhus group ( p-values 0 . 06 and 0 . 08 respectively , Fig 3 ) , the findings merit further investigation of measuring nucleosomes and ELA complexes for assessing disease severity in analogy to findings in sepsis patients [21 , 27] . We found significantly higher bacterial loads for O . tsutsugamushi than R . typhi , which is in line with recent findings of higher bacteremic burden during scrub typhus than in murine typhus [39] . The increased circulating biomass of O . tsutsugamushi could contribute to a more pronounced systemic inflammation and exposure to neutrophils , resulting in marked systemic neutrophil activation , and a pro coagulant and inflammatory state , as reflected by higher nucleosome and neutrophil activation levels [12 , 40] . These data are congruent with recent reports that demonstrated neutrophil activation and nucleosome levels to correlate with disease severity and fatality in systemic inflammations , such as severe sepsis , septic shock and meningococcal sepsis [21 , 27] . FSAP circulates as a single-chain molecule in plasma and is activated upon contact with apoptotic and necrotic cells [22 , 23] . Recently , cell-free histones , RNA and glycosaminoglycans have been reported as FSAP activators as well [41–43] . Activated FSAP induces cleavage of DNA by DNAse resulting in nucleosome release from apoptotic cells and necrotic cells [22 , 44] . In the absence of any of these enzymes , DNA will not be released , therefore , the local concentration of DNAse , which acts in concert with FSAP to remove DNA from necrotic cells , critically influences nucleosome release as well [45] . In this study we found FSAP activation in both sympatric scrub and murine typhus patients , suggestive of involvement of FSAP in the release of nucleosomes in both diseases . Interestingly , strong correlations with nucleosomes were only observed in murine typhus , but not scrub typhus patients ( Table 2 ) . The activators of FSAP have not been identified for these forms of typhus , but our findings may indicate different mechanisms of FSAP activation , derived from potentially different sources of extracellular DNA for scrub and murine typhus . Although , the actual source of circulating nucleosomes remains difficult to establish , in inflammatory diseases nucleosomes are released from both dead hematopoietic and parenchymal cells . The different correlations between nucleosomes , FSAP and neutrophil activation in these intracellular infections suggest distinct cell death patterns occur , but the source of dead cells remains to be determined . If the cellular tropism of R . typhi and the associated cytotoxic T cell responses providing protective immunity in part by cytotoxicity towards infected cells contribute to nucleosome release remains to be determined [8 , 46] . However , in scrub typhus the data point towards neutrophils as a potential source of nucleosomes , contributing to disease exacerbation reflected by association to clinical disease severity ( Table 2 ) . The finding of high circulating plasma nucleosomes and ELA complexes as independent risk factors for developing severe complications in scrub typhus suggests that increased neutrophil activation could have a detrimental influence to disease severity , and no evidence points towards elevated nucleosomes and ELA complexes contributing towards resolving infection . In previous studies , nucleosomes and histones exposed on NETs have been reported to be highly cytotoxic to endothelial cells in vitro and are strongly pro-coagulant [24–27 , 40] . Further , concentrations of circulating nucleosomes have been associated with mortality and correlated with markers of coagulation , inflammation and neutrophil activation in sepsis [21 , 27] . Scrub typhus is a systematic vasculopathy with pronounced perivascular infiltrates and a procoagulant inflammatory coagulation profile [12] . Upon admission , scrub typhus patients with increased plasma levels of nucleosomes or ELA complexes above the specific cut off values ( 1 , 040 U/ml for circulating nucleosomes and 275 U/ml for ELA complexes ) were associated with a 4 . 8-fold and 4-fold increased risk of developing severe disease . Neutrophil activation correlated strongly with nucleosome levels , which suggests that neutrophils contribute to histone circulation , leading to subsequent vascular damage and resulting in a pro-coagulant profile , thus actually contributing to exacerbation of disease . That ELA complexes also correlated with fibrinolysis and high IL-8 plasma levels , a recently identified predictor of severe disease and mortality [17] , further corroborates the harmful effect of neutrophil activation . These results confirm that high levels of circulating nucleosomes and ELA complexes are independent risk factors for developing severe scrub typhus . Studies published by the Wagner group demonstrate that circulating nucleosomes and markers for neutrophil activation are reliable PMN markers associated with NET formation in the circulation [25 , 28] . By using comparable markers we find strong correlation of nucleosome levels with neutrophil activation in scrub typhus patients suggesting that neutrophils represent—at least in part—the source of nucleosomes in these patients , providing indirect evidence for NET formation . In summary , the data suggest that increased neutrophil activation relates to disease progression and severe complications in scrub typhus , but not murine typhus , and that increased plasma levels of nucleosomes and ELA complexes represent independent risk factors for developing severe scrub typhus .
|
Tropical rickettsial illnesses , especially scrub typhus and murine typhus , are increasingly recognized as a leading cause of treatable undifferentiated febrile illness in Asia , but remain severely neglected and under appreciated diseases in many areas . In this study we investigated the relationship of markers of neutrophil activation and cell death with disease severity in patients with acute scrub typhus and murine typhus in Laos . These easily measurable circulating markers were associated with a 4 to 5-fold increased risk of developing severe clinical disease manifestations in scrub typhus and represent independent predictors of severe disease , and possibly death . We also found strong correlations between circulating markers of cell death and neutrophil activation in patients with scrub typhus , but not murine typhus , providing indirect evidence that neutrophil extracellular traps could contribute to the vascular damage and pro-coagulant state leading to exacerbation of disease in scrub typhus , thus indicating a detrimental role of neutrophil activation . The data suggest that increased neutrophil activation relates to disease progression and severe complications , and increased plasma levels of nucleosomes and ELA complexes represent independent risk factors for developing severe scrub typhus .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Increased Nucleosomes and Neutrophil Activation Link to Disease Progression in Patients with Scrub Typhus but Not Murine Typhus in Laos
|
KSHV is etiologically associated with Kaposi's sarcoma ( KS ) , an angioproliferative endothelial cell malignancy . Macropinocytosis is the predominant mode of in vitro entry of KSHV into its natural target cells , human dermal microvascular endothelial ( HMVEC-d ) cells . Although macropinocytosis is known to be a major route of entry for many viruses , the molecule ( s ) involved in the recruitment and integration of signaling early during macropinosome formation is less well studied . Here we demonstrate that tyrosine phosphorylation of the adaptor protein c-Cbl is required for KSHV induced membrane blebbing and macropinocytosis . KSHV induced the tyrosine phosphorylation of c-Cbl as early as 1 min post-infection and was recruited to the sites of bleb formation . Infection also led to an increase in the interaction of c-Cbl with PI3-K p85 in a time dependent manner . c-Cbl shRNA decreased the formation of KSHV induced membrane blebs and macropinocytosis as well as virus entry . Immunoprecipitation of c-Cbl followed by mass spectrometry identified the interaction of c-Cbl with a novel molecular partner , non-muscle myosin heavy chain IIA ( myosin IIA ) , in bleb associated macropinocytosis . Phosphorylated c-Cbl colocalized with phospho-myosin light chain II in the interior of blebs of infected cells and this interaction was abolished by c-Cbl shRNA . Studies with the myosin II inhibitor blebbistatin demonstrated that myosin IIA is a biologically significant component of the c-Cbl signaling pathway and c-Cbl plays a new role in the recruitment of myosin IIA to the blebs during KSHV infection . Myosin II associates with actin in KSHV induced blebs and the absence of actin and myosin ubiquitination in c-Cbl ShRNA cells suggested that c-Cbl is also responsible for the ubiquitination of these proteins in the infected cells . This is the first study demonstrating the role of c-Cbl in viral entry as well as macropinocytosis , and provides the evidence that a signaling complex containing c-Cbl and myosin IIA plays a crucial role in blebbing and macropinocytosis during viral infection and suggests that targeting c-Cbl could lead to a block in KSHV infection .
KSHV is etiologically associated with Kaposi's sarcoma ( KS ) , the most common AIDS related malignancy , as well as with two lymphoproliferative diseases , primary effusion lymphoma ( PEL ) and multicentric Castleman's disease [1] , [2] . KSHV infects a variety of target cells both in vivo and in vitro . Entry into the target cells is the most crucial step in the establishment of a successful infection for all viruses . KSHV utilizes different modes of endocytosis to enter different target cells in vitro [3] . For example , KSHV enters human foreskin fibroblasts ( HFF ) via clathrin mediated endocytosis and enters HMVEC-d cells via macropinocytosis [3] , [4] , [5] . During the early stages of infection of HMVEC-d cells , KSHV forms a multi-molecular complex with host cell heparan sulfate , integrins ( α3β1 , αVβ3 and αVβ5 ) and transporter protein xCT with the subsequent induction of overlapping signal cascades [3] . Our studies show that KSHV induces a complex set of signaling molecules that are involved in diverse biological functions to regulate the various aspects of KSHV endocytosis including internalization , trafficking in the cytoplasm and nuclear delivery [3] . KSHV activates FAK , Src , PI3-K , Rho-GTPases and cytoskeleton rearrangement which are all critical for entry of virus [6] , [7] , [8] , [9] . KSHV also activates other downstream molecules such as PKC-ζ , MEK , ERK1/2 and NFkB which are essential for viral gene expression [6] , [7] , [8] , [9] . The Cbl family of adaptor proteins include three mammalian isoforms , c-Cbl , Cbl-b and Cbl-c or Cbl-3 [10] , [11] . Cbl proteins play important roles in signal transduction as negative regulators by mediating the ubiquitinilation and down-regulation of proteins while it acts as a positive regulator through their scaffold function in assembling signaling complexes [10] , [11] . c-Cbl has been shown to bind to several molecules critical in signal transduction [10] , [11] . Tyrosine phosphorylation of c-Cbl has been shown to be crucial for c-Cbl mediated adaptor functions in most circumstances [11] , [12] , [13] . However , the adaptor functions of c-Cbl and a c-Cbl mediated signaling pathway during virus infection has not been demonstrated . Macropinocytosis provides a major route for the productive infection of many viruses including KSHV . Macropinocytosis is an actin dependent membrane associated process which involves recruitment and integration of several signaling molecules necessary for cytoskeletal rearrangement and membrane remodeling . However , there is little information about the molecules involved in the recruitment and integration of signaling during macropinosome formation . Even though c-Cbl has been shown to recruit and link different signaling molecules in a signaling pathway , a direct role for c-Cbl in the process of macropinocytosis has not been established yet . Here we identified that c-Cbl is involved in KSHV entry and critical for triggering the macropinocytic event . Our data provide evidence that the interaction between c-Cbl and myosin IIA , a motor protein that binds to the proline rich domain of c-Cbl , regulates macropinocytosis of KSHV . This study on the functional organization of the c-Cbl and myosin IIA complex and its effect on viral entry provide an important insight into understanding the role of c-Cbl in virus infection .
HMVEC-d cells ( CC-2543; Clonetics , Walkersville , Md ) were grown in endothelial cell medium ( EBM2; Cambrex , Walkersville , MD ) . Induction of the KSHV lytic cycle in BCBL-1 cells , supernatant collection , and virus purification procedures were described previously [14] . KSHV DNA was extracted from the virus , and the copies were quantitated by real-time DNA PCR using primers amplifying the KSHV ORF 73 gene as described previously [15] . A pool of lentivirus shRNA specific for human c-Cbl and non-specific control shRNA were purchased from Santa Cruz Biotechnology ( Santa Cruz , CA ) . HMVEC-d cells were transduced with control lentivirus shRNA and c-Cbl lentivirus shRNA according to the manufacturer's instructions and selected by puromycin hydrochloride ( 10 µg ml−1; Santa Cruz Biotechnology ) . The following antibodies were used: mouse anti-c-Cbl , mouse anti-phospho Cbl 700 ( phosphorylated at Tyr700 ) , and mouse anti-p85 ( PI-3K ) antibodies ( BD Transduction Laboratories , San Diego , CA ) ; anti-phospho MLC II , anti-phospho Cbl 731 , anti-phospho Cbl 774 ( phosphorylated at Tyr731 and Tyr774 ) , isoform specific anti-myosin II heavy chain antibodies myosin IIA , IIB and IIC ( Cell Signaling Technology , Danvers , MA ) ; mouse anti-phospho tyrosine ( 4G10 clone; Millipore , Temecula , CA ) ; mouse anti-tubulin , mouse anti-beta actin antibodies ( Sigma , St Louis , MO ) ; rabbit anti-lamin B ( Abcam , Cambridge , MA ) ; rabbit anti-HA ( Zymed , Invitrogen , Carlsbad , CA ) ; mouse anti-ubiquitin ( P4D1 ) , mouse ant-GFP , mouse anti-GST ( Santa Cruz , CA ) ; rabbit anti-gB and mouse anti-gpK8 . 1A antibodies were created in our laboratory[16] , [17]; anti-goat , anti-rabbit and anti-mouse antibodies linked to horseradish peroxidase ( KPL Inc . , Gaithersburg , Md . ) ; DAPI , rhodamine conjugated dextran , Alexa 594 or Alexa 488 conjugated phalloidin and anti-rabbit and anti-mouse secondary antibodies conjugated to Alexa 488 , Alexa 594 ( Invitrogen ) ; protein A and G–Sepharose CL-4B beads ( Amersham Pharmacia Biotech , Piscataway , NJ ) ; blebbistatin , U0126 ( Calbiochem , La Jolla , CA ) ; TPA , LY294002 ( Sigma ) . Unless stated otherwise , cells were infected with KSHV at 10 DNA copies ( multiplicity of infection [MOI] ) per cell at 37°C . Entry was measured by infecting the cells with KSHV for 30 min . The cells were washed with HBSS to remove the unbound virus , treated with 0 . 25% trypsin-EDTA for 5 min at 37°C to remove the bound but non-internalized virus , and washed . Cells were recovered by centrifugation and total DNA was isolated from infected or uninfected cells using a DNeasy kit ( QIAGEN , Valencia , CA ) as described previously [15] . To calculate percent of inhibition of KSHV entry , internalized KSHV DNA was quantitated by amplification of the ORF73 gene by real-time DNA PCR [15] . The KSHV ORF73 gene cloned in the pGEM-T vector ( Promega ) was used for the external standard . The cycle threshold ( Ct ) values were used to generate the standard curve and to calculate the relative copy numbers of viral DNA in the samples . Percentage inhibition was calculated by considering the ORF73 copy numbers in untransduced cells as 100% . Total RNA was prepared from infected or uninfected cells using an RNeasy kit ( QIAGEN ) as described previously [15] . To quantitate viral gene expression , isolated RNA was subjected to ORF73 and ORF50 RNA expression by real-time reverse transcription ( RT ) -PCR using gene specific real-time primers and specific TaqMan probes [15] . The relative copy numbers of the transcripts were calculated from the standard curve plotted using the Ct values for different dilutions of in vitro-transcribed transcripts . These values were normalized to each other using the values of the GAPDH control reactions . Percentage inhibition was calculated by considering ORF73 and ORF50 gene expression in untransduced cells as 100% . Cells were lysed in RIPA buffer ( 15 mM NaCl , 1 mM MgCl2 , 1 mM MnCl2 , 2 mM CaCl2 , 2 mM phenylmethylsulfonyl fluoride , and protease inhibitor mixture ( Sigma ) ) and centrifuged at 12 , 000 rpm at 4°C for 15 min . Lysates were normalized to equal amounts of protein and the proteins were separated by 7 . 5–12 . 5% gradient SDS-PAGE , transferred to nitrocellulose and probed with the indicated primary antibodies . Detection was by incubation with species-specific HRP-conjugated secondary antibodies . Immunoreactive bands were visualized by enhanced chemiluminescence ( Pierce , Rockford , IL ) according to the manufacturer's instructions . The bands were scanned and quantitated using the FluorChem FC2 and Alpha-Imager Systems ( Alpha Innotech Corporation , San Leonardo , CA ) . Two hundred micrograms of cell lysates prepared as described in the above section were incubated for 2 h with immunoprecipitating antibody at 4°C , and the immune complexes were captured by protein A or G-Sepharose . The samples were tested by Western blot with specific primary and secondary antibodies . HMVEC-d cells were infected with KSHV for different time points . The samples were resolved on an SDS-PAGE gel and the gel was stained with Coomassie blue . The bands of interest were excised , digested with trypsin , separated by reverse phase nano-chromatography and analyzed by mass spectrometry . Immunofluorescence assay was performed using HMVEC-d cells seeded on 8 well chamber slides ( Nalge Nunc International ) . Infected and uninfected cells were fixed with 3% paraformaldehyde for 15 min , permeabilized with 0 . 2% Triton X-100 , and blocked with Image-iTFX signal enhancer ( Invitrogen ) . The cells were then immunostained with primary antibodies against the specific proteins , followed by fluorescent dye-conjugated secondary antibodies . For colocalization with dextran and transferrin , cells were incubated with the fluid-phase marker dextran Texas Red ( 40 kD , 0 . 5 mg ml−1; Invitrogen ) or Alexa 594 transferrin ( 35 µg ml−1; Invitrogen ) at 37°C in the presence or absence of KSHV followed by immunostaining with the appropriate antibodies . Cells were imaged with a Nikon fluorescence microscope equipped with a Metamorph digital imaging system . DIC ( Differential Interference Contrast ) images were acquired with objectives equipped with DIC optics . For confocal analysis , the Olympus Fluoview 300 fluorescence confocal microscope was used for imaging , and analysis was performed using Fluoview software ( Olympus , Melville , NY ) . All experiments were performed at least three times . HMVEC-d cells , incubated with dextran Texas Red ( 0 . 5 mg ml−1 , 40 kD; Invitrogen ) and KSHV for 30 min , were washed twice in HBSS . To remove surface bound dextran , cells were treated with 0 . 25% trypsin-EDTA and the cells were harvested . Quantitative analysis of dextran uptake was determined by counting the number of cells stained positive for dextran under immunofluorescence microscope . At least 10 different microscopic fields of 50 cells each were counted for each experiment and the results displayed as percentage of dextran positive cells . Flow cytometry analysis was used to quantify the uptake of dextran during KSHV internalization in control shRNA and c-Cbl shRNA transduced cells . Cells were incubated with 500 µg ml−1 FITC-dextran in the presence or absence of virus at 37°C for 30 min . The cells were washed , harvested using trypsin EDTA , fixed and analyzed by flow cytometry . Mean fluorescence intensity was determined using a Becton Dickinson FACS system and CellQuest software . Cells incubated with dextran alone were used as controls . HeLa cells ( ATCC CCL-2 ) were cultured in DMEM containing 10% fetal bovine serum . Wild type , mutants and deletion constructs of c-Cbl , c-Cbl C-terminal domain encompassing PRD ( Cbl-C ) and c-Cbl N-terminal domain ( Cbl-N ) constructs were generously provided by Dr . Hamid Band [18] ( Eppley Institute for Cancer and Allied Diseases , University of Nebraska Medical Center ) . Cells were transiently transfected with wild type , mutants and deletion constructs . Transfection was performed using 5 µg of plasmid DNA , lipofectamine 2000 ( Invitrogen ) , and Opti-MEM medium ( Invitrogen ) according to the manufacturer's instructions . After transfection , cells were cultured for 48 h . Cells were then serum starved for 4 h and stimulated with TPA ( 100 ng ml−1 ) at 37°C for 5 min . Lysis was performed in RIPA buffer plus protease inhibitors . The cell lysate was used for immunoprecipitation and immunoblotting . E . coli BL21 ( DE3 ) cells were transformed with pGEX4T . 1 GST-Cbl C ( Cbl residues 358–906 ) plasmids which encode regions encompassing the C-terminal PRD domain of c-Cbl and pGEX4T . 1 GST-Cbl N which encodes the N-terminal region of c-Cbl . Expression of the GST-Cbl fusion proteins was induced with IPTG ( isopropyl-D-1-thiogalactopyranoside ) 1 mM for 3 h at 37°C . The bacterial lysates ( 500 µg ) were incubated with glutathione-sepharose beads ( GE Healthcare , U . K . ) for 2 h at 4°C . The beads were washed with lysis buffer three times . 293T cells were transiently transfected with 2 µg pEGFP C3 myosin IIA plasmids ( Addgene ) . After 48 h of transfection , cells were lysed in RIPA buffer and 500 µg of the lysates were incubated with the glutathione-Sepharose beads bound with the GST-Cbl fusion proteins . The beads and the bound proteins were collected by centrifugation , washed and the interaction of GST-Cbl with myosin IIA was analyzed by SDS-PAGE and Western blotting using anti-GFP antibody . HMVEC-d cells infected with KSHV for 5 min were fixed and stained with DAPI . DIC images were acquired and the cells presenting blebs or no blebs were counted visually . At least 10 random microscopic fields per experiment were counted and expressed as a proportion of the total number of DAPI stained cells . Infected and uninfected cells were washed three times with HBSS and lysed in homogenization buffer ( 250 mM sucrose , 20 mM HEPES , 10 mM KCL , 1 mM EDTA , 1 mM EGTA and protease inhibitors ) . The homogenate was subjected to centrifugation at 3 , 000 rpm for 5 min . Post-nuclear supernatant was centrifuged at 8 , 000 rpm for 5 min at 4°C . The supernatant was again centrifuged at 40 , 000 rpm for 1 h at 4°C , and the supernatant and the pellet were considered the cytosolic and the membrane fractions , respectively . The membrane pellet was solubilized using RIPA buffer and used for Western blot .
To determine whether c-Cbl and the c-Cbl mediated signaling pathway play roles in KSHV infection , we first examined the early tyrosine phosphorylation kinetics of c-Cbl in KSHV infected cells . HMVEC-d cells infected with KSHV induced rapid tyrosine phosphorylation of c-Cbl , which was detectable as early as 1 min post-infection ( p . i . ) , reaching maximum levels at 5 min ( 4 . 1-fold ) , followed by a decrease which was constent for as much as 30 min p . i . ( Figure 1a ) . To determine whether the phosphorylation of c-Cbl is specifically induced by KSHV , cells were infected with KSHV pre-incubated with heparin . Heparin is known to block the binding of KSHV to the target cells [19] . Compared to the untreated virus , heparin treated virus considerably reduced the phosphorylation of c-Cbl ( Figure 1a ) which demonstrated the specificity of KSHV induced c-Cbl phosphorylation . The efficient tyrosine phosphorylation of c-Cbl is suggestive of the possible involvement of a c-Cbl mediated signaling pathway in KSHV infection . We next investigated the link between c-Cbl phosphorylation and other signaling molecules activated during KSHV infection . It is well documented that the interaction of KSHV glycoproteins with integrins and other cellular receptors activate FAK and the downstream molecules Src and PI3-K [7] , [9] , [14] , [20] , [21] . c-Cbl has been shown to form a complex with PI3-K p85 in the integrin mediated signaling pathway [12] . We therefore examined whether the association of PI3-K with c-Cbl occurred during KSHV infection . KSHV infection led to an increase in the interaction of c-Cbl with PI3-K p85 in a time dependent manner ( Figure 1b ) . To verify that the c-Cbl-PI3-K interaction is specifically induced by virus , cells were infected with heparin treated virus which notably decreased the association of c-Cbl with PI3-K ( Figure 1b ) . The c-Cbl-PI3-K association was further confirmed by confocal analysis ( Figure 1c ) . Consistent with previous studies [13] , our results demonstrated that activated c-Cbl leads to the association of c-Cbl with PI3-K . Previous studies have shown that ERK1/2 is activated during KSHV infection and is a key signaling molecule implicated in viral gene expression [22] . To examine whether an ERK1/2 associated pathway is involved in c-Cbl mediated signaling , we investigated the association of ERK1/2 with c-Cbl in KSHV infected cells . No colocalization was observed between ERK1/2 and c-Cbl ( Figure 1c ) which suggested that the ERK associated pathway is not involved in c-Cbl mediated signaling in KSHV infected cells . Taken together , our data suggests that a signaling complex which contains c-Cbl and PI3-K but not ERK1/2 is involved in the integrin mediated signaling pathway of KSHV infection . To further demonstrate the relationship between the interaction of c-Cb1 with PI3-K but not with ERK1/2 , we studied the effect of PI3-K and ERK1/2 inhibitors in KSHV induced c-Cbl phosphorylation . HMVEC-d cells pretreated with the PI3-K inhibitor , LY294002 , and the ERK1/2 inhibitor , U0126 were infected with KSHV for 10 min and the lysates were analyzed for c-Cbl phosphorylation . KSHV induced c-Cbl phosphorylation was abolished by the PI3-K inhibitor LY294002 , whereas the ERK1/2 inhibitor U0126 did not show any inhibition on c-Cbl phosphorylation ( Figure 1d ) . The failure of ERK1/2 inhibitor to abolish c-Cbl phosphorylation confirmed that ERK1/2 is not associated with c-Cbl induction in KSHV infected cells . We next used c-Cbl lentivirus encoding shRNAs to knockdown c-Cbl activity in HMVEC-d cells ( c-Cbl shRNA cells ) to analyze the functions of c-Cbl in KSHV infected cells . The c-Cbl specific shRNA inhibited 90% of c-Cbl expression as detected by Western blotting with antibodies to c-Cbl ( Figure S1a ) . Untransduced , control shRNA and c-Cbl shRNA transduced cells were infected with KSHV for 2 and 24 h , and viral gene expression was determined by real-time RT-PCR analysis . Compared with control cells , c-Cbl shRNA transduced cells showed about 60–70% inhibition of the latency associated ORF 73 gene ( Figure 2a ) and 70–80% inhibition of the lytic switch ORF 50 gene ( Figure 2b ) expression . We next determined whether the inhibition of viral-gene expression by c-Cbl shRNA was due to a blockage at the entry stage of the virus . To determine c-Cbl's role in KSHV entry , internalization of viral DNA was determined by measuring viral ORF 73 DNA copy numbers by real-time DNA PCR . We observed ∼65% inhibition of KSHV entry in c-Cbl shRNA cells compared to control cells ( Figure 2c ) . Internalized KSHV ORF 73 DNA copy numbers and ORF 50 and ORF 73 RNA copy numbers are shown as histograms in supplementary Figure S1 . Taken together , these studies demonstrated that the decreased viral gene expression observed in c-Cbl shRNA cells was due to a decrease in the entry of KSHV . These results further suggested that a c-Cbl containing signaling complex may be crucial for the initiation of entry and for a productive infection . In our earlier studies we have demonstrated that macropinocytosis is the major pathway of KSHV entry leading to a productive infection in HMVEC-d cells [4] . Since c-Cbl inhibited KSHV's entry , we theorized that c-Cbl might be playing a role in macropinocytosis and associated signaling events . Viruses such as vaccinia virus that use macropinocytosis as a mode of entry induce signaling molecules and cytoskeletal rearrangements in the form of blebs which ultimately retract and ingest viral particles [23] , [24] . To determine whether blebs were involved in KSHV infection , we used DIC image analysis and observed the association of KSHV with blebs . As shown in Figure 3a , bleb formation and the association of individual blebs with KSHV was observed by 5 min p . i . . The DIC microscopic analysis of a single bleb for viral particles confirmed that the blebs , formed during viral infection , were associated with viral particles ( Figure 3b ) . To further investigate whether KSHV infection induces blebbing , HMVEC-d cells were infected with KSHV for 2 and 5 min and then actin , a well known determinant of cell shape and blebbing , was stained with phalloidin [25] . Within 2 min of infection , membrane protrusions appeared along the cell surface which rapidly enlarged into well-formed blebs at 5 min ( Figure S2 ) . We also observed the association of viral particles at the bleb forming site as well as with retracting blebs ( Figure S2 ) . Unlike the well formed blebs , the retracting blebs were characterized by a thick actin cortex [26] . These results demonstrated that early during infection , KSHV induces actin reorganization and the subsequent formation of blebs that may be involved in its entry . Since we observed that c-Cbl shRNA inhibited KSHV entry , which involves bleb formation , we hypothesized that c-Cbl and its phosphorylation might be involved in the dynamics of virus induced blebbing . To understand the function of c-Cbl in blebbing , we examined the localization of phosphorylated c-Cbl ( p-Cbl ) in KSHV infected cells . Confocal microscopy analysis showed that KSHV induced p-Cbl localized to blebs as early as 2 min p . i . ( data not shown ) . Bleb formation , and its association with p-Cbl , was maximal at 5 min , and by 10 min blebs containing p-Cbl started internalizing and p-Cbl was mostly observed at the nuclear periphery by 15 min p . i . ( Figure 3c ) . A similar pattern of localization was exhibited by p-Cbl and virus at the blebs as early as 5 min p . i . , and accumulation of p-Cbl and virus around the nuclear periphery was observed at 15 min p . i . ( Figure 3d ) . These results suggested that the recruitment of phosphorylated c-Cbl to the sites of bleb formation was involved in bleb associated entry of the virus . To further explore the role of c-Cbl in bleb associated macropinocytosis , we performed a confocal immunofluorescence colocalization study between p-Cbl and the macropinocytosis marker dextran in infected cells . This analysis showed that dextran colocalized with p-Cbl at 10 min p . i . ( Figure 4a ) in the infected cells . Next , we performed a dextran uptake study since the uptake of dextran has been used as a biochemical marker of macropinocytosis . We incubated the cells with dextran in the presence or absence of virus for 30 min and then quantitated the level of uptake . As shown in Figure 4b and supplementary information Figure S3a , c-Cbl shRNA cells showed a drastic inhibition of dextran uptake compared to control shRNA cells infected with KSHV . This indicated that the uptake of dextran or macropinocytosis in KSHV infected cells was a c-Cbl dependent process . The uptake of dextran and colocalization with KSHV in non-specific control shRNA and c-Cbl shRNA cells were confirmed by immunofluorescence colocalization and DIC analysis . In control shRNA cells infected with KSHV , intracellular KSHV was highly colocalized with dextran , whereas in c-Cbl shRNA cells infected with KSHV , most of the viral particles remained at the membrane periphery although minimal colocalization of KSHV with dextran was observed in some cells ( Figure 4c and Figure S3b ) . Control shRNA cells incubated with KSHV and Alexa 594 transferrin , a marker for clathrin-mediated endocytosis , did not show any significant colocalization with KSHV ( Figure 4c and Figure S3b ) which demonstrated the specificity of macropinocytosis mediated entry in HMVEC-d cells [4] . These results were consistent with the results of the dextran uptake study , confirming that c-Cbl was critical for inducing the macropinocytic process that promoted the internalization of KSHV . The uptake of dextran in control shRNA and c-Cbl shRNA cells was further quantified by FACS analysis . Cells were incubated with dextran in the presence or absence of virus for 30 min and the uptake was measured using flow cytometry . As shown in Figure 4d , compared to the control shRNA cells , c-Cbl shRNA cells showed a notable decrease in mean fluorescence intensity . These results are consistent with the results of the immunofluorescence analysis and thus confirmed the role of c-Cbl in KSHV induced macropinocytosis . c-Cbl is a multi-domain protein that interacts with a number of signaling molecules and performs multiple functions [10] , [11] . To decipher the molecular partners interacting with c-Cbl during KSHV infection , we used mass spectrometric analysis . HMVEC-d cells were infected with KSHV for 1 , 5 and 10 min , lysed and the lysates were immunoprecipitated with anti-c-Cbl antibodies . Samples were separated by SDS-PAGE , followed by Coomassie blue staining and mass spectrometry analysis . Mass spectrometry identified several novel c-Cbl interacting proteins in the infected samples ( Supplementary Table S1 ) . The most prominent protein identified in the infected samples was myosin IIA which is one of three isoforms of the non-muscle myosin II family of proteins [27] , [28] . The other novel interacting partners of c-Cbl in the infected cells included vimentin , HSP70 , BiP protein , Rho GEF and a solute carrier anion exchanger ( Table S1 ) . To confirm the mass spectrometry data , uninfected and KSHV infected cell lysates were immunoprecipitated with anti-c-Cbl antibody and blotted for the three isoforms of the non-muscle myosin II family , IIA , IIB and IIC , with isoform specific antibodies . Our results confirmed that c-Cbl interacts with myosin IIA in the lysates of infected cells , whereas the other isoforms did not show any interaction with c-Cbl ( Figure 5a ) . To elucidate the functional domain of c-Cbl involved in myosin IIA interaction , a series of truncated and mutant constructs of c-Cbl with HA epitope tags were used ( Figure S4 ) . Since HMVEC-d cells are not easily transfectable , we used HeLa cells for this study . HeLa cells were transfected with vector alone , Cbl wild-type , Cbl-tyrosine kinase binding domain ( TKB ) mutant , RING domain mutant , and two truncation mutants ( Cbl-Δ357 and Cbl-Δ421 ) . As TPA ( phorbol ester ) has been shown to induce membrane blebbing [29] , we used TPA induced HeLa cells to analyze the interaction of over-expressed c-Cbl with endogenous myosin IIA . Transfection of the Cbl-TKB mutant and RING mutant induced the interaction of c-Cbl with myosin IIA similar to full length wild-type Cbl . The truncated versions Cbl-Δ357 and Cbl-Δ421 lacking a C-terminal proline rich domain ( PRD ) decreased the interaction with myosin IIA considerably ( Figure 5b ) . Expression of all constructs determined by Western blotting with HA revealed comparable levels of protein ( Figure 5b ) . Taken together , these results indicated that the C-terminal region encompassing the PRD of c-Cbl was sufficient for association with myosin IIA . To further confirm that the C-terminal PRD of c-Cbl interacts with myosin , an in vitro binding assay was performed using bacterially expressed GST fusion proteins of c-Cbl C-terminal ( Cbl-C , encompassing PRD ) and N-terminal ( Cbl-N ) domains . GST Cbl-C and Cbl-N proteins adsorbed on glutathione sepharose beads were incubated with 293T cell lysates expressing GFP-tagged myosin IIA . The interaction between GFP-myosin IIA and GST-Cbl was analyzed by Western blotting with anti-GFP antibody . Our results demonstrated that myosin IIA predominantly interacted with Cbl-C domains compared to Cbl-N ( Figure 5c ) . The interaction of myosin IIA with c-Cbl suggested that their association could be playing a role in blebbing and macropinocytosis of KSHV . To investigate this , we used blebbistatin , a specific inhibitor of myosin II ATPase activity that has been shown to inhibit myosin II induced blebbing [30] , [31] and macropinocytosis [23] . As shown in Figure 6a , we observed a dose dependent inhibition of KSHV internalization in 25 µM ( ∼35% ) and 50 µM ( ∼60% ) concentrations of blebbistatin indicating that the entry process was dependent on myosin II activity . As reported previously [4] , chlorpromazine , an inhibitor of clathrin dependent endocytosis , did not show any notable decrease in entry of KSHV ( Figure 6a ) . These findings suggested that c-Cbl associated myosin IIA was involved in bleb mediated macropinocytosis of KSHV . To determine whether blebbistatin treatment affects other internalization pathways , we investigated the effect of blebbistatin on clathrin-mediated internalization , a major and well characterized endocytic pathway of eukaryotic cells . To study this , untreated or blebbistatin treated HMVEC-d cells were induced with FBS in the presence of Alexa 594 labelled transferrin or Texas Red labelled dextran . The endocytic uptake of transferrin and dextran were then analyzed using immunofluorescence . As indicated in Figure 6b and d , blebbistatin strongly inhibited the uptake of dextran , whereas the uptake of transferrin was unaffected ( Figure 6c and e ) . This demonstrated that blebbistatin specifically inhibits macropinocytosis but not clathrin mediated endocytosis pathways . The above studies demonstrated that the c-Cbl interacting partner myosin IIA is a biologically significant component of the c-Cbl signaling pathway . We then explored the role of c-Cbl in myosin II induced blebbing in KSHV infected cells . If c-Cbl is an upstream molecule of myosin IIA , the loss of function of c-Cbl should prevent the formation of myosin II mediated blebs in c-Cbl shRNA cells . To test this hypothesis , control shRNA and c-Cbl shRNA transduced HMVEC-d cells were infected with KSHV and the percentage of cells with blebs was quantitated . As expected , in c-Cbl shRNA transduced cells , the blebs were considerably reduced compared to control shRNA cells ( Figure 7a and b ) . This suggested that c-Cbl and associated myosin IIA molecules were linked to induce membrane blebbing in HMVEC-d cells . Our results indicated that c-Cbl plays an upstream role in the regulation of bleb formation which occurs as a result of myosin II induced cortical contractility [25] , [26] . To further demonstrate that c-Cbl is upstream to myosin IIA , we infected blebbistatin treated cells with KSHV and the membrane localization of c-Cbl was observed by immunofluorescence . As shown in Figure 7c , blebbistatin did not inhibit the localization of c-Cbl to the plasma membrane , whereas it prevented the formation of blebs in the infected cells . This suggested that myosin IIA was downstream to c-Cbl and was not involved in the localization of c-Cbl to the plasma membrane . A subclass of myosins , the class II myosins are hexameric motor proteins composed of two identical heavy chains ( MHC ) , and two pairs of light chains ( MLC ) . It has been well accepted that phosphorylation of the myosin light chain II is a major determinant of force generation and actomyosin dynamics during apoptotic membrane blebbing [32] , [33] . Hence , we examined phosphorylation of myosin light chain II ( p-MLC II ) during KSHV infection . Compared to the uninfected cells , KSHV infection results in rapid and strong phosphorylation of MLC II with maximal phosphorylation at 10 min p . i . ( 5 . 8-fold increase ) and decreased thereafter ( Figure 7d ) . The specificity of virus induced MLC II phosphorylation was shown using heparin treated virus which did not induce MLC II phosphorylation ( Figure 7d ) . Since light chain phosphorylation has been shown to regulate blebbing [32] , [33] , our results suggested that KSHV induced MLC II may be participating in the induction of blebbing during infection . During virus induced and apoptotic membrane blebbing , the signaling molecules associated with cytoskeletal function are recruited to the blebs [23] , [26] . It has been demonstrated that the recruitment of functional myosin II heavy and light chain complexes drive the process of bleb retraction [26]; however , it is not clear how individual myosin II molecules are recruited to the blebs . It is possible that c-Cbl interaction with myosin IIA leads to recruitment of the complex to the blebs . Therefore , we infected control shRNA and c-Cbl shRNA cells with KSHV and tested the association of c-Cbl with myosin II in the blebs . Punctate staining of p-Cbl and p-MLC II was observed in the interior of blebs with a predominant colocalization between p-Cbl and p-MLC II in control shRNA infected cells ( Figure 8a ) suggesting that phosphorylated Cbl recruits individual myosin molecules to the blebs . In contrast , c-Cbl shRNA cells infected with KSHV did not show bleb formation and the recruitment and localization of myosin II to the bleb membrane ( Figure 8a ) . To further confirm the membrane localization of myosin IIA and c-Cbl to the blebs , membrane fractions from control shRNA infected cells and c-Cbl shRNA infected cells were isolated and analyzed by Western blotting . Compared to the uninfected cells , control shRNA cells infected with KSHV showed a 3 . 2 and 2 . 9-fold increase in membrane localization of myosin IIA and c-Cbl , respectively , whereas in c-Cbl shRNA-KSHV cells , membrane localization of myosin IIA and c-Cbl was almost completely absent ( Figure 8b ) . This suggested that a decrease in membrane localization of myosin IIA in c-Cbl shRNA cells may be caused by a deficiency in the association of c-Cbl with myosin IIA . Myosin II molecules recruited to the membrane blebs form contractile foci in association with actin under the bleb membrane which is critical for bleb retraction [26] . Therefore , we asked whether a similar kind of association occurs between p-MLC II and actin in control shRNA-KSHV and c-Cbl shRNA-KSHV cells . As has been previously reported [26] , we observed the association of actin and p-MLC II in the membrane blebs in control shRNA KSHV infected cells ( Figure 9a ) . Bleb formation and the association of actin with p-MLC II in the blebs were not seen in c-Cbl shRNA-KSHV cells ( Figure 9a ) . The association of p-MLC II with actin , which is a known interacting partner of c-Cbl [10] , coupled with the detection of an association with c-Cbl in the infected cells ( Table S1 ) suggested that actin , myosin IIA and c-Cbl are part of a signaling complex which might be essential for the formation of blebs and bleb retraction . To examine whether the interaction of c-Cbl with myosin IIA is actin dependent , we investigated the interaction of c-Cbl with myosin IIA in cells treated with actin inhibitor cytochalasin D . The interaction between c-Cbl and myosin IIA was then examined by coimmunoprecipitation and Western blot analysis . As shown in Figure 9b , cytochalasin D did not inhibit the interaction of myosin IIA with c-Cbl . The interaction of myosin IIA with c-Cbl in cytochalasin D treated cells suggested that actin is not essential for the initial association of myosin IIA with c-Cbl in KSHV infected cellular environment . c-Cbl is an E3 ubiquitin ligase , which has been shown to be involved in poly or monoubiquitination of a number of proteins [34] , [35] , [36] . To further analyze the functional significance of the interaction between c-Cbl , myosin and actin , the role of c-Cbl in ubiquitination of actin and myosin was analyzed . Both myosin and actin ubiquitination were determined in control shRNA and c-Cbl shRNA cells infected with KSHV . We observed multiple bands of ubiquitinated myosin and actin probably indicating monoubiquitination on multiple sites and not polyubiquitination in the infected cells ( Figure 9c and d ) . c-Cbl shRNA abolished the KSHV induced ubiquitination of actin and myosin suggesting that it was mediated by c-Cbl . This indicated that upon KSHV infection , c-Cbl binds to actin and myosin which in turn is responsible for ubiquitination .
Several viruses utilize macropinocytosis to gain access to target cells [24] . Macropinocytosis is strictly an actin driven process which includes the formation of membrane ruffles , lamellipodia and blebs [24] . Blebbing is a phenomenon which is mainly observed during cellular processes such as embryogenesis , cytokinesis and apoptotic cell death [25] . Bleb associated macropinocytosis provides a mechanism of virus entry into host target cells and has been shown to be an efficient tactic of the virus to enter the target cells by mimicking the apoptotic cells [23] . Macropinocytosis has been observed as a major route of entry of KSHV into HMVEC-d cells , the natural target cells of infection [3] , [4] . During KSHV infection , the interaction of KSHV glycoproteins with integrins and other cellular receptors activate host cell signaling molecules FAK , Src , PI3-K , RhoA GTPase and are all recruited to the entry site [3] . The inhibition of any of these proteins significantly reduces the entry of virus suggesting that KSHV exploits preexisting host cell signaling machinery for a successful infection [3] . Our current study shows that c-Cbl is also required for efficient macropinocytic uptake , suggesting that the previously observed signaling molecules are linked via c-Cbl to perform their downstream functions . Since bleb associated macropinocytosis is an actomyosin dependent process , c-Cbl and its interaction with myosin playing a role in macropinocytosis further strengthens the possibility that c-Cbl is a critical molecule involved in linking the signaling molecules in the virus induced macropinocytic process . Based on the strong evidences presented here , we propose that efficient bleb mediated macropinocytic uptake enables a productive infection of KSHV to occur in cells supporting a signaling cascade that contains the c-Cbl-myosin IIA complex , and that a defect in c-Cbl-myosin IIA association results in lowered macropinocytic uptake , entry and infection . The simultaneous decrease in macropinocytic uptake and blebbing by blebbistatin treatment and c-Cbl silencing with shRNA strongly suggests that bleb associated macropinocytosis is the predominant pathway of KSHV infection in HMVEC-d cells . The defect in macropinocytosis in c-Cbl shRNA cells could be due to a defect in linking myosin II molecules to the membrane associated events which is necessary for blebbing and bleb mediated macropinocytosis [24] , [25] , [26] . Although the molecular details of bleb associated macropinosome formation are yet to be uncovered , our study demonstrates that during KSHV infection , the interaction of c-Cbl with myosin IIA leads to bleb formation and the recruitment of myosin IIA into the blebs , where the myosin IIA molecules could be interacting with actin to accelerate actomyosin contraction and bleb retraction [26] . Retracting blebs form macropinosomes along with the viral particles close to the blebs ( Figure 10 ) [24] . Myosins provide the ATP-dependent force to generate the movement required for the process of bleb retraction [26] . Myosin IIA is the major isoform of myosin II implicated in membrane associated functions such as the maintenance of cell shape and movement [37] , [38] . The functional ending of ubiquitination is related to the type of ubiquitin chains added to the substrate protein [39] . Monoubiquitination promotes internalization of cell surface receptors and subsequent lysosomal degradation [40] , whereas polyubiquitinated proteins are targeted for proteasomal degradation [41] . Whether the complex events associated with the monoubiquitination of actin and myosin by c-Cbl could be related to an increase in the activity of actomyosin contraction or directing subcellular compartmentalization during KSHV infection and macropinocytosis remains to be studied in detail . Further studies are required to understand the molecular aspects of the interaction between c-Cbl and myosin IIA and their role in subsequent stages of bleb mediated macropinocytosis . Further studies are also required to validate the specificity and the functional significance of other identified c-Cbl interacting proteins , their association with c-Cbl during infection and whether c-Cbl induces the ubiquitination of cell surface molecules recognized by KSHV . In conclusion , our results provide for the first time clear evidence demonstrating that c-Cbl , and the interaction between c-Cbl and myosin IIA , is critical for triggering bleb mediated macropinocytic events during KSHV entry into target cells ( Figure 10 ) . This study also provides the first evidence that c-Cbl and a c-Cbl mediated signaling pathway as well as ubiquitination play roles in viral infection and that c-Cbl function as an adaptor protein for PI3-K and other KSHV induced signaling events . This also identifies c-Cbl as a potential target to intervene in KSHV infection .
|
KSHV is etiologically associated with Kaposi's sarcoma ( KS ) , the most common AIDS related neoplasm . The first key step in KSHV infection is its initial contact with target cells and entry . While it is known that KSHV uses macropinocytosis for its infectious entry into its natural target cells , HMVEC-d cells , we know little about the molecule ( s ) involved in this event . Here , we show that the adaptor protein c-Cbl plays a major role in regulating bleb associated macropinocytosis of KSHV . The results demonstrate that c-Cbl protein functions as an adaptor for the myosin II hexameric complex in macropinocytic events . Knocking down c-Cbl by shRNA induces defects in myosin II dependent blebbing and KSHV entry , indicating that c-Cbl uses myosin II to coordinate signaling pathways , resulting in bleb formation and bleb retraction . This work provides a clear understanding of the role of c-Cbl in the recruitment and integration of signaling molecules around the macropinosome during virus infection , and identifies potential targets to intervene in KSHV infection .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases/viral",
"infections",
"virology/host",
"invasion",
"and",
"cell",
"entry",
"virology/viruses",
"and",
"cancer",
"virology"
] |
2010
|
Interaction of c-Cbl with Myosin IIA Regulates Bleb Associated Macropinocytosis of Kaposi's Sarcoma-Associated Herpesvirus
|
Single-cell RNA sequencing ( scRNA-seq ) technology allows researchers to profile the transcriptomes of thousands of cells simultaneously . Protocols that incorporate both designed and random barcodes have greatly increased the throughput of scRNA-seq , but give rise to a more complex data structure . There is a need for new tools that can handle the various barcoding strategies used by different protocols and exploit this information for quality assessment at the sample-level and provide effective visualization of these results in preparation for higher-level analyses . To this end , we developed scPipe , an R/Bioconductor package that integrates barcode demultiplexing , read alignment , UMI-aware gene-level quantification and quality control of raw sequencing data generated by multiple protocols that include CEL-seq , MARS-seq , Chromium 10X , Drop-seq and Smart-seq . scPipe produces a count matrix that is essential for downstream analysis along with an HTML report that summarises data quality . These results can be used as input for downstream analyses including normalization , visualization and statistical testing . scPipe performs this processing in a few simple R commands , promoting reproducible analysis of single-cell data that is compatible with the emerging suite of open-source scRNA-seq analysis tools available in R/Bioconductor and beyond . The scPipe R package is available for download from https://www . bioconductor . org/packages/scPipe .
Advances in single-cell transcriptomic profiling technologies allow researchers to measure gene activity in thousands of cells simultaneously , enabling exploration of gene expression variability [1] , identification of new cell types [2] and the study of transcriptional programs involved in cell differentiation [3] . The introduction of cellular barcodes , sequences distinct for each cell attached to the dT-primer , has increased the throughput and substantially reduced the cost of single-cell RNA sequencing ( scRNA-seq ) . These barcodes allow for the demultiplexing of reads after cells are pooled together for sequencing . Apart from cellular barcodes , molecular barcodes or unique molecular identifiers ( UMIs ) , are frequently employed to remove PCR duplicates and allow identification of unique mRNA molecules , thereby reducing technical noise . The multiple levels of barcoding used in scRNA-seq experiments create additional challenges in data processing together with new opportunities for quality control ( QC ) . Different protocols use different barcode configurations , which means a flexible approach to data preprocessing is required . A large number of software tools have already been tailored to scRNA-seq analysis [4] , the majority of which are focused on downstream tasks such as clustering and trajectory analysis . Methods for preprocessing tend to focus on specific tasks such as UMI-tools [5] , umitools ( http://brwnj . github . io/umitools/ ) and umis [6] which have been developed for handling random UMIs and correcting UMI sequencing errors . Other tools such as CellRanger [7] , dropEst [8] and dropseqPipe ( https://github . com/Hoohm/dropSeqPipe ) on the other hand offer a complete preprocessing solution for data generated by droplet based protocols . Other packages such as scater [9] , and scran [10] work further downstream by preprocessing the counts to perform general QC and normalization of scRNA-seq data . scPipe was developed to address the lack of a comprehensive R-based workflow for processing sequencing data from different protocols that can accommodate both UMIs and sample barcodes , map reads to the genome and summarise these results into gene-level counts . Additionally this pipeline collates useful metrics for QC during preprocessing that can be later used to filter genes and samples . In the remainder of this article we provide an overview of the main features of our scPipe software and demonstrate its use on various in-house generated and publicly available scRNA-seq datasets .
scPipe is an R [13] / Bioconductor [14] package that can handle data generated from all popular 3’ end scRNA-seq protocols and their variants , such as CEL-seq , MARS-seq , Chromium 10X and Drop-seq . Data from non-UMI protocols generated by Smart-seq and Smart-seq2 can also be handled . The pipeline begins with FASTQ files and outputs both a gene count matrix and a variety of QC statistics . These are presented in a standalone HTML report generated by rmarkdown [15] that includes various plots of QC metrics and other data summaries . The scPipe package is written in R and C++ and uses the Rcpp package [16 , 17] to wrap the C++ code into R functions and the Rhtslib package [18] for BAM input/output . The key aspects are implemented in C++ for efficiency .
The scPipe package is available from https://www . bioconductor . org/packages/scPipe . Code for each of the example analyses described in the ‘Using scPipe’ section above is available from http://bioinf . wehi . edu . au/scPipe/ . Bug reports and questions about using scPipe should be posted on the Bioconductor support site ( https://support . bioconductor . org/ ) . With the growing popularity of scRNA-seq technology , many tools have been developed for normalization , dimensionality reduction and clustering . There are relatively few packages designed to handle the raw data obtained from the various 3’ end sequencing protocols with their associated UMIs and cell-specific barcodes from beginning to end and collect detailed quality control information . The scPipe package bridges this gap between the raw FASTQ files with mixed barcode types and transcript sequences and the gene count matrix that is the entry point for all downstream analyses . scPipe outputs numerous QC metrics obtained at each preprocessing step and displays these results in an HTML report to assist end users in QC evaluations . Future improvements that are planned for scPipe include support for new scRNA-seq protocols as they emerge and parallelization of the various preprocessing steps to enable scalability to larger datasets . We also plan to generate a more comprehensive scRNA-seq benchmark dataset to ensure the default UMI correction and quality control methods used in scPipe are optimal that will also allow for a more detailed comparison of scPipe with other relevant analysis pipelines , such as zUMIs .
|
Biotechnologies that allow researchers to measure gene activity in individual cells are growing in popularity . This has resulted in an avalanche of custom analysis methods designed to deal with the complex data that arises from this technology . Although hundreds of analysis methods are available , relatively few deal with raw data processing in a holistic way . Our scPipe software has been developed to fill this gap . scPipe is the first fully integrated R package that deals with the raw sequencing reads from single cell gene expression studies , processing them to the point where biologically interesting downstream analyses can take place . By following community developed standards , scPipe is compatible with many other software packages for single cell analysis available from the open-source Bioconductor project , facilitating a complete beginning to end analysis of single cell gene expression data . This allows various biological questions to be answered , ranging from the identification of novel cell types to the discovery of new marker genes . scPipe promotes reproducibility and makes it easier for researchers to share results and code .
|
[
"Abstract",
"Introduction",
"Design",
"and",
"implementation",
"Availability",
"and",
"future",
"directions"
] |
[
"engineering",
"and",
"technology",
"industrial",
"engineering",
"quality",
"control",
"molecular",
"biology",
"techniques",
"information",
"technology",
"data",
"processing",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"exon",
"mapping",
"computer",
"and",
"information",
"sciences",
"sequence",
"alignment",
"bioinformatics",
"gene",
"mapping",
"gene",
"expression",
"chemistry",
"chromium",
"molecular",
"biology",
"database",
"and",
"informatics",
"methods",
"software",
"engineering",
"genetics",
"biology",
"and",
"life",
"sciences",
"preprocessing",
"physical",
"sciences",
"software",
"tools",
"chemical",
"elements"
] |
2018
|
scPipe: A flexible R/Bioconductor preprocessing pipeline for single-cell RNA-sequencing data
|
A recent study of plasma neutralization breadth in HIV-1 infected individuals at nine International AIDS Vaccine Initiative ( IAVI ) sites reported that viral load , HLA-A*03 genotype , and subtype C infection were strongly associated with the development of neutralization breadth . Here , we refine the findings of that study by analyzing the impact of the transmitted/founder ( T/F ) envelope ( Env ) , early Env diversification , and autologous neutralization on the development of plasma neutralization breadth in 21 participants identified during recent infection at two of those sites: Kigali , Rwanda ( n = 9 ) and Lusaka , Zambia ( n = 12 ) . Single-genome analysis of full-length T/F Env sequences revealed that all 21 individuals were infected with a highly homogeneous population of viral variants , which were categorized as subtype C ( n = 12 ) , A1 ( n = 7 ) , or recombinant AC ( n = 2 ) . An extensive amino acid sequence-based analysis of variable loop lengths and glycosylation patterns in the T/F Envs revealed that a lower ratio of NXS to NXT-encoded glycan motifs correlated with neutralization breadth . Further analysis comparing amino acid sequence changes , insertions/deletions , and glycan motif alterations between the T/F Env and autologous early Env variants revealed that extensive diversification focused in the V2 , V4 , and V5 regions of gp120 , accompanied by contemporaneous viral escape , significantly favored the development of breadth . These results suggest that more efficient glycosylation of subtype A and C T/F Envs through fewer NXS-encoded glycan sites is more likely to elicit antibodies that can transition from autologous to heterologous neutralizing activity following exposure to gp120 diversification . This initiates an Env-antibody co-evolution cycle that increases neutralization breadth , and is further augmented over time by additional viral and host factors . These findings suggest that understanding how variation in the efficiency of site-specific glycosylation influences neutralizing antibody elicitation and targeting could advance the design of immunogens aimed at inducing antibodies that can transition from autologous to heterologous neutralizing activity .
HIV-1 has proven difficult to vaccinate against due , in part , to its ability to generate high levels of genetic diversity . During error-prone reverse transcription , up to 3 . 4 x 10−5 mutations per base pair can be introduced , and recombination can occur between the two viral genomes contained within a single virion [1 , 2] . Since HIV-1 group M emerged in the human population , these diversification mechanisms have led to the evolution of nine genetically distinct subtypes , as well as numerous circulating recombinant forms ( CRFs ) [3 , 4] . An ideal prophylactic vaccine would therefore elicit immune responses capable of blocking the full range of HIV-1 variants to which a population might be exposed , which differs based on geography . The HIV-1 envelope ( Env ) glycoprotein subunits gp120 and gp41 are the targets of neutralizing antibodies , which are perhaps the most important correlate of vaccine-mediated protection against other viral diseases . HIV-1 Env gp120 exhibits the highest amount of amino acid variation throughout the entire proteome , creating a formidable obstacle for generating neutralizing antibody-based protection [5–7] . This is exemplified by the fact that most HIV-1 infected individuals develop robust neutralizing antibodies within a few months against the infecting virus , but these antibodies are strain-specific , and lead to viral escape [8–17] . However , in a substantial fraction of infected individuals , these strain-specific antibodies evolve to acquire neutralization activity against heterologous HIV-1 Env variants over time [18–22] . To gain insight into this process , a previous study evaluated whether early Env diversity impacted neutralization breadth , which was measured at approximately 5 years post-infection in 26 HIV-1 individuals infected predominantly with subtype A [23] . In that study , diversity was measured in the Env gp120 V1-V5 region using proviral DNA-derived sequences sampled between 17 to 299 days after infection . Early Env diversity was positively associated with later development of neutralization breadth , leading the authors to propose a model in which the process of neutralization breadth begins in early infection . Indeed , others have shown that the T/F Env and/or early escape variants can trigger the development of a specific broadly neutralizing antibody lineage [15 , 24–26] . However , our inability to reproduce this phenomenon via vaccination reflects a significant gap in our understanding of this process . The development of neutralization breadth during HIV-1 infection is best understood as being a complex and continuous interplay between the host immune response and the evolving viral quasispecies [15 , 17 , 24–32] . Viral replication capacity , Env diversity , super-infection , and high viral load may all contribute to the antigenic stimulation necessary to augment heterologous neutralization breadth [21 , 23 , 27 , 31 , 33–36] . Certain HIV-1 Env characteristics , often related to subtype or glycans , may also favor the development of neutralization breadth [9 , 19 , 33 , 36–38] . Host immunologic factors such as T follicular helper cell activity and activation within the B and T cell compartments could also modulate the development and maintenance of neutralization breadth [39–41] . MHC Class I alleles that target CD8 T cell epitopes and influence HIV-1 disease progression could also impact the development of neutralization breadth [36 , 42] . On the other hand , despite strong evidence that HIV-1 pathogenesis and immune activation differs between males and females [43–48] , a recent study carried out by the IAVI Protocol C team of investigators was among the first to report that there was no difference in the development of neutralization between males and females [36] . In that study , 439 participants from IAVI Protocol C at nine sites across Uganda , Kenya , Rwanda , Zambia , and South Africa were evaluated for heterologous neutralization potency and breadth starting at approximately 24 months after infection . These individuals were infected predominantly with HIV-1 subtypes A1 , C , and D , the proportion of which varied depending on the location of the site . The strongest correlates of neutralization breadth that emerged in the IAVI cohort study were high viral load , low CD4 T cell count , subtype C HIV-1 infection , and HLA-A*03 genotype . Thus , from a broad perspective , there are viral and host factors that create a more favorable environment for neutralization breadth to develop during natural infection . However , none of these factors can fully predict whether an individual will develop strong neutralization . We therefore investigated early viral and immune events in individuals from this same cohort to determine their role in priming the development of neutralization breadth within a natural infection setting . In the present study , we have examined features of the T/F Envs , patterns of early Env diversification , and the autologous neutralization responses in early HIV-1 infection . Our analysis revealed that T/F Envs with fewer NXS-encoded glycan motifs , which are associated with less efficient N-linked glycosylation , were correlated with the development of greater neutralization breadth . Of note , a previous study reported that the opposite association was found in early Envs from a clade B cohort , suggesting that Env glycosylation could be a strong determinant of breadth in a clade-specific manner [38] . Neutralizing antibody-driven diversification in the Env gp120 V2 , V4 , and V5 regions was also a strong contributor to the development of heterologous neutralization breadth . Thus , our finding that correlates of neutralization breadth are present in a panel of individuals in the first weeks to months after infection suggests that this process must be initiated and perpetuated during early infection by Env and antibody interactions , but is further amplified over time by a complex constellation of additional viral and host factors [36] .
As the first step to understanding the development of neutralization breadth , we characterized the T/F Env variants , early longitudinal Env variants , and the autologous neutralizing antibody response in HIV-1 infected individuals from Lusaka , Zambia ( n = 12 ) and Kigali , Rwanda ( n = 9 ) . We have reported that early autologous neutralizing antibody responses that develop in individuals in these cohorts are often potent but are invariably specific for the infecting strain [9–12] . Indeed , in the parent IAVI cohort study , heterologous neutralization was detected in less than 2% of individuals at 24 months post-infection [36] . The ability to neutralize heterologous Env variants that the immune response has not encountered , loosely defined as neutralization breadth , has been shown to develop in plasma by around 2–4 years after HIV-1 infection , in anywhere from 10 to 80% of individuals , depending on the nature of the cohort , timing of sample collection , and the methods used to evaluate and quantify breadth [18 , 21 , 22 , 33 , 34 , 36 , 41 , 49 , 50] . In agreement with these findings , neutralization scores indicative of breadth in the IAVI study were generally first observed at a mean of 3 . 5 years [36] . Therefore , to assess the level of neutralization breadth that developed in our smaller subset of individuals , we independently evaluated plasma samples collected at a median of 3 . 0 years ( ranging from 1 . 1 to 3 . 5 years ) after the first p24 antigen positive test ( Fig 1; collectively referred to as 3-year plasma ) for their capacity to neutralize a panel of 12 globally representative , tier 2 Env pseudoviruses , from a parent panel of 219 HIV-1 Env variants , as described by [51] . Numerous other studies have characterized HIV-1 plasma neutralization breadth using similar panels of 6 to 12 genetically diverse Env variants [11 , 36 , 40 , 50–56] . Since neutralization by the plasma samples at the lowest dilution tested ( 1:20 ) did not always achieve 50% inhibition , we calculated the Area Under the Curve ( AUC ) for each plasma-Env combination ( 21 plasma samples x 12 Envs ) for a total of 252 infectivity curves each containing a series of 5-fold dilutions . A heatmap of the neutralization AUC values is shown in S1A Fig . Neutralization IC50 titers were also calculated for each plasma-Env combination , and are provided in a heatmap shown in S1B Fig . These parameters were highly correlated as expected ( Spearman’s r = -0 . 99 , p < 0 . 0001 ) . Using this approach , plasma samples with the most potent neutralizing activity yielded the smallest AUCs , while large AUC values were observed for poor neutralizers , thus inversely related to the IC50 titer dilution . The median AUC was calculated for each individual’s plasma S1A Fig and used as a continuous variable to rank neutralization breadth [51 , 57] . Individual plasma neutralization curves for the subjects with the highest ( Z1800M ) and lowest ( R1135M ) overall neutralization breadth against the 12 reference Envs are shown in Fig 2A and 2B . Z1800M plasma completely neutralized ten of the reference panel Envs , which included subtypes C , B , A , G , and recombinant forms CRF01_AE , CRF07_BC , and AC , with IC50 titers in the 1:100 to 1:1 , 000 range ( Fig 2A ) . The neutralizing activity against the two remaining Envs did not reach 100% but was substantial . This participant was also ranked as the best neutralizer out of 439 individuals tested in the larger IAVI study [36] . In contrast , plasma from subject R1135M had very limited neutralizing activity against all panel Envs , including the clade-matched subtype A reference Env , which did not reach 50% neutralization ( Fig 2B ) . R1135M was ranked at #372 out of 381 participants whose follow-up extended across 24 to 48 visit months in the IAVI study , and thus exhibited poor neutralization breadth in both studies [36] . Overall , the median AUC values for the 21 plasma samples ranged from high ( AUC = 0 . 57 ) to low ( AUC = 4 . 13 ) neutralization breadth ( Fig 2C ) . For the 12 Protocol C participants that were independently analyzed in the present study and in the larger IAVI study ( i . e . had an IAVI breadth score from [36] , shown in Fig 1 ) , we compared the median AUC calculated here to the maximum breadth score ( sMAX ) assigned in the larger study , and found that the ranking of these individuals was highly correlated as expected ( S2 Fig; Spearman’s r = -0 . 79 , p = 0 . 002 ) . Median AUC values were also significantly correlated to breadth scores at Protocol C visits 36 and 42 ( S2 Fig; r = -0 . 76 , p = 0 . 002; r = -0 . 65 , p = 0 . 006 , respectively ) . Thus , despite using a smaller and less diverse population of HIV-1 infected subjects , a continuous variable ranking system , a different reference Env panel , and evaluation of a single time point , we detected a quantitatively similar spectrum of neutralization breadth as that reported in the IAVI study . Features of HIV-1 Env have been previously associated with the level and type of neutralization breadth that develops within an individual [9 , 33 , 38] . Indeed , in the IAVI study involving multiple HIV-1 suptypes , subtype C infection was a significant correlate of higher neutralization breadth score [36] . However , depending on the segment of the viral genome used and the timing of sample collection , subtype determination may not fully reflect the Env glycoproteins , and also may not represent the Env of the T/F virus . Here , we assigned subtypes based on the full-length T/F Env sequences , which are the antigens that initiated the autologous neutralizing antibody response . We utilized plasma and/or PBMC samples collected at a median of 28 estimated days after infection ( range 22 to 65 days; Fig 1 ) to capture the T/F Env sequences using single genome PCR amplification ( SGA ) [9 , 61–64] . We utilized the LANL Highlighter tool ( https://www . hiv . lanl . gov/content/sequence/HIGHLIGHT/highlighter_top . html ) and the LANL Poisson Fitter v2 tool ( http://www . hiv . lanl . gov/content/sequence/POISSON_FITTER/poisson_fitter . html ) to characterize the T/F Env sequences ( Highlighter plots of all T/F Env sequences are shown in S3 Fig ) . Between 5 and 31 T/F Env sequences were analyzed for each individual , noting that fewer sequences were available for Z185F , Z221M , Z205F , Z153M , and Z201M , the subjects that were studied prior to the initiation of Protocol C . Two individuals , R53F and Z1800M , were infected with two distinguishable variants that differed systematically at one or two polymorphisms , respectively . These observations could indicate infection by two highly similar but distinct variants from the donor quasispecies , or reflect early selection pressure on Env . Since we do not have the donor Env sequences from these individuals , we cannot distinguish between the two scenarios . Regardless , only the dominant T/F polymorphism was observed at subsequent time points for both individuals . Overall , the analysis of T/F Env sequences suggested infection by a single variant in majority of individuals ( S3 Fig ) . This predominance of single variant infections is consistent with our previous study of subtype A and C HIV-1 infected transmission pairs in PSF and ZEHRP , where 18 out of 20 infections were initiated by a single variant [64] . The subtype of the T/F Envs was then determined using the LANL Recombinant Identification Program ( RIP ) ( http://www . hiv . lanl . gov/content/sequence/RIP/RIP . html ) [65 , 66] or the REGA HIV-1 Automated Subtyping Tool ( http://dbpartners . stanford . edu:8080/RegaSubtyping/stanford-hiv/typingtool/ ) [67] . Seven of the 9 PSF participants in Kigali were infected with an HIV-1 subtype A1 Env variant , and the remaining 2 were each infected with a unique A/C recombinant Env ( S4 Fig ) . All 12 individuals from the ZEHRP site in Lusaka were infected with a subtype C HIV-1 Env variant . This distribution of subtypes is also consistent with our previous studies in the ZEHRP and PSF cohorts [9 , 62 , 64 , 68] and with the larger IAVI study [36] . A number of studies have attempted to identify amino acid signatures , glycosylation patterns , and effects of variable loop lengths in Env that are associated with the development of neutralization breadth , as this could reveal an attractive target for vaccine immunogen design [38 , 69 , 70] . To this end , we sought to identify signatures of breadth in the T/F Envs in our cohort . Amino acid signatures of neutralization breadth previously identified by others did not emerge as correlates in our cohort [38 , 69] . We also examined whether the length of the variable loops or the total number of N-linked glycosylation motifs correlated with neutralization breadth ( Fig 3A ) . Interestingly , the feature that was significantly correlated with breadth in our cohort was the ratio of NXS to NXT glycosylation sites found within the gp120 region of T/F Envs ( Fig 3B , S5 Fig ) . N-linked glycosylation motifs are generally encoded with the amino acid motifs NXS or NXT , with X indicating any amino acid except proline; however , glycosylation at NXT occurs at a higher probability than NXS that can be as much as 40% [71 , 72] . Interestingly , van den Kerkhof et al previously reported a connection between the NXS/total glycans ratio in early Envs and the development of breadth in a subtype B HIV-1 infected cohort [38] . However in that study , the authors observed a higher ratio of NXS to total glycosylation sites in patients who developed breadth , postulating that a lower probability of Env gp120 glycosylation could favor breadth . We observed a contrasting effect in our mixed subtype , non-B cohort ( Fig 3 ) . While the total number of glycosylation sites in gp120 did not correlate with breadth , the ratio of NXS to NXT sites was correlated with neutralization breadth AUC ( r = 0 . 56 , p = 0 . 008 ) . That is , individuals infected with T/F Env variants encoding fewer NXS glycan motifs ( r = 0 . 50 , p = 0 . 02 ) and more NXT sites ( r = -0 . 45 , p = 0 . 04 ) in gp120 , and by extension a higher probability of glycosylation , were more likely to develop greater breadth . The discrepancy between van der Kerkhof et al and our study could reflect differences in the cohorts , experimental methods , Env sampling times , or time frame when neutralization breadth was measured . Furthermore , numerous biological distinctions have been reported in Env glycosylation , antigenicity , and immunogenicity , as well as in transmission and disease progression , based on HIV-1 subtype [6 , 9 , 36 , 46 , 64 , 68 , 73–80] . Indeed striking differences in conservation and variability of glycosylation sites are seen within and between HIV-1 subtypes [76 , 81 , 82] . Even the presence of relatively ‘conserved’ glycosylation positions can vary from less than 25% to over 90% , depending on subtype [76] . It is therefore not surprising that the efficiency of early Env glycosylation could impact the development of neutralization breadth in a clade-specific manner . We next sought to gain insight into whether early Env diversification influenced subsequent neutralization breadth in our cohort of subjects . To understand the shifts from the T/F Env to a more diverse quasispecies due to selective pressures , we analyzed SGA derived full-length ‘early’ Env sequences derived from plasma and/or PBMC samples collected after the development of potent strain-specific neutralizing antibodies [9 , 11 , 12] . Because nucleotide-based dN/dS analyses are problematic when resampling the same population over short periods of time [83] , and do not fully reflect the range of diversity that occurs within Env at the amino acid level ( conservative vs . non-conservative amino acid changes , insertions and deletions , alterations to glycosylation sites ) , we developed and implemented a novel approach for statistically quantifying early Env diversity . First , FASTA amino acid alignments of the T/F Envs and early Envs from the same patient were subjected to Sequence Harmony analysis to identify amino acid positions that were significantly different between the two groups of sequences ( Z-score ≤ -3 ) [38 , 63 , 84] . As a proof of concept for this approach , we first analyzed sequences from subject R880F , whose early neutralizing antibody responses and escape pathways had been previously mapped in our laboratory in great detail [11] . Sequence Harmony analysis not only confirmed our previous findings of neutralizing antibody escape mechanisms at 3-months and 6-months after infection ( Fig 4A and 4B , respectively ) , which were characterized through extensive Env mutagenesis and neutralization assays , but also identified an additional four positions where the amino acid composition at 6-months post-infection had significantly shifted from the T/F Envs ( Fig 4B ) . Sequence Harmony analysis was therefore performed on Env sequences from 12 subjects who had sufficient numbers of Env sequences available ( minimum of 15 total sequences ) from an early longitudinal plasma sample , which ranged from 3 . 7 to 7 . 7 months after infection . Next , to take into account the biochemical nature of the identified changes in amino acid composition , we developed an Immunotype Diversity Index ( IDI ) that reflected the complexity of diversification that occurred in Env , particularly in highly variable regions . This approach is based on the premise that antibodies must be exposed to variation within their epitopes , i . e . immunotypes , to acquire heterologous neutralization breadth [19 , 27] . The IDI score was calculated as follows: For amino acid positions with Z-scores < -3 , a conservative amino acid change was given 0 . 5 points; while a non-conservative change , an insertion or deletion , and a purifying selection event were each given 1 point . It is important to note , that this approach takes into account all SGA derived T/F Env sequences , as opposed to using a consensus T/F Env as a point of comparison . This analysis can therefore factor in the presence of amino acid diversity near the time of infection , thus allows for the identification of purifying events . Multiple amino acid changes at a single position ( e . g . in R880F , position 338 shifts from E in the T/F Env to G , A , D , or K in the 6-month Env population , Fig 4B ) were given appropriate points for each possible change ( 0 . 5 or 1 , depending on the nature of the change ) , and added together . Furthermore , if the amino acid change resulted in an introduction , deletion , or shift of a glycosylation site , the significant position was given an additional point . For example , in R880F at position 335 ( Fig 4A ) , there was a non-conservative change from S to N ( 1 point ) that shifted a glycosylation motif ( 1 point ) , giving amino acid position 335 a total of 2 points . The final value for each identified position was multiplied by the absolute value of the calculated Z-score , to factor in the statistical significance of the change . That is , a non-conservative amino acid change that just meets the significance threshold ( -3 ) will have 3 points , but is not weighted equally to the same change with a highly significant Z-score of -30 , which will result in 30 points . The final IDI score was calculated for each patient by taking the sum of all points for all significant positions ( Fig 5 ) . Overall , the IDI scores varied over almost 10-fold , ranging from 263 in Z1800M to 28 in R66M , demonstrating broad variation in the amount of early Env diversity in our cohort ( Fig 5 ) . S6A Fig shows an amino acid alignment for Z1800M , with regions that factored into the IDI score highlighted , as a representative example of high Env diversity that is concentrated in V2 , V4 , and V5 . The same is shown for R66M in S6B Fig to represent low Env diversity . When the IDI scores for full-length T/F Env were plotted against median AUC neutralization breadth scores , there was a significant negative correlation ( Spearman’s r = -0 . 60 , p = 0 . 04; Fig 6A ) . Thus , complex and dramatic amino acid changes associated with stronger selection in Env were correlated with greater breadth . However , much of the significant Env amino acid diversity was occurring in the V2 , V4 , and V5 regions of Env gp120 ( see S6A Fig for an example ) . When IDI scores were calculated using only these three regions , a much stronger correlation was observed ( Spearman’s r = -0 . 80 , p = 0 . 003; Fig 6B ) . Thus , complex changes in V2 , V4 , and V5 of gp120 appear to play a particularly strong role in driving the development of neutralization breadth in our cohort . Sequence diversity within Env during early HIV-1 infection is likely to be the result of pressure exerted predominantly by replicative capacity , cytotoxic T-cell responses , and autologous neutralizing antibodies [10–12 , 85–87] . For 11 of the 21 individuals , we had sufficient samples available to directly evaluate autologous neutralization against the T/F Env and contemporaneous Envs at an early time point ranging from 2 to 12 months ( median of 5 months ) . As autologous neutralizing antibody responses are generally more potent that heterologous responses , we calculated the neutralization IC50 titers for the T/F Env and contemporaneous Envs for these 11 individuals . If we were unable to calculate the IC50 titer due to lack of neutralization activity , we used a value of 20 . The autologous neutralization IC50 titers for the T/F Envs ranged from 32 to 2 , 580 , while the titers for neutralization of the contemporaneous Envs were lower , ranging from 20 to 267 ( Figs 7 and 8 ) . As a group , the T/F Envs were significantly more sensitive to neutralization by the early autologous plasma than were the contemporaneous Envs , indicating that escape from autologous neutralizing antibodies had occurred during this time period ( Fig 7 , Wilcoxon matched-pairs signed rank test , p = 0 . 002 ) . The magnitude of escape was then quantified by dividing the T/F IC50 titer by the contemporaneous titer . The magnitude of escape ranged from 1 ( i . e . Z201M ) , meaning there was essentially no difference between neutralization susceptibility of the T/F Env and the contemporaneous Envs , to 112 ( Z185F ) ( Fig 8 ) . We also evaluated the ability of the 3-year ‘late’ plasma ( that was used to determine breadth ) to neutralize the T/F Env , and these IC50 titers ranged from 850 to 25 , 342 . We next performed a non-parametric Spearman’s correlation analysis using these variables , along with the ratio of NXS:NXT sites in the T/F Env and the 12-month viral load , to examine potential relationships between these factors and the development of neutralization breadth . As expected , the neutralization IC50 and AUC values against the global reference panel were strongly inversely correlated ( Fig 9; Spearman’s r = -0 . 99 , p < 0 . 0001 ) . Interestingly , even within this smaller subset of subjects , the ratio of NXS to NXT glycan motifs in the T/F Env gp120 proteins was again significantly correlated with the development of neutralization breadth using either measurement ( Fig 9; Spearman’s r = -0 . 82 , p = 0 . 003 for IC50; r = 0 . 80 , p = 0 . 005 for AUC ) . The NXS:NXT ratio was also strongly and inversely correlated with the magnitude of escape ( Spearman’s r = -0 . 83 , p = 0 . 002 ) . Thus , potentially higher efficiency of glycosylation in gp120 was associated with a higher magnitude of early escape from autologous neutralizing antibodies and the development of neutralization breadth . The potency of T/F Env neutralization was also correlated directly and significantly with the magnitude of escape ( Spearman’s r = 0 . 76 , p = 0 . 008 ) , underscoring the importance of a dynamic relationship between the T/F Env , early neutralizing antibodies , and viral escape . The importance of this early co-evolutionary process is further substantiated by the fact that both the V2-V4-V5 IDI and the magnitude of escape were also strongly correlated with the development of greater breadth ( V2-V4-V5 IDI: Spearman’s r = 0 . 89 , p = 0 . 012 for IC50; r = -0 . 89 , p = 0 . 012 for AUC; Escape: Spearman’s r = 0 . 75 , p = 0 . 010 for IC50; r = -0 . 77 , p = 0 . 008 for AUC ) .
A major goal for HIV-1 vaccine development is to induce neutralizing antibodies that are capable of broadly protecting against a globally diverse population of viral variants . To this end , various studies have attempted to identify signatures within HIV-1 Env or clinical/host factors that correlate with the development of breadth [10 , 21 , 22 , 31 , 36 , 38 , 69 , 77 , 88] , on the basis that they could be targeted by immunization strategies . We postulated that factors present as early as several weeks after infection could also be determinants for subsequent development of breadth , as early Env diversity was previously linked with breadth [23] . Using a panel of individuals exhibiting a spectrum of neutralization breadth that was highly correlated with that found for the larger parent cohort [36] , we analyzed the inter-dependent relationships between the T/F Env; Env diversification , neutralizing antibodies , and viral escape during the first months of infection; and development of neutralization breadth several years later . Our focus on early infection stems partly from the observation that generic correlates ( i . e . high viral load , subtype C infection ) cannot fully explain the development of breadth , as there are many individuals that have these characteristics but fail to develop broad heterologous neutralization capacity . In addition , understanding which early viral and immune events are associated with generating desirable antibody responses during infection could most directly inform strategies to transition from vaccine-induced strain-specific autologous neutralizing antibodies , which have been recently elicited by trimer-based immunogens [89 , 90] , to those with broader neutralizing capacity . Our observations suggest that the long process of developing neutralization breadth is initiated at least in part by glycosylation of the T/F Env , and is perpetuated by antibody-driven Env diversification in gp120 hyper-variable regions . Our findings are also consistent with previous reports that tracked broadening of a neutralizing antibody lineage within single individuals [24 , 26 , 27] . However , further augmentation by factors linked to viral load or disease progression may also be necessary to develop high levels of neutralization breadth , but are more difficult to translate into immunization strategies [36] . Currently , it is unknown how parameters such as viral subtype and HLA-A alleles could contribute to the broadening of the neutralizing antibody response . One possible explanation is that subtype C HIV-1 infection generates higher plasma viral loads than subtype A infection [46] . Furthermore , HLA alleles can influence the immune response to HIV-1 infection and disease progression via both innate and adaptive immune pathways in the same cohort , and thus could have durable or transient effects that act at different stages of infection [46] . Regarding the impact of the T/F Env , our finding that fewer NXS glycan motifs favor the development of neutralization breadth highlights the important and incompletely understood relationship between Env glycosylation and immunogenicity . Various models of glycan site occupancy have revealed a strong preference for glycosylation at NXT over NXS [71 , 72 , 91–94] . This could result from NXT providing a more optimal conformation of the acceptor sequence , with increased nucleophilicity of the asparagine amide group [91] . Furthermore , kinetics studies have demonstrated that the eukaryotic oligosaccharyltransferase enzyme has a higher affinity for NXT sites compared to NXS [95 , 96] . NXT motifs are also positively selected for in adjacent sequons , and are more likely than NXS to be glycosylated in this setting [97] . Other factors , such as the flanking sequences , tertiary protein structure , and proximity to the C-terminus of the protein , as well as the amino acid at the X position , could also influence the probability and extent of glycosylation at individual sites , but have not systematically been investigated [98 , 99] . Although glycosylation is a common protein modification , lentiviral proteins are particularly heavily glycosylated [100] , with carbohydrate moieties accounting for about 50% of the mass of HIV-1 Env [101 , 102] . Remarkable variation in the number and position of glycan motifs across variants has long been recognized as a defining feature of HIV-1 Env [76 , 81 , 82] , and evasion from neutralizing antibodies has been proposed as a major function of this ‘glycan shield’ [8] . In the context of the Env trimer , an extensive glycan ‘canopy’ consists of around 90 N-linked carbohydrate moieties that exist in crowded and dispersed configurations , are under-processed compared to host glycans , and form inter-dependent clusters [103 , 104] . The oligomeric nature and dense packing of glycans of HIV-1 Env restricts its glycan processing in the golgi such that most glycans are oligomannose , but micro-heterogeneity in carbohydrate forms is observed at individual glycan addition sites [103–106] . The glycan array is known to have a strong influence on Env structure and antigenicity , as well as modulating antibody recognition . Indeed , removal of a single glycan site can influence neutralizing antibody recognition at distal locations as well as reducing the overall oligomannose content by more than predicted [104] . Glycosylation of gp120 could also influence the mucosal transmissibility of HIV-1 , depending on the clade and cohort examined [64 , 68 , 78 , 79 , 107 , 108] . Our findings , combined with those from previous studies , suggest that the use of NXS vs . NXT in HIV-1 Env could be an additional means for the virus to modulate its structure , immunogenicity , and sensitivity to neutralizing antibodies [10 , 38] . Recovery and characterization of numerous bnAbs from chronically infected individuals has demonstrated that many of these antibodies have the capacity to target glycans directly [56 , 109–122] . Even bnAbs that do not target glycans directly , such as VRC01 , have acquired unique adaptations that allow the antibody to avoid glycan clashes and tolerate Env diversity [123–125] . Further underscoring the importance of Env glycosylation to vaccine development , Env SOSIP trimer immunogens based on clade B JR-FL and clade A BG505 Envs elicited tier 2 neutralizing antibodies , but their activity was restricted at least in part by the fact that the neutralizing antibodies that targeted glycan ‘holes’ , where a particular glycan motif is absent [89 , 90] . Although we did not find that the absence or presence of a specific glycan ( s ) was associated with the development of breadth in our cohort , it is possible that T/F Envs with fewer NXS sites are less likely to have glycan ‘holes’ , and preferentially elicit antibodies with a greater capacity to acquire breadth in response to Env diversification . Thus , as stated by [126] , glycans have ‘enormous relevance’ to HIV-1 vaccine design . It is therefore notable that a potential marker for efficiency of Env glycosylation has emerged in this study , and in a distinct study by [38] , as an early clade-specific signal that may contribute to the development of neutralization breadth . Currently , a substantial effort is being devoted to designing and testing novel HIV-1 Env immunogens that preserve features of the native trimer and/or are derived from patients who developed broadly neutralizing antibody activity [89 , 127–133] . These Env immunogens could be readily varied in terms of glycosylation and immunotype diversity , and assessed experimentally , whereas it is less clear how to incorporate sustained viral replication , continuing Env diversity , genotypic host features , and other correlates of breadth via a vaccination protocol . Our results are encouraging for vaccine design because they suggest that modulating Env glycosylation in a clade-specific manner could produce immunogens that are even better suited to elicit neutralizing antibodies with the potential to transition to heterologous breadth . Then , if one can develop immunotype mimics based on diversification of the hyper-variable domains , antibody lineages may be able to develop tolerance for diversity and glycans , and acquire heterologous neutralizing activity to relevant targets . On the other hand , Env diversity presented in the absence of the neutralizing antibody response that provided the selective pressure may be unable to fully recapitulate the evolution of bnAb in immunized animal models [134 , 135] . However , the strategic selection and presentation of the priming Env appears to be of vital importance , as we have shown for SIV vaccination in nonhuman primate studies [136] . It is important to consider that , during HIV-1 infection , neutralizing antibodies tend to have limited specificities , including the CD4 binding site , V1V2 , and the base of the V3 loop , that differ between individuals [10–12 , 15–17 , 27 , 137] . It is unclear why some individuals , such as R66M studied here , produce antibodies with limited neutralizing capacity against the autologous T/F Env while others , such as Z1800M , produce potent neutralizing antibodies within the same time frame . It is possible that , in addition to differences in subtype , the higher NXS to NXT ratio of the R66M T/F Env compared to the Z1800M T/F Env could also have initiated this disparity . Our findings highlight a void in our understanding of the importance of the immunogenicity and antigenicity of the T/F Env , and the early anti-Env antibody landscape , in terms of germline activation , clonality , and binding affinity , on the ability to transition from strain-specific to heterologous neutralization . Broadly neutralizing antibody lineages , such as VRC01 or CAP256 , do not develop in isolation; they stem from autologous neutralizing antibodies in an environment that includes competition and influence by a milieu of other B cells and antibodies , occurring in the presence of ongoing viral diversity [27 , 138] . Current studies have begun to bridge the gap between strain-specific and broadly neutralizing antibodies by focusing on the longitudinal evolution of an individual bnAb lineage [24–27 , 138] . Others have developed strategies to design Env antigens that engage specific B-cell germline lineages with known potential to evolve into a broadly neutralizing antibody [139 , 140] . However , even this strategically designed Env immunogen engaged other germlines , in addition to the germline of interest , suggesting that the complexity of the human immune system will be a formidable obstacle to reproducing the activation and maturation of individual bnAb lineages [139] . Taken together , our observations provide new insight into the importance of T/F Env glycosylation , autologous neutralizing antibodies , diversification in gp120 , and viral escape in setting the course to neutralization breadth . With the successful elicitation of tier 2 autologous neutralizing antibodies using trimer-based immunogens [89 , 90] , additional promising Env immunogens on the horizon , and a more complete understanding of glycosylation in the context of several genetically diverse Env trimers [103 , 104] , it may be possible to optimize the glycosylation of these reagents so that they preferentially elicit neutralizing antibodies with the potential to acquire heterologous neutralizing capacity . Whether necessary cycles of antibody-virus co-evolution , and other factors , can then be reproduced in the absence of infection to augment this process will remain to be determined .
The ZEHRP , PSF , and IAVI Protocol C participants were selected based on rapid screening of adults with recent history of HIV exposure in Rwanda and Zambia . After obtaining written informed consent , blood samples were collected from HIV-1 infected participants longitudinally . The procedures for written informed consent and research activities were approved by institutional review boards at all collaborating clinical research centers , with further compliance to human experimentation guidelines set forth by the United States Department of Health and Human Services . The study was reviewed and approved by the Republic of Rwanda National Ethics Committee , Emory University Institutional Review Board , and the University of Zambia Research Ethics Committee . The 21 HIV-1 infected individuals studied were identified shortly after a transmission event through their enrollment in two HIV-discordant couple cohorts ( see Fig 1 ) . Zambian participants were from the Zambia-Emory HIV Research Project ( ZEHRP , established in 1994 by Dr . Susan Allen ) in Lusaka and the Rwandan participants were from Projet San Francisco ( PSF , established in 1986 by Dr . Susan Allen ) in Kigali . Together , these two sites accounted for roughly 43% of participants in the larger IAVI study [36] . These projects were originally designed as ‘couples voluntary counseling and testing’ clinics and have been described previously [64 , 141] . ZEHRP and PSF are also part of the Rwanda Zambia HIV Research Group at Emory University ( RZHRG; http://www . rzhrg . org ) . More recently Protocol C , a uniform vaccine-preparedness study developed and implemented by the International AIDS Vaccine Initiative ( IAVI; http://www . iavi . org ) , was initiated and carried out at multiple sites in Africa , including ZEHRP and PSF [80] . The 21 individuals studied here were selected prior to the initiation of the larger IAVI study . The subjects were chosen based on three factors: a p24 antigen positive test ( i . e . infection detected in the early Fiebig stages ) ; the availability of early longitudinal plasma samples; and for Protocol C participants , the availability of viable PBMC collected during early infection ( for studies that are not included in these analyses ) . Z185F , Z221M , Z205F , Z153M , and Z201M were enrolled in ZEHRP prior to Protocol C . The remaining 16 subjects were enrolled in Protocol C at the time of sampling . Twelve of 16 the Protocol C participants were analyzed in both the present study and in the larger IAVI study ( see Fig 1; note that while there are 13 Parent Study participants with PC codes listed in the second column , Z1022M/PC138 was not analyzed in the larger study and thus has no breadth score in the fifth column ) [36] . The procedures for written informed consent and research activities were approved by institutional review boards at all collaborating clinical research centers , with further compliance to human experimentation guidelines set forth by the United States Department of Health and Human Services . Plasma viral load determinations were underwritten by IAVI and performed at Contract Lab Services ( CLS ) in South Africa using an Abbott m2000 system . The typical range of detection was between 160 and 4×107 RNA copies/ml . HLA genotyping was performed using a combination of PCR-based techniques as described previously [60 , 142] . For 16 of 21 individuals , cDNA synthesis and 384-well single genome PCR amplification ( SGA ) was performed essentially as described in [61] S3 Fig . Briefly , RNA was extracted from cryopreserved patient plasma samples using the QIAmp viral RNA , and reverse transcription was performed using the SuperScript III kit ( Invitrogen ) with reverse primer OFM19 ( 5’- GCACTCAAGGCAAGCTTTATTGAGGCTTA-3’ ) . cDNA was diluted to result in <30% positive wells for SGA . First round PCR was performed in a 15 μL volume using the Phusion Hotstart II High Fidelity DNA Polymerase ( Thermo Scientific ) with forward primer Vif1 ( 5’-GGGTTTATTACAGGGACAGCAGAG-3’ ) and OFM19 . Cycling conditions were 98°C for 2 min; 10 cycles of 95°C for 15 s , 54°C for 60 s , and 68°C for 4 min; 25 cycles of 95°C for 15 s , 54°C for 60 s , and 68°C for 4 min , adding 5 s to the extension per cycle; 72°C for 30 min; and 4°C hold . Second round PCR was performed with the same enzyme in a 10 μL volume with 1 μL of the first round of PCR and EnvA-TOPO ( 5’-CACCGCCTTAGGCATCTCCTATGGCAGGAAGAA-3’ ) and EnvN ( 5’-CTGTCAATCAGGGAAGTAGCCTTGTGT-3’ ) . Cycling conditions were 95°C for 2 min; 30 cycles of 95°C for 15 s , 54°C for 60 s , and 72°C for 2 . 5 min; 72°C for 10 min; and 4°C hold . PCR amplicons were purified using Qiagen PCR Clean-Up Kit . For Z201M , Z205F , Z221M , Z185F , and Z153M , full-length env genes were amplified previously using 96-well SGA from plasma and PBMC , as described [9] . PCR amplified env genes plus flanking sequences were T/A-cloned into one of several pCR3 . 1-based expression vectors as described [9 , 11 , 61] . On average , 13 SGA PCR amplicons per patient ( range 5 to 31 ) per time-point were sequenced with Beckman Coulter Genomics using the following primers: For13 ( 5’- GAGAAAGAGCAGAAGACAGTGG-3’ ) ; For15 ( 5’-CAGCACAGTACAATGTACACATGGAA-3’ ) ; For17 ( 5’- AGCAGCAGGAAGCACTATGGGCGC-3’ ) ; For19 ( 5’-GGAACCTGTGCCTCTTCAGCTACC-3’ ) ; and Rev14 ( 5’-ACCATGTTATTTTTCCACATGTTAAA-3’ ) ; Rev16 ( 5’-ATGGGAGGGGCATACATTGCT-3’ ) ; Rev17 ( 5’- CCTGGAGCTGTTTAATGCCCCAGAC-3’ ) ; and Rev19 ( 5’-ACTTTTTGACCACTTGCCACCCAT-3’ ) . Sequencher v5 was used to generate nucleotide sequence contigs , and sequences with evidence of mixed peaks were omitted from the analysis . Additionally , previously reported sequences were utilized for patients Z205F , Z185F , Z201M , Z221M , Z153M , and R880F [9–12 , 143 , 144] . Geneious v6 . 1 . 7 was used to align and translate nucleotide sequences . Amino acid alignments were exported from Geneious in FASTA format and used to generate Highlighter plots ( http://www . hiv . lanl . gov/content/sequence/HIGHLIGHT/highlighter_top . html ) , to define Env features ( http://www . hiv . lanl . gov/content/sequence/GLYCOSITE/glycosite . html ) and ( http://www . hiv . lanl . gov/content/sequence/VAR_REG_CHAR/index . html ) , and perform Sequence Harmony comparisons ( http://www . ibi . vu . nl/programs/shmrwww/ ) . Subtype reference Env nucleotide sequences were obtained from the Los Alamos Sequence Database ( http://www . hiv . lanl . gov/content/sequence/NEWALIGN/align . html ) . A Neighbor-joining nucleotide phylogenetic tree was generated in Geneious v6 . 1 . 7 . HIV-1 subtyping was performed with REGA HIV-1 Automated Subtyping Tool ( http://dbpartners . stanford . edu:8080/RegaSubtyping/stanford-hiv/typingtool/ ) [67] . The days since the most common recent ancestor ( MRCA ) are presented in Fig 1 , and were determined by analyzing the T/F Env sequences for each subject in the Poisson Fitter v2 tool from the Los Alamos HIV Database ( http://www . hiv . lanl . gov/content/sequence/POISSON_FITTER/pfitter . html ) [58 , 145] . The estimated days since infection values , also presented in Fig 1 , were calculated using the methods of [59] . The T/F env nucleotide sequences have been submitted under Genbank accession numbers KX983471-KX983929 . For calculation of IDI scores , Sequence Harmony comparisons were used to identify amino acid positions significantly different between the T/F and 4–8 month sequences ( Z-score < -3 ) . For significant positions , points were given as follows: 0 . 5 points: conservative changes; 1 point: non-conservative changes , deletions , insertions , purifying selection; 1 point: shift , deletion , or insertion of glycosylation motif . Multiple changes at significant positions were given cumulative points . Sum of points at significant positions were multiplied by the absolute value of the Z-score at corresponding position to weight scores by statistical significance . Final IDI for each patient was calculated by adding all points for all positions . For V2-V4-V5 scores , significant positions were restricted to these locations , as identified by LANL HIV Variable Region Characterization tool . A global panel of 12 HIV-1 Env reference clones developed by deCamp et al . was obtained though the NIH AIDS Reagent Program ( Catalog number 12670 ) [51] . Generation of Env pseudoviruses and performing the TZM-bl neutralization assay has been described previously [9–12 , 40 , 61 , 68 , 136 , 144 , 146–150] . Briefly , Env-expressing plasmids were co-transfected with the HIV-1 SG3ΔEnv proviral backbone into 293T cells using Fugene HD ( Promega ) , and pseudovirus stocks were collected 48h post-transfection , clarified by centrifugation , and frozen at -80°C . In all cases except one , plasma was used to evaluate heterologous and autologous neutralization . Because R66M began antiretroviral therapy approximately 1 . 25 years post-infection , IgG antibodies were purified from 36-month plasma using a GE Healthcare Life Sciences Ab SpinTrap , according to manufacturer instructions ( GE 28-4083-47 ) . The concentration of the purified IgG was determined by ELISA , and was used in place of plasma in neutralization assays . Five-fold serial dilutions of heat-inactivated plasma samples or purified IgG were assayed for their inhibitory potential against the Env pseudoviruses using the TZM-bl indicator cell line , with luciferase as the readout . At 48 hours post-infection , cells were lysed and luciferase activity was measured using a BioTek Cytation 3 imaging reader with Gen5 v2 . 07 Software . The average background luminescence from a series of uninfected wells was subtracted from each experimental well . Assays were run in duplicate and repeated independently at least twice . All graphs were generated in Prism v6 . 0 . Virus infectivity curves were generated and Area Under the Curve was calculated in Prism and used for comparisons of neutralization activity . Neutralization IC50 titers were also calculated in Prism from the virus infectivity curve using log transformation of x-values , normalization of y-values , and linear regression of dose-response inhibition with variable slope . A Mann-Whitney test or Wilcoxon matched pairs analysis was used to compare two groups , while a Kruskal-Wallis test was used for to compare more than two groups . A non-parametric Spearman’s test was used to assess correlations between variables in Prism v6 . 0 . P values less than 0 . 05 were considered to be significant .
|
HIV-1 has proven difficult to vaccinate against due to its ability to generate high levels of genetic diversity , particularly in the envelope glycoproteins . An ideal prophylactic vaccine would therefore elicit immune responses capable of blocking the full range of HIV-1 variants to which a population might be exposed . An essential component of protective immunity against HIV-1 is likely to be an antibody response that is capable of neutralizing genetically diverse HIV-1 viral variants . In an effort to understand how this type of ‘broad’ antibody response develops during natural HIV-1 infection , a large cohort study recently found that certain factors , such as high viral load , HLA-A*03 genotype , and subtype C infection , were correlated with the development of greater neutralization breadth . Here we investigated the viral envelope proteins and antibody responses in early infection in a small subset of individuals from two of the African sites included in the larger cohort study . We found that a marker for the efficiency of envelope glycosylation in the infecting viral variant was strongly correlated with the development of antibodies with greater neutralization breadth . We also found that extensive viral changes in the V2 , V4 , and V5 regions of the envelope gp120 protein , were strongly associated with the development of antibodies with greater neutralization breadth . Based on these results , we propose that more efficient glycosylation of the envelope protein of the infecting viral variant elicits neutralizing antibodies that drive early and complex amino acid changes in gp120 , which triggers the development of antibodies whose neutralization breadth can be augmented over time by additional viral and host factors . These findings suggest that a better understanding of the efficiency of envelope glycosylation in HIV-1 could inform current vaccine strategies aimed at eliciting antibodies with neutralization breadth .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"amino",
"acid",
"sequence",
"analysis",
"rna",
"viruses",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"glycosylation",
"antibodies",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"immune",
"system",
"proteins",
"sequence",
"alignment",
"proteins",
"medical",
"microbiology",
"hiv",
"antigens",
"microbial",
"pathogens",
"hiv-1",
"molecular",
"biology",
"biochemistry",
"post-translational",
"modification",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"lentivirus",
"glycobiology",
"organisms"
] |
2016
|
Diversification in the HIV-1 Envelope Hyper-variable Domains V2, V4, and V5 and Higher Probability of Transmitted/Founder Envelope Glycosylation Favor the Development of Heterologous Neutralization Breadth
|
Although DNA methylation patterns in somatic cells are thought to be relatively stable , they undergo dramatic changes during embryonic development , gametogenesis , and during malignant transformation . The enzymology of DNA methyltransferases is well understood , but the mechanism that removes methylated cytosines from DNA ( active DNA demethylation ) has remained enigmatic . Recently , a role of the growth arrest and DNA damage inducible protein GADD45A in DNA demethylation has been reported [1] . We have investigated the function of GADD45A in DNA demethylation in more detail using gene reactivation and DNA methylation assays . Contrary to the previous report , we were unable to substantiate a functional role of GADD45A in DNA demethylation . The mechanism of active DNA demethylation in mammalian cells remains unknown .
Mammalian DNA methylation patterns are thought to be generally quite stable throughout cell divisions due to faithful maintenance of DNA methylation patterns during DNA replication [2] . There is , however , published data showing that in somatic cells methylated CpGs can be converted to unmethylated CpGs in the absence of DNA replication [3] , [4] , [5] , [6] , [7] . This replication-independent DNA demethylation would imply the existence of a mammalian DNA demethylase enzyme that can either actively remove the methyl group from 5-methylcytosine or can remove the entire methylated base or nucleotide , perhaps in a base excision repair-like pathway . Direct breakage of a carbon-carbon bond seems energetically unfavorable . If not strand-specifically coordinated , the excision repair pathway would put the genome at risk for DNA double strand breakage . The mechanistic steps and proteins involved in DNA demethylation in mammals are currently unknown . In plants , a demethylase pathway involving a DNA glycosylase activity has been identified [8] , [9] , [10] but these proteins do not appear to have mammalian homologues . The mammalian cytidine deaminases AID and APOBEC1 in vitro have 5meC deaminase activity [11] and the deaminated base , thymine , may be removed by base excision repair pathways . One of the best pieces of evidence for an active DNA demethylation pathway in mammalian cells comes from studies of pre-implantation mouse embryos . It has been shown that the paternal genome becomes almost completely demethylated within less than six hours after fertilization . This process must be independent of DNA replication . Astonishingly , the maternal genome resists this genome-wide demethylation process [12] , [13] , [14] , [15] . The nature of the mammalian DNA demethylase has remained obscure . Earlier , the recombinant MBD2b protein has been reported to possess DNA demethylase activity in vitro [16] , although this finding has so far not been repeated in other laboratories [17] , [18] . Staining of 5-methylcytosine with an antibody in fertilized oocytes containing an Mbd2 knockout allele gave results that were the same as wild type controls [19] arguing that MBD2 is not involved in zygotic DNA demethylation . Recently , it was reported that the GADD45A protein promotes demethylation of CpG-methylated DNA [1] . GADD45A ( Growth arrest and DNA-damage-inducible gene 45 alpha ) is a gene induced by a variety of growth arrest conditions and DNA damaging agents . The GADD45A protein is an 18 kDa acidic nuclear protein involved in maintenance of genomic stability , DNA repair , cell cycle checkpoints and suppression of cell growth [20] , [21] . Barreto et al reported that GADD45A has a key role in active DNA demethylation . GADD45A overexpression activated a methylation-silenced reporter plasmid and promoted DNA demethylation [1] . Here , we have further investigated the role of GADD45A in DNA demethylation . However , we were not able to substantiate a role of GADD45A in demethylation of mammalian DNA .
The best evidence for active DNA demethylation occurring in mammalian genomes comes from studies of fertilized oocytes where it has been shown that the paternal genome is actively demethylated before the onset of DNA replication [12] , [13] . Barreto et al identified GADD45A as a putative factor for DNA demethylation by screening a Xenopus cDNA expression library for cDNAs able to activate a methylation-silenced luciferase reporter gene [1] . However , in contrast to mice , the paternal genome of Xenopus is not subjected to active demethylation of 5-methylcytosine immediately after fertilization [22] . We hypothesized that Gadd45a should be expressed in the oocyte or zygote if it plays a role in the mammalian DNA demethylation pathway . The gene expression profile of Gadd45a in mouse developmental stages was assessed by examining the UniGene database . Although this database cannot replace direct analysis of gene expression , it can provide a useful guide . Also , while it is possible that a demethylase pathway is exclusively expressed in oocytes/zygotes , this may not necessarily be the case , as tissue-specific posttranscriptional factors may modify protein levels or activity . With this caveat noted , we observed that Gadd45a is expressed in late developmental stages or in somatic tissues ( Table 1 ) . We compared the expression profile of Gadd45a with the expression profile of PGC7/Stella , a protein known to be important for the prevention of demethylation of the maternal genome in fertilized oocytes [23] . As expected , PGC7/Stella was expressed at high levels in oocytes , unfertilized ova , zygotes and in preimplantation embryos but PGC7/Stella mRNA was virtually absent at later embryonic stages and in adult tissue ( Table 1 ) . Thus , the expression of Gadd45a and PGC7/Stella are almost mutually exclusive , a situation not expected if Gadd45a functions in mammalian demethylation , at least in zygotes . Over the past decade , it has become a difficult issue to explore the biological function of GADD45A [21] . Its role in the induction of apoptosis is unclear . We have not seen apoptotic cells by TUNEL staining in cells transfected with GADD45A and have not seen a substantial cell cycle arrest by FACS analysis of cells transfected with GADD45A ( data not shown ) . The phenotype of cells with deleted Gadd45a is centrosome amplification and mitotic failure [20] . GADD45A interacts with the mitotic kinase Aurora A [24] . The transfected GADD45A protein does indeed interact with Aurora A ( Figure S1 ) attesting to the functionality of the protein . Barreto et al have shown that overexpression of GADD45A can lead to reactivation of a methylation-silenced EGFP reporter gene [1] . We first tested if overexpression of GADD45A has an effect on the expression of methylated reporter plasmids in somatic cells . As a simple assay for detecting demethylation activity , in vitro methylated promoter and reporter genes have been used to detect methylation changes in mammalian cell lines [25] , [26] . As done by Barreto et al . [1] , we used this assay system to quantify a change in the state of expression of the methylated EGFP reporter plasmid ( pEGFP-N2 ) when GADD45A was overexpressed in HEK293 cells . The EGFP reporter gene , which is controlled by the CMV promoter , was in vitro methylated at all CpG sites with SssI DNA methylase and transiently transfected or co-transfected with a GADD45A mammalian expression vector into HEK293 cells . After 48 hours , expression of EGFP was determined by fluorescence microscopy ( Figure 1 ) . The methylated reporter gene was not transfected at a lower efficiency than the unmethylated one as assessed with a co-transfected luciferase expression plasmid ( data not shown ) . While EGFP was expressed from the unmethylated plasmid , the methylated EGFP plasmid was not expressed . EGFP expression in GADD45A overexpressing HEK293 cells was observed at a level similar to that of the control , suggesting that GADD45A does not have an effect on the methylation-silenced EGFP reporter plasmid and does not release methylation silencing ( Figure 1 ) . It was reported that GADD45A promotes active DNA demethylation in mammalian cell lines [1] . We replaced the CMV promoter in the pEGFP-N2 plasmid with the 2 . 4-kb mouse Oct4 gene promoter ( Figure 2A ) . DNA demethylation has been previously studied with murine Oct4-EGFP gene constructs [27] , [28] , [29] . We asked whether GADD45A overexpression induces demethylation of the in vitro methylated mouse Oct4 promoter plasmid in HEK293 cells . The methylation status of the Oct4 promoter-EGFP reporter gene was determined by Southern blot analysis with methylation-sensitive restriction enzyme digestion , as was done by Barreto et al . [1] ( Figure 2 ) , and by bisulfite sequencing ( Figure 3 ) . The pOct4-EGFP construct was in vitro methylated with HpaII and HhaI methylases . The efficiency of methylation of the Oct4-EGFP plasmid was tested by bisulfite sequencing and it was judged to be at least 95% ( Figure 2B ) . In vitro methylated pOct4-EGFP plasmids were recovered from control and GADD45A overexpressing HEK293 cells 48 hours after transfection with or without GADD45A mammalian expression vector . The GADD45A overexpression in HEK293 cells was verified by Western blot analysis with GADD45A-specific antibody ( Figure 2D ) . The transiently transfected HEK293 cells were grown under conditions of serum starvation to separate active DNA demethylation from passive demethylation events due to any DNA replication . It is noted that Barreto et al observed demethylation in both dividing and non-proliferating cells [1] . The recovered plasmids were digested with the methylation-sensitive restriction enzyme HpaII , which cleaves the sequence 5′-CCGG only when it is not methylated . The digested fragments were subjected to Southern blot analysis with a 32P-labeled EGFP probe to determine the degree of demethylation as shown in Figure 2C . However , we failed to detect HpaII-digested fragments from methylated plasmids recovered from GADD45A overexpressing cells . This demonstrates that no significant demethylation occurred in methylated and transfeced pOct4-EGFP plasmids in HEK293 cells with or without GADD45A overexpression . This assay was repeated with HpaII-only methylated pOct4-EGFP , because Barreto et al . showed HpaII-digested EGFP gene fragments when HpaII-only methylated plasmid was recovered from GADD45A overexpressing cells [1] . But , as shown in Figure 2C , we also failed to detect demethylation of pOct4-EGFP when it was in vitro methylated with HpaII only . This result was further confirmed by a more sensitive technique , sodium bisulfite sequencing . As shown in Figure 3 , bisulfite sequencing of six HpaII methylation sites within and upstream of the EGFP gene clearly shows that the level of methylation on the pOct4-EGFP plasmid , which was in vitro methylated by either HpaII single methylation or by HpaII and HhaI double methylation , was not affected by expression of GADD45A . These data suggest that GADD45A does not promote active DNA demethylation . It was reported that GADD45A promotes substantial demethylation of the endogenous Oct4 promoter in mouse NIH3T3 cells although the transfection efficiency for these cells was only ∼30% [1] . Using Amaxa nucleofection technology , we achieved a transfection efficieny of over 50% for NIH3T3 cells as tested with a GFP expression plasmid ( Figure 4A ) . However , despite of the higher transfection efficiency , we were not able to see any significant demethylation of the Oct4 promoter using bisulfite sequencing analysis ( Figure 4 ) . In control vector transfected cells , 34% of the CpGs were unmethylated and in GADD45A-transfected cells , 36 . 5% of the CpGs were unmethylated , a difference which was statistically not significant . At this point , we wished to further verify GADD45A activity towards methylation of other genomic sequences . To address this issue , we evaluated the methylation status of two endogenous single copy genes when GADD45A was overexpressed in HEK293 cells . Recently , we reported that the promoter-associated CpG islands of the RASSF1A and TIG1 genes are highly methylated in HEK293 cells [30] . Therefore , we carried out bisulfite sequence analysis to see if the levels of DNA methylation at the promoters of RASSF1A and TIG1 can be affected by overexpression of GADD45A in HEK293 cells . As illustrated in Figure 5 , the dense methylation status of the RASSF1A and TIG1 promoters generally was maintained in GADD45A-overexpressing cells . We then analyzed methylation of endogenous LINE1 sequences in HEK293 cells . After bisulfite treatment of DNA from control cells and GADD45A overexpressing cells , we used consensus PCR primers for the LINE1 promoter . Combined bisulfite restriction analysis ( COBRA ) with HinfI digestion was used to assess the methylation status of these repetitive elements ( Figure 6 ) . Quantitative analysis of the cleavage products indicates that there was no difference in methylation of LINE1 sequences between control and GADD45A-transfected cells . The DNA excision repair nuclease XPG contributes to repair of various types of DNA adducts . Barreto et al reported that the DNA demethylation and gene reactivation activity of GADD45A was further enhanced by co-expression of XPG or XPG in combination with XPB [1] . We used a methylated SV40-luciferase construct to test the activities of GADD45A and XPG on reporter gene activity ( Figure S2 ) . Neither the co-expression of GADD45A alone nor that of GADD45A plus XPG led to reactivation of the methylated reporter . In contrast to the previous report [1] , expression of the methylated reporter gene was even further suppressed by co-expression of GADD45A .
It has been reported that GADD45A promotes epigenetic gene activation by active DNA demethylation presumably involving a repair-mediated process [1] . This was a surprising result since GADD45A is not expressed at the developmental stage of murine embryos at which active DNA demethylation has been well documented ( Table 1 ) . In addition , it is not immediately apparent why genomic DNA demethylation should occur under conditions where GADD45A is induced , i . e . growth arrest or DNA damage . We have not been able to confirm the biochemical results showing a role of GADD45A in DNA demethylation ( Figures 1–6 ) . The reasons for the discrepancy between our results and those of Barreto et al are not clear . We used the same cell line ( HEK293 ) for transfection of GADD45A and the methylated reporter plasmids . The sequences analyzed for demethylation in the Oct4 promoter-EGFP construct were the same as those analyzed by Barreto et al . The methylases used and the restriction enzyme sites tested were the same . In addition to HpaII cleavage , as done by Barreto et al , we used sodium bisulfite sequencing which is considered the method of choice for high resolution DNA methylation analysis . We were also unable to confirm GADD45A-induced DNA demethylation of methylated endogenous genes ( LINE1 promoter , RASSF1A and TIG1-associated CpG islands in HEK293 cells and the Oct4 promoter in NIH3T3 cells ) . Barreto et al have also used GADD45A-specific siRNAs to support a role of GADD45A in DNA demethylation [1] . However , we point out that an almost 3-fold increase in global genomic methylation levels seen after knockdown of GADD45A ( see Fig . 3c , d of their report ) is virtually impossible since there are not enough unmethylated CpG dinucleotides in the human genome to allow for such an increase . In summary , one can conclude that the identity and composition of the mammalian DNA demethylase machinery remains unknown . Viable candidate genes would be those expressed in oocytes , zygotes , and 1- or 2- cell stage embryos . AID and APOBEC1 are expressed in ovaries and oocytes [11] . Another example is the MBD3L2 protein , which is predominantly expressed at these stages ( Unigene database ) . MBD3L2 is a homologue of the methyl-CpG binding proteins MBD2 and MBD3 that by itself does not contain a methyl-CpG binding domain [31] , [32] . Work is underway in our laboratory to construct gene-targeted mice with a deletion of Mbd3l2 in order to test the role of this gene in early mammalian development and DNA demethylation .
HEK293 and NIH3T3 cells were maintained as a monolayer in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum at 37°C in a 5% CO2 standard incubator . To serum-starve the cells , confluent HEK293 or NIH3T3 cells were cultured in a medium containing 0 . 5% fetal bovine serum for 48 hours after transfection . Transfection of HEK293 cells was carried out in Opti-MEM-I medium ( Invitrogen; Carlsbad , CA ) using Lipofectamine 2000 ( Invitrogen ) according to the manufacturer's instructions . Transfection of NIH3T3 cells was performed by nucleofection using the Amaxa nucleofector and kit R ( Amaxa; Gaithersburg , MD ) , following the protocol recommended by the manufacturer . More than 50% of the NIH3T3 cells and more than 60% of the HEK293 cells were routinely transfected . For the details of the co-immunoprecipitation experiments , see Text S1 . Total cell extracts were prepared with RIPA buffer ( 25 mM Tris-HCl pH 7 . 6 , 150 mM NaCl , 1% NP-40 , 1% sodium deoxycholate and 0 . 1% SDS ) and were separated by SDS/PAGE and blotted onto PVDF membranes ( BioRad ) . The membranes were blocked with 5% nonfat milk at 4°C overnight . After washing , GADD45A protein was detected using rabbit polyclonal IgG ( Santa Cruz , sc-797 ) at a 1∶3 , 000 dilution , followed by peroxidase-conjugated anti-rabbit IgG ( Santa Cruz , sc-2004 ) at 1∶8 , 000 dilution . β-tublin protein was detected using mouse monoclonal IgG ( Santa Cruz , sc-5274 ) at 1∶400 dilution , followed by peroxidase-conjugated anti-mouse IgG ( Santa Cruz , sc-2005 ) at 1∶5 , 000 dilution . The signal was visualized by using ECL-Plus ( Amersham Pharmacia Biotech ) . HEK293 cells were plated at a density of 2 . 5 × 105 cells per well directly on coverslips in a six-well tissue culture dish . The transfection was achieved using Lipofectamine 2000 ( Invitrogen ) according to the manufacturer's instructions using 2 µg of plasmids . At 48 hours after transfection , cells were washed twice with PBS and then fixed in 3 . 7% formaldehyde for 5 min at room temperature . The cells were washed again , stained with 0 . 25 µg/ml of 4′ , 6′-diamidino-2-phenylindole ( DAPI ) ( Sigma ) , and mounted with 90% glycerol in PBS . Images were visualized with an Olympus IX81 fluorescence microscope . The full-length cDNA encoding human GADD45A was obtained by reverse transcription-PCR . For creating a GADD45A mammalian expression vector ( pcDNA3 . 1-GADD45A ) , the GADD45A cDNA was cloned into the BamHI-XhoI sites of pcDNA3 . 1 ( + ) vector ( Invitrogen ) . The inserted cDNA of the construct was sequenced and verified by comparing with the human GADD45A transcript ( RefSeq; NM_001924 . 2 ) in the Ensemble database . For the pOct4-EGFP construct , the cytomegalovirus ( CMV ) promoter of the pEGFP-N2 vector ( Clontech; Palo Alto , CA ) was replaced with the 2 . 4-kb mouse Oct4 gene promoter . For this purpose , the AseI and EcoRI fragment of the pEGFP-N2 plasmid encoding the CMV promoter was excised by restriction nuclease cleavage . Next , the 2 . 4-kb Oct4 promoter region upstream of the translational start site was amplified by PCR followed by AseI and EcoRI digestion of the PCR product and insertion into the AseI and EcoRI sites of the pEGFP-N2 plasmid . The constructed pOct4-EGFP plasmid was verified by DNA sequencing . For in vitro methylation of plasmids , 10 µg each of pEGFP-N2 or pOct4-EGFP were methylated in vitro with 20 units of SssI , HpaII alone or HhaI plus HpaII DNA methylases ( New England Biolabs ) , respectively . These methylated plasmids were then phenol/chloroform extracted , ethanol-precipitated and resuspended in TE buffer . The extent of methylation was confirmed by methylation-sensitive restriction enzyme , HpaII , digestion and evaluated with bisulfite sequencing for pOct4-EGFP ( Figure 2 ) . Transfected plasmids were recovered by using QIAquick PCR-purification kits ( Qiagen ) . The plasmids were digested with HincII and HpaII and analyzed by Southern blotting using an EGFP gene probe . Bisulfite conversion and purification of DNA for methylation analysis was accomplished using the EpiTect Bisulfite kit ( Qiagen ) . One µg of recovered plasmids and genomic DNAs from HEK293 or NIH3T3 cells was used for the analysis . Bisulfite modified DNA was amplified using the following primers: for analyzing the six HpaII methylation sites downstream of the Oct4 promoter of pOct4-EGFP , the forward primer 5′-TTAGAGGTTAAGGTTAGAGGGTGG-3′ and reverse primer 5′-ATAATACAAATAAACTTCAAAATCAACTTA were used . For the CpG island of the RASSF1A promoter , the forward primer 5′-AGTTTTTGTATTTAGGTTTTTATTG-3′ and reverse primer 5′-AACTCAATAAACTCAAACTCCCC were used . For the CpG island covering the TIG1 promoter , the forward primer 5′-AGGAGTGGTTTTATGGGGAT-3′ and reverse primer 5′-AACCCGAACCAAAAAACAAACA-3′ [33] were used . For analyzing the methylation sites of the endogenous mouse Oct4 promoter in NIH3T3 cells , covering sequences from −1 , 148 to −754 upstream of the transcription start site , the forward primer 5′-GTTAGTATAGGAATGGGGGAGG-3′ and reverse primer 5′-CCATAAAACCTACACCCAAACTC-3′ were used; for sequences covering positions −289 to −24 , the forward primer 5′-GGGTGTAGTGTTAATAGGTTTTGTG-3′ and reverse primer 5′-AACCAAATATCCAACCATAAAAAAA-3′ were used . The reaction buffer contained dNTPs and Hotstart Taq polymerase ( Qiagen ) and the samples were incubated at 95°C for 15 min and then 40 cycles of PCR at 94°C for 45 sec , 55°C for 30 sec and 72°C for 1 min , followed by a final extension step at 72°C for 5 min , were performed . The PCR products were purified using QIAquick PCR purification kits ( Qiagen ) and were then ligated into the pCR2 . 1 vector ( Invitrogen ) . 16 to 25 colonies for each sample were sequenced . Any clones with apparent non-CpG methylation ( an indication of incomplete bisulfite conversion ) were excluded from the dataset , and these clones made up less than 2% of all clones sequenced . Global DNA methylation was measured by a LINE1 methylation assay as previously reported [34] . Methylation of LINE1 elements was analyzed by bisulfite conversion of genomic DNA followed by PCR with consensus primers for the LINE1 promoter . PCR was carried out in a reaction buffer containing dNTPs and Hotstart Taq polymerase ( Qiagen ) , and the samples were incubated at 95°C for 15 min and then 42 cycles of PCR at 94°C for 45 sec , 53°C for 1 min and 72°C for 1 min followed by a final extension step at 72°C for 7 min . The following primers were used: forward primer 5′-TTGAGTTGTGGTGGGTTTTATTTAG-3′ and reverse primer 5′-TCATCTCACTAAAAAATACCAAACA-3′ . The PCR products were digested with the HinfI restriction enzyme , which cleaves only methylated DNA after bisulfite conversion . The digested PCR products were separated by electrophoresis on 2% agarose gels . The percentage of methylation was determined after imaging of the gels . The cut bands representing methylated DNA were quantitated using image analysis . The percentages represent the mean with standard deviation for triplicate samples .
|
During mammalian development , genome-wide DNA demethylation occurs both in developing germ cells and in fertilized oocytes . This rapid DNA demethylation is an active process that occurs in the absence of DNA replication . The mechanism of active DNA demethylation represents a conundrum for researchers in this field , i . e . the breakage of a carbon-carbon bond to remove a methyl group from the DNA cytosine ring appears energetically unfavorable , and the elimination of approximately 30 million 5-methylcytosine bases from both DNA strands within a short time window raises questions about the maintenance of genome stability during this process . Recently , it has been reported that the protein GADD45A , a small acidic protein that has been implicated in the DNA damage response , plays a crucial role in promoting active DNA demethylation in several mammalian cell lines . We noticed that GADD45A does not fulfill one likely requirement for a mammalian DNA demethylase factor in that it is not expressed in oocytes or zygotes . We then investigated the role of GADD45A in DNA demethylation using methylated reporter plasmids and DNA methylation analysis of several endogenous genes in cell lines overexpressing GADD45A . Contrary to the previous report , we were not able to demonstrate a role of GADD45A in DNA demethylation . The activity that promotes DNA demethylation at a genome-wide level in mammals remains to be identified .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetics",
"and",
"genomics/epigenetics",
"genetics",
"and",
"genomics"
] |
2008
|
GADD45A Does Not Promote DNA Demethylation
|
For Hepatitis C virus ( HCV ) , initiation of translation is cap-independently mediated by its internal ribosome entry site ( IRES ) . Unlike other IRES-containing viruses that shut off host cap-dependent translation , translation of HCV coexists with that of the host . How HCV IRES-mediated translation is regulated in the infected cells remains unclear . Here , we show that the intracellular level of 40S ribosomal subunit plays a key role in facilitating HCV translation over host translation . In a loss-of-function screen , we identified small subunit ribosomal protein 6 ( RPS6 ) as an indispensable host factor for HCV propagation . Knockdown of RPS6 selectively repressed HCV IRES-mediated translation , but not general translation . Such preferential suppression of HCV translation correlated well with the reduction of the abundance of 40S ribosomal subunit following knockdown of RPS6 or other RPS genes . In contrast , reduction of the amount of ribosomal proteins of the 60S subunit did not produce similar effects . Among the components of general translation machineries , only knockdowns of RPS genes caused inhibitory effects on HCV translation , pointing out the unique role of 40S subunit abundance in HCV translation . This work demonstrates an unconventional notion that the translation initiation of HCV and host possess different susceptibility toward reduction of 40S ribosomal subunit , and provides a model of selective modulation of IRES-mediated translation through manipulating the level of 40S subunit .
Viruses lack translational apparatus , and so they rely exclusively on host machinery for their protein synthesis . Competition for components of the translational machinery between cellular mRNA and viral RNA is therefore inevitable . To gain translational advantage , viruses have evolved various strategies , among which the employment of internal ribosome entry site ( IRES ) -mediated initiation of translation accounts for one [1] . By adopting an initiation mechanism distinct from the predominant cellular cap-dependent initiation , differential regulation of host and viral translation is enabled , and virus translation is thus favored . For example , when cap-dependent translation is selectively repressed during picornavirus ( e . g . , poliovirus and enterovirus ) infection , viral IRES-mediated translation prevails [2] . These viruses encode proteases capable of shutting off host translation by cleaving eukaryotic initiation factor ( eIF ) 4G , whose structural integrity is essential for cap-dependent , but not viral IRES-mediated , initiation of translation [3] . Although hepatitis C virus ( HCV ) also employs IRES-mediated initiation mechanism , no HCV protein has been reported to suppress cap-dependent translation [2] . In addition , cell death often follows the shut-off of host protein synthesis caused by virus infection [4] , and yet HCV establishes chronic infection with little consequence of cytotoxicity . How HCV IRES-mediated translation is regulated in the virus-infected cells remains unclear . HCV IRES is located at the 5′-untranslated region of HCV RNA , and is composed of highly conserved stem-loop secondary structures with specific tertiary folding [5] , [6] . Skipping the requirement for eIFs in the process of directly recruiting 40S ribosomal subunit is one distinct feature of HCV IRES-mediated initiation of translation [7] . Based on the in vitro translation study using cell homogenate supernatant ( Hela S10 ) containing complete set of translation machinery , Otto and Puglisi demonstrated that the formation of binary complex ( HCV IRES and 40S ribosomal subunit ) precedes the formation of 48S-like pre-initiation complex ( HCV IRES , 40S subunit , eIF3 and eIF2 ternary complex ) [8] . This result suggests that HCV IRES directly recruits 40S subunit and subsequently the other factors ( eIF3 , eIF2 ternary complex ) to form 48S-like pre-initiation complex . In contrast to the simplicity of 40S recruitment mediated by HCV IRES , cap-dependent initiation adopts a more sophisticated process , namely , it takes the coordination of various eIFs ( eIF1 , eIF1A , eIF2 , eIF3 , eIF4A , eIF4E , eIF4G ) to sequentially recruit 40S ribosomal subunit to the 5′ end of capped mRNA , and then the 40S ribosomal subunit scans ( energy-dependently ) for the initiation codon , in a 5′ to 3′ direction [9] . The differences in the 40S ribosomal subunit recruitment process between the two distinct modes of translation initiation [7] might provide clues to the mechanism by which HCV differentially regulates host and viral translation . Formation of a stable binary complex consisting of 40S ribosomal subunit and HCV-IRES is the very first step of HCV translation . In reconstitution experiments , the binary complex formation positively correlates with increasing concentrations of purified 40S ribosomal subunit [10] . Individual single point mutations in the IRES that compromise the binary complex forming efficiency invariably led to diminished activities of HCV IRES-mediated translation [10] . As the amount of ribosomes is largely determined by the rate of ribosomal RNA ( rRNA ) transcription , it is notable that HCV viral protein NS5A stimulates the transcription of rRNA by more than 10 folds [11] through hyper-phosphorylation and consequent activation of pol I DNA binding transcription factor , namely , upstream binding factor ( UBF ) . These imply that HCV stimulates ribosome biogenesis , and thus , preferentially viral translation . Here , in a loss-of-function screen , we have identified 40S ribosomal protein 6 ( RPS6 ) as an indispensable host factor for HCV propagation in vivo . Although RPS6 is considered a house-keeping gene , its knockdown , unexpectedly , posed little deleterious effect to the cells , while specifically reduced the HCV RNA level . We found that the abundance of 40S ribosomal subunit was attenuated by RPS6 knockdown , which preferentially suppressed HCV IRES-mediated translation , but left global ( cap-dependent ) translation largely unperturbed . Knockdown of other 40S ribosomal proteins also yielded similar effects . Such selective repression of HCV translation was unique to attenuation of 40S , but not 60S , ribosomal subunit abundance . Furthermore , meta-analysis of a genome-wide screen revealed that , among the components of translation machinery , only knockdown of the 40S ribosomal subunit proteins preferentially suppressed HCV replication . These results thus suggest a distinctive role of 40S ribosomal subunit abundance in facilitating HCV translation and imply a possible role of 40S/60S ribosomal subunit ratio in differential translational regulation .
HCV relies on host factors to complete its life cycle . To identify such cellular factors , we employed a Huh7-cell-based knockdown screening system , in which a modified tri-cistronic HCV replicon ( encoding firefly luciferase ) was used as a reporter ( Figure 1A ) . In this approach , lentivirus-based shRNAs were transduced into cells for specific knockdown of the corresponding genes [12] . We simultaneously measured the luciferase activity ( L ) and cell viability ( M ) ( Figure 1B ) . Clones that exhibited low L/M values , after normalizing against the lacZ shRNA treatment , presumably represented an outcome of reduction of HCV level per cell . Candidate genes were chosen on the basis of three stringent criteria ( Figure 1C ) for further analysis . Out of the ∼1200 human genes screened ( Table S1 ) , RPS6 rose up to the top of the list of candidate genes , as four of the five shRNAs targeting RPS6 caused a significant reduction of the L/M value ( >85% for shRNA1 , 2 , and 3 ) ( Figure 1D ) . To rule out the possibility of false-positive results arising from the tricistronic HCV replicon , which contains several non-HCV elements , the initial candidates were further validated in an infectious HCV system using Jc1 strain [13] . These four shRNAs targeting RPS6 consistently suppressed the HCV RNA level in correlation with the decreased RPS6 mRNA level ( Figure 1E ) . It is interesting to note that shRNA5 inhibited RPS6 mRNA marginally , but caused an increase in HCV RNA level . The reason for this paradoxical effect is not clear . Nevertheless , the good correlation between the RPS6 silencing and HCV Jc1 inhibitory effects at both RNA ( Figure 1E ) and protein level ( Figure 1F ) overall suggests an indispensable role of RPS6 in HCV replication . In addition to RPS6 , we also identified two other hits in the same screen study , polo-like kinase 1 ( PLK1 ) and proline-serine-threonine phosphatase interacting protein 2 ( PSTPIP2 ) . The roles of these two genes in HCV replication were reported recently [14] , [15] . To investigate the mechanism of inhibition of HCV by RPS6 knockdown , we first examined if RPS6 knockdown caused non-specific global effects on the cells . Global RNA degradation is one possible unintended outcome that may result in decreased HCV RNA level . Such a possibility was excluded , because a time-course analysis showed that HCV RNA level was reduced by more than 80% by day 7 post-transduction , and yet the mRNA levels of four randomly selected cellular genes , including PBGD , PKR , AK3 , and APOB , continued to increase ( Figure 2A ) . This result suggests indirectly that RPS6 knockdown specifically reduces HCV RNA synthesis , without causing global RNA degradation . As RPS6 encodes a component of the 40S ribosomal subunit , its knockdown may shut down global protein synthesis and cause cell death , rendering cells unable to support HCV replication . However , we found that cells expressing RPS6 shRNA remained healthy and showed no signs of morphological alteration or deterioration of membrane integrity for at least 12 days ( Figure 2B ) , except for a slower proliferation rate than that of the lacZ shRNA-treated cells ( Figure S1 ) . When cells were treated with cycloheximide ( CHX ) , a global protein synthesis inhibitor , the entire cell population died within 6 days ( Figure 2B ) . In contrast , the cells expressing RPS6 shRNA or lacZ shRNA did not show any sign of cell death for at least 12 days post-transduction . Consistent with published evidence [16] , this result demonstrate that RPS6 knockdown did not cause cell death , at least within the 12-day period . Another possible mechanism of inhibition of HCV replication is by confluence-associated cytostatic effect . Nelson et al . demonstrated that , masked by confluence-associated cytostatic effect , it is the shutoff of de novo nucleotide synthesis that is responsible for the inhibitory effect on HCV replication [17] . Importantly , this published study provided several critical pieces of evidence to prove that cytostatic effect is not the major cause of inhibition of HCV replication . Firstly , addition of nucleotide can recover intracellular HCV level without altering cytostatic status of the cells . Secondly , cytostatic effect alone can not reproduce confluence-associated inhibitory effect on HCV replication , as serum starvation or DNA synthesis inhibitor ( aphidicolin ) perturb cell cycle progression but not intracellular HCV level . These results suggest that cell growth arrest is not sufficient to account for the inhibition of HCV replication . These results together demonstrate that the HCV inhibitory effect is not a consequence of non-specific global effects , and that RPS6 down-regulation can be tolerated by cells without significant deleterious effects . To understand the mechanism of differential effects of RPS6 knockdown on the virus and host cells , we first examined whether 40S and 60S ribosomal subunits and polysomes were differentially affected by RPS6 knockdown . Polysome profile analysis was performed and each ribosome species was quantified by integrating the areas under each peak . The amount of free 40S ribosomal subunit was significantly decreased in the RPS6-knockdown cells as compared to the control lacZ shRNA-transduced cells; in contrast , the amounts of 60S , 80S and polysomes were not affected by RPS6 knockdown ( Figure 3A ) . These results suggest that when RPS6 was knocked down , ribosomes that are actively engaged in translation ( shown as 80S and polysomes ) , stayed relatively unchanged , and only free 40S subunits , which are not involved in translation ( shown as the 40S peak ) , was preferentially affected . In the presence of EDTA , which dissociates ribosomes into 40S and 60S subunits , the total amount of 40S ribosomal subunit was reduced by 50% in the RPS6-knockdown cells . These 40S ribosomes were most likely derived from the 80S ribosomes and polysomes; these results suggest that there were still a substaintial amount of 40S ribosome in the polysomes , even though there was little free 40S ribosome in the RPS6-knockdown cells . In contrast , the 60S ribosome peak remained relatively unchanged in the RPS6-knockdown cells , as compared to that in lacZ shRNA-treated cells ( Figure 3B ) . These results suggested that the knockdown of RPS6 affected primarily the amount of free 40S ribosome while leaving the polysomes essentially intact . These data are highly reproducible . These results also suggest that overall translation in the RPS6 knockdown cells should not be affected . Consistent with the polysome profile analysis , L-[35S]-methionine incorporation study showed that knockdown of RPS6 did not significantly reduce global protein synthesis ( Figure 3C ) . As a control , cycloheximide treatment reduced protein synthesis by more than 80% . These data together demonstrate that despite the reduction of 40S ribosome abundance following RPS6 knockdown , the overall translation in the cells was not significantly affected . Different from the general translation of the cell , the majority of which utilizes cap-dependent initiation of translation , HCV translation relies on its IRES-mediated initiation , where direct binding of eIF-free 40S ribosome to IRES is the crucial first step [8] . To test whether cap-dependent translation and IRES-dependent translation exhibit differential sensitivity toward RPS6 knockdown , we employed a standard bicistronic reporter assay , in which an HCV IRES element was inserted between Renilla and firefly luciferase ORFs ( Figure 4A ) . We found that the relative ratio of the two luciferase activities dropped over time to about 50% after transduction with each one of the two different shRNAs of RPS6 ( Figure 4B ) . In contrast , lacZ shRNAs did not alter the ratio ( Figure 4B ) . These results showed that RPS6 depletion selectively suppressed HCV IRES-mediated translation . To test whether inhibition of HCV IRES-mediated translation is a general property of the reduction of 40S subunit abundance , or a unique property of RPS6 , we examined the effects of knockdown of other small subunit ribosomal protein ( RPS ) genes . The shRNAs targeting RPS9 , RPS15A and RPS20 efficiently reduced target gene mRNA expression level by more than 90% ( Figure 5A ) ; knockdown of these RPS genes specifically caused reduction of the ratio of the 40S/60S ribosomal subunit by 20–50% ( Figures 5B and 5C ) ( Figure S2 and S3 ) . Correspondingly , the relative ratio of the luciferase activity of the dual reporter RNA , representing IRES-dependent vs . cap-dependent translation , also dropped by 20∼40% following the knockdown of these RPS ( Figure 5D ) . Furthermore , there was an 80% reduction of HCV Jc1 RNA level in the cells expressing shRNA targeting these RPS genes ( Figure 5E ) . The finding that HCV RNA level was reduced by a higher extent ( 80% ) than was the translation activity ( 20–50% ) can be explained by the fact that the HCV RNA level reflects the cumulative effects of translation and replication , the latter of which is also affected by the amount of viral polymerase protein . Therefore , through the cumulative effects , the suppression of HCV IRES activity would result in magnified inhibitory effect on HCV replication ( Figure 5 ) . These results demonstrated that individual RPS knockdown caused suppression of HCV IRES-mediated translation , and thus lowered HCV replication efficiency . Also , they suggest that the HCV inhibitory effect is most likely due to the reduction of the amount of 40S ribosomal subunit . To determine if such preferential suppression of HCV is specific to the attenuation of 40S ribosomal subunit or is due to perturbation of the translation machinery in general , we tested the effect of 60S attenuation on HCV replication . Knockdown of large ribosomal protein 6 ( RPL6 ) caused a more than 90% reduction of RPL6 mRNA amount ( Figure 5A ) and a correspondingly reduced amount of 60S ribosome ( Figure 5B ) ( ∼40% increase of the 40S/60S ratio , Figure 5C ) . However , unlike the RPS gene knockdown , the RPL6 knockdown did not alter the dual reporter ratio ( Figure 5D ) ; also , it suppressed HCV RNA level only slightly ( Figure 5E ) . These results suggest that the suppression of HCV IRES-mediated translation is specific to the attenuation of the abundance of 40S subunit . Other than ribosomal subunits , the translation machinery also contains components such as translation initiation factors , which are involved in the regulation of initiation of cap-dependent translation [18] . To determine whether knockdown of these components would result in inhibitory effect on HCV replication , we performed a meta-analysis of a genome-wide screen result based on a published report [19] . This genome-wide screen used siRNA library to monitor the effects of knockdown of cellular genes on the firefly luciferase activity that reflects replication efficiency of a bicistronic subgenomic HCV replicon ( Figure 6A ) . The result of analysis is shown as Z score , which is the number of standard deviations of the experimental luciferase activity above the median plate value in the genome-wide screen . Negative Z scores indicate inhibition of HCV replication . The meta-analysis ( box plot ) ( Figure 6B ) reveals that , among all components of translation machinery , only those that disturb the expression of 40S ribosomal proteins ( RPS , 34 genes ) consistently exhibited significant HCV inhibitory effect ( median Z score around -5 ) . In contrast , knockdown of the 60S ribosomal proteins ( RPL , 55 genes ) , components of eukaryotic translation initiation factors ( eIF1 to eIF5 , 44 genes in total ) , cell cycle regulation-related genes such as cyclins ( CYL , 26 genes ) , cyclin-dependent kinase ( CDK , 18 genes ) and cell division cycle-related genes ( CDC , 36 genes ) , did not preferentially inhibit HCV replication ( median Z score around 0 ) ( Figure 6B ) . These results suggest that the components of 40S ribosomal subunit are distinct from any other components of the translation machinery in playing key roles in HCV replication . It is noteworthy that the result of the knockdown of cell cycle-related genes is consistent with our conclusion ( Figure 2 ) that decrease of cell proliferation rate does not lead to preferential suppression of HCV replication . It has not escaped our notice that the genetic structure of the bicistronic HCV replicon used in the screening of the cellular genes [19] for this meta-analysis might have caused overestimation of the effects on viral replication , because silencing of RPS genes would also directly affect the expression of luciferase reporter , whose translation is driven by HCV IRES . The same reservation applies to the constructs used in our study . Nonetheless , the possibility of overestimation in our case was minimal , since our shRNA treatments were verified by direct biochemical evidence that silencing of small ribosomal protein genes significantly inhibited the replication of infectious HCV Jc1 by more than 70% ( Figure 5E ) . In summary , our results show that RPS6 knockdown attenuates the abundance of 40S ribosomal subunit , which , in turn , preferentially suppresses HCV IRES-mediated translation , and thus inhibits HCV replication without perturbing general translation and cell survival . Importantly , such selective HCV inhibitory effect is specific to the attenuation of 40S , but not that of other components of the translation machinery .
The findings reported here support an unconventional concept that 40S ribosomal subunits can differentially affect different modes of translation . The translation efficiency of HCV , but not that of the cell , is highly correlated with the abundance of 40S ribosomal subunit . Reduction of 40S subunit after knockdown of the various 40S ribosomal proteins ( RPS ) , consistently suppressed the translational activity mediated by HCV IRES ( Figures 5B , 5C and 5D ) , and led to decrease of HCV RNA level ( Figure 5E ) . In contrast , cellular general translation , where cap-dependent initiation predominates , remained largely unperturbed despite the near 50% reduction of total 40S subunit ( Figure 3B ) , as demonstrated by polysome profile analysis ( Figure 3A ) and S[35]-methionine incorporation study ( Figure 3C ) . Furthermore , the reduction of 60S ribosomal protein ( RPL ) ( Figure 5 ) or other translation initiation factors ( eIF ) did not have such an effect ( Figure 6 ) . Taken together , these results showed that the reduction of 40S subunit differentially affected the two modes of translation initiation adopted by HCV and host cell respectively , indicating a unique role of 40S ribosomal subunit in translational regulation . Although the reduction of HCV translation by RPS gene knockdown was only modest ( approximately 20–50% ) , the reduction of HCV RNA level was substantial ( 75% ) ( Figure 5E ) . This is most likely due to the fact that the reduction of viral protein synthesis ( e . g . RNA polymerase ) also led to reduction of viral RNA replication , which determines the amount of positive strand HCV RNA available for translation , thus amplifying the suppressive effects . In addition to the cumulative effect originating from translation-replication coupling , the stability of HCV RNA might be another contributing factor . It is likely that when HCV RNA cannot be efficiently translated or replicated , it is rapidly degraded . This possibility will need to be verified by future experiments . Several possibilities may explain the general increase of cellular mRNA in RPS6 knockdown cells ( Figure 2A ) . One explanation is that the cellular energy originally devoted to ribosome biogenesis , which accounts for ∼80% [20] of cellular energy consumption , has been significantly reduced in RPS6 knockdown cells , and the saved energy was then shifted to other energy-competing cellular processes , such as mRNA transcription . The other possibility is that it might reflect an adaptive feedback to maintain homeostatic gene expression level in response to the slightly decreased translation efficiency caused by the reduction of 40S ribosomal subunit . The cellular mRNA levels were elevated to compensate for the reduction of protein expression level . It has been shown that quantitative reduction of RPS leads to defects at various stages of 18S rRNA maturation [21] , [22] , causing accumulation of precursor rRNA ( 45S ) and reduction of mature 18S rRNA . The reduction of both RPS and 18S rRNA will lead to decreased level of 40S ribosomal subunits . Our study showed that , in RPS knockdown cells , the total amount of 40S ribosomal subunit was reduced by various degrees ( Figures 3B and 5B ) , whereas that of 60S ribosomal subunit remained unaffected . These results suggest that , despite the fact that 40S and 60S subunits share the same rRNA precursor , selective reduction of 40S ribosomal subunit is achievable through reducing its corresponding ribosomal proteins . Knockdown of each one of the RPS proteins studied in this report resulted in the reduction of 40S ribosomal subunit . However , we could not rule out the possibility that these RPS proteins may also affect certain specific functions of ribosome . It has been reported that knockdown of RPS25 does not affect the level of 40S ribosomal subunit , but affects HCV translation [23] . Thus , RPS25 probably affects structure of 40S ribosome; in either case , the abundance and structural integrity of 40S ribosome play key roles in differentiating HCV and host translation . Unlike the traditional view that IRES-containing viruses are particularly sensitive to the inhibition of general translation [24] , our findings indicate that , for HCV , the effect of 40S attenuation is different from that of 60S attenuation or perturbations of translation initiation factors , and suggest a distinction between HCV and other IRES-containing viruses . The key difference between HCV IRES-mediated translation and cap-dependent or other viral IRES-mediated translation lies in the steps of 40S subunit recruitment that use different forms of 40S subunit [7] . Based on the in vitro reconstitution assays , it has been shown that eIFs are not required for the initial binding of 40S subunits to the HCV IRES [8] , [10] , whereas most eIFs are utilized in those of cap-dependent and encephalomyocarditis virus ( EMCV ) IRES-mediated translation [7] . For the latter cases , 43S pre-initiation complex ( consisting of 40S subunit , eIF1 , eIF2 , eIF3 , initiator tRNA and GTP [18] ) is involved in 40S ribosomal subunit recruitment , and the eIFs ( eIF4A , eIF4B ) are indispensible in facilitating the migration of 40S subunit to the translation initiation sites . However , the in vivo evidence for this model is still not available . These differences may explain the differential sensitivity of HCV translation to the amounts of 40S ribosome . In RPS6 knockdown cells , the near 50% reduction of total 40S ribosomal subunit ( Figure 3B ) , affected only the amount of free 40S subunit ( the 40S that are not engaged in translation ) but not that of 60S , 80S and polysomes ( Figure 3A ) . It is likely that 40S ribosomal subunits are recycled from polysomes immediately after translation termination . Recycling of ribosome after translation termination is facilitated by the binding of eIF1 and eIF3 to 40S ribosomal subunit [25]; once eIF2-GTP-Met-tRNAMet joins , this form of 40S subunit readily forms 43S pre-initiaition complex [9] , and thus supports cap-dependent translation . Therefore , when the levels of 40S ribosomal subunits were partially depleted after RPS6 knockdown ( Figure 3B ) , cap-dependent translation was maintained ( Figure 3C ) , presumably by the recycled ribosome , while that of HCV IRES was preferentially suppressed ( Figure 4B ) , very likely due to the shortage of eIF-free 40S subunit . In all the polysome profiling experiments presented here ( Figures 3A , 3B and Figure 5B ) , each treatment was performed using 10 ( 6-well- ) plates , which were pooled together for polysome analysis . The boundaries used for quantifying peaks are shown in Figure S4 . Our results are consistent with published evidence [16] , [21] , [22] demonstrating that silencing of ribosomal protein reduces the abundance of the corresponding ribosomal subunit . Together with published data , we propose a model of how the translational advantage of HCV varies in response to the rises and falls of 40S subunit abundance , as illustrated in Figure 7 . The amount of free 40S ribosomal subunit fluctuates in response to the total amount of 40S ribosomal subunit , whose biogenesis is boosted by HCV infection [11] and attenuated by RPS knockdown . Thus , during HCV infection , ribosome synthesis increases [11]; as a result , the level of free 40S subunit increases and so does that of the eIF-free 40S subunit , which then facilitates HCV IRES-mediated translation . On the other hand , as RPS knockdown blocks the maturation of nascent 40S ribosomal subunits , the reduction of total 40S subunit then selectively reduces availability of free 40S ribosomal subunit . Accordingly , 40S subunit recycled from translation termination sustains the formation of 43S pre-initiation complex [25] but not that of eIF-free 40S; as a result , cap-dependent translation remains unaffected , resulting in preferential suppression of HCV IRES-mediated translation . The conclusion that the availability of the 40S ribosomal subunit is crucial for HCV proliferation is compatible with the result that depletion of RPL6 increased translation of the HCV IRES-driven firefly luciferase in the bicistronic reporter ( Figure 5D ) . Since the ratio of 40S∶60S subunits was increased ( Figure 5C ) in cells depleted of RPL6 , more free 40S ribosomal subunit would become available for forming a complex with the HCV RNA and thus compete with cellular mRNAs for available 60S ribosomal subunits . Interestingly , the increase in HCV translation nonetheless suppressed the HCV RNA levels in Jc1-infected cells to 70% of the control . According to published data [22] , silencing of RPS only had slight effects on cellular polysomes , whereas silencing of RPL significantly decreased total amounts of polysomes . Since general translation is affected substantially by knockdown of RPL , it is conceivable that , in RPL knockdown cells , the decrease of overall translation activity ultimately will negatively affect the production of HCV viral proteins , despite the elevated 40S/60S ratio favoring HCV IRES-mediated translation . There are growing lines of evidence showing that ribosomal protein may be highly regulated to exert specific translational control in gene expression [26] , which leads to a provocative model of “ribosome code” [27] . Consistent with this concept , our data provide a model that alteration in ribosomal subunit 40S/60S ratio can differentially affect cap-dependent and HCV IRES-mediated translation , pointing out a possible mechanism that ribosomal proteins exert specific control through fine tuning the ratio of 40S/60S ribosomal subunit . In this loss-of-function screen study , our tricistronic replicon was originally designed for probing host factors involved in HCV replication in general , not limited to the translation of HCV ( here the HCV IRES drives translation of the neo gene ( Figure 1A ) , which is irrelevant in the time frame of the screen experiment ) . Nonetheless , this system can still identify essential factors required by HCV IRES-mediated translation for two main reasons . One is that all these viral IRES elements that drive the expression of reporter and HCV polyprotein respectively may share the same cellular factors [24] . The candidates that are not caused by the effects on HCV IRES would be excluded by the confirmation study using infectious HCV . The other reason is that the 5′UTR region of HCV harbors not only IRES crucial for translation , but also the sequence crucial for genome replication [28]; therefore , due to tight coupling of translation and replication of IRES-containing virus [29] , [30] , [31] , including HCV [32] , many host factors , such as La autoantigen [33] , [34] , PTB [35] , and PCBP2 [36] , are involved in both functions . RPS6 was selected under such a condition . Current therapy against HCV infection relies mainly on interferon ( IFN ) -based regimens [37] , based on the anti-viral defenses activated by IFN signaling [38] . However , HCV has evolved various strategies to evade host defense , such as blocking IFN-mediated antiviral defense by perturbing the production and signaling of IFN [39] . Moreover , some HCV genotypes are intrinsically resistant to IFN [40] . Besides , the current standard therapy is only effective for <50% of genotype-1- and ∼80% of genotypes 2 and 3-infected patients [41] . Although novel therapies developed to target HCV viral proteins are promising , they may eventually be compromised by rapid emergence of resistance-associated mutations of HCV [42] . Here , our findings provide a new anti-HCV strategy by manipulating host factors , namely , the level of 40S ribosomal subunit , to act against the translational advantage of HCV . This strategy not only inhibits viral propagation effectively , but also provides a solution to current mutation-associated drug resistance problems [43] .
Cells ( Huh-7 . 5 , HCV-tricistronic replicon cell , and 293T ) were cultured in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% fetal bovine serum , nonessential amino acids , 100 units/ml penicillin , and 100 µg/ml streptomycin at 37°C in a CO2 ( 5% ) incubator . To ensure cell line consistency , no more than 15 passages of subculture were employed . The HCV tricistronic replicon used in this study was modified from the HCV-1b-neo45 replicon construct [44] by inserting an EV71-IRES-driven firefly luciferase ORF between Neo ORF and the EMCV IRES . The tricistronic replicon RNA was produced by in vitro transcription , and electroporated into Huh-7 cells ( ECM630 Electroporator , BTX Harvard Apparatus; 975 µF and 220 V ) and selected with G418 . The colony with the highest luciferase activity and the highest HCV protein level was chosen and maintained in G418 ( 0 . 5 mg/ml ) -containing medium for subsequent experiments . All plasmids for lentivirus production were provided by the National RNAi Core Facility , Academia Sinica , Taiwan . For lentiviral production , pCMV-ΔR8 . 91 ( for packaging ) , pMD . G ( for envelope proteins ) , and an individual shRNA construct were transfected into 293 T cells using Trans-IT ( Mirus Bio ) ( see Protocol S1 ) . The viral titer was determined in Huh-7 cells by using a cell viability assay ( i . e . , the Relative Infectious Unit [RIU] method ) ( see Protocol S2 ) . The HCV-EV71-Luc-replicon-containing cells were plated in 96-well plates ( 1×104 cells per well ) 24 h prior to transduction . Cells were transduced with lentiviruses at a multiplicity of infection ( MOI ) of ∼3 in the presence of Polybrene ( hexadimethrine bromide; 8 µg/ml ) by spin infection ( 1 , 100×g , 15 min , 37°C ) and incubated at 37°C for 24 h . The cells were incubated with fresh medium for another 24-h incubation and finally incubated in the media containing puromycin ( 2 µg/ml ) . Luciferase activity ( L ) and cell viability ( M ) were measured 5 days after the lentiviral transduction . L was determined by using a Bright-Glo Luciferase Assay System ( Promega ) and M was measured by using an MTS assay ( CellTiter 96 Aqueous Non-Radioactive Cell Proliferation Assay , Promega ) . The relative L/M ratios , which were normalized against the L/M value of lacZ-shRNA-transduced control cells , were used to evaluate the knockdown effects on HCV replication . All lentivirus-based shRNA knockdown experiments were done at MOI = 2 to ensure high transduction efficiency and uniform utilization of the microRNA processing machinery in the cells . Transduction and puromycin selection were done as described above . For the shRNA sequences used in this study , please refer to Table S2 . The pJc1 plasmid , which contains a chimera genome of HCV J6CF/JFH1 , was constructed as previously described [13] . Full-length JC1 genomic RNA was produced by in vitro transcription of pJc1 , and electroporated into Huh7 . 5 cells , and incubated for 24 hrs . Jc1 viral particles were then collected from the cell culture supernatant at 3 day after transfection for further expansion . Viral particles were titered by counting the number of infected cell colonies by immunostaining HCV core protein as described [45] . For studying the effects of shRNA on Jc1 infection , Huh7 . 5 cells were first transduced with lentivirus-based shRNA for 2 h and then incubated with Jc1 virus suspension ( MOI = 1 ) for 1 h at 37°C . The intracellular virus RNA was determined at various time points after infection . Total RNAs were extracted from cells by using RNeasy Mini Kit ( Qiagen ) and converted into cDNAs using SuperScript III First-Strand Synthesis System ( Invitrogen ) . The primers for reverse transcription are oligo ( dT ) 20 and an HCV-specific primer ( 5′-CACTCGCAAGCACCCTATCA-3′ ) . For real-time PCR analysis , we followed the standard TaqMan strategy using the Universal Probe Library and the LightCycler 480 Real-Time PCR System ( Roche Diagnostics ) ( Table S3 ) . Each quantitative PCR reaction was performed in duplicates . Data were normalized against the quantity of PBGD , which served as an internal control . Cells were stained based on membrane integrity and intracellular esterase activity by using LIVE/DEAD Viability/Cytotoxicity Kit for mammalian cells ( Invitrogen ) . Cells seeded in 12-well plates were washed with 1× PBS two times to remove serum esterase in the medium , and followed by incubation with the staining reagent ( 2 µM Calcein AM , 4 µM Ethidium homodimer in 1× PBS ) at room temperature for 30 min . Images of stained cells were acquired by using a Zeiss inverted fluorescent microscope system ( Axio Observer A1 ) . Cell lysate was prepared by mixing 2 . 5×106 cells with 1-ml RNase-free lysis buffer ( 0 . 5% Nonidet P-40 , 500 U/ml RNAse inhibitor ( Invitrogen ) , 1 mM PMSF , 20 mM DTT , 150 µg/ml cycloheximide ) on ice and centrifuged at 12 , 000×g for 5 min at 4°C to remove nuclei . The RNA concentration in the supernatant was then determined . Linear sucrose gradient ( 11 ml ) was prepared by mixing equal volume of 10% and 60% sucrose stock solutions ( in 300 mM KCl , 5 mM MgCl2 , 10 mM HEPES [pH 7 . 4] ) using Gradient Master ( Bio-Comp ) . An equivalent of 300-µg RNA of lysate was loaded onto a sucrose gradient , which was then centrifuged in a SW41 rotor ( Beckman ) at 21 , 000×g for 2 h at 4°C . We used the Density Gradient Fractionation System ( ISCO ) to fractionate the gradients at a flow rate of 0 . 75 ml/min with the UV-detector sensitivity set at 1 . 0 . To dissociate ribosomes into subunits for similar analysis , EDTA was included in the lysis buffer ( 100 mM ) and sucrose gradient buffer ( 25 mM ) . For quantitative analyses of polysome profile , the area of integration of each peak was defined as the area enclosed by the trace lines and the boundary lines . The boundary line is defined by two points at which the slope of the trace line exhibits the biggest difference with that of the adjacent point ( See Figure S3 and S4 ) . Paper boards were cut along the contour of these integration areas and weighed; the weights of these paper boards were proportional to the sizes of the integration areas and used for quantitative analyses . Lentivirus-transduced cells ( 1×106 ) were first incubated in methionine-free medium for 1 h at 37°C . The medium was then replaced with the same media containing 0 . 2 mCi of L-[35S]-methionine ( >1000 Ci/mmol; NEG709A005MC , PerkinElmer ) in 1-ml volume and further incubated for 15 min . Cells were then washed with PBS and resuspended in 100 µl of BSA ( 1 mg/ml ) /0 . 02% NaN3 ( 0 . 02% [w/v] ) . Half of the cell suspension was then mixed with 1 ml of 10% trichloroacetic acid ( TCA ) . TCA precipitates were collected onto a 25 mm glass microfiber filter paper ( Whatman GF/C ) , and the [35S] radioacitivity was measured in a scintillation counter . The TCA-precipitable counts were normalized against the counts of the other halves of the samples that were not precipitated by TCA . Each experiment was done in triplicates . The bicistronic dual reporter construct was modified from psiCHECK2 ( Promega ) , which harbors a Renilla luciferase ORF and a firefly luciferase ORF under the control of SV40 and HSV-TK promoters , respectively . The psiCHECK2 was first digested with PmeI and ApaI ( 30 bp downstream of the firefly luciferase gene ) for replacing the region between the two luciferase genes with a DNA fragment containing HCV-IRES ( genotype-1 ) . This HCV-IRES-containing fragment , which was obtained from PCR , has the following features ( from 5′ to 3′ ) : PmeI site , HCV-IRES sequence ( i . e . , HCV 5′UTR and coding sequence for the first 12 amino acids of the core protein ) , an alanine codon , coding sequence for the first 10 amino acids of firefly luciferase , and ApaI site . Transduction of Huh7 . 5 cells was done as described above , except that cells were plated in 96-well plates ( 5 , 000 cells per well ) . Transfection of the bicistronic dual reporter plasmid DNA ( 100 ng per well ) was performed at post-transduction day 3 using Trans-IT LT1 transfection reagent ( Mirus Bio ) . Transfected cells were harvested 20 h after transfection for luciferase activity assay using Dual-Glo Luciferase Assay System ( Promega ) . Data of the published functional genomic screen [19] were retrieved in Excel format and key-word sorted into the following categories: small subunit ribosomal proteins , large subunit ribosomal proteins , translation initiation factors , cyclin , cyclin-dependent kinases , and cell division cycle-related genes ( see Dataset S1 ) . The genomic screen employed a bicistronic subgenomic HCV replicon , in which the first cistron encodes a fused ORF containing a firefly-luciferase gene and a neomycin-resistance gene . Firefly luciferase activities therefore reflect the abundance of the bicistronic subgenomic HCV replicon . The retrieved Z score is the number of standard deviations of the experimental luciferase activity above the median plate value . Negative Z scores indicate inhibition of HCV replication . The distribution of the Z scores in each category was then analyzed and displayed as box plot using the NCSS 2007 software , with each box representing the interquartile range within which the middle 50% of the ranked data were found .
|
Hepatitis C virus ( HCV ) infection causes chronic liver diseases that threaten ∼2% of the world population . There is no effective vaccine , and the current standard therapy , the combination of interferon and ribavirin , is effective to less than 50% of genotype-1 infected patients . While antivirals targeting at specific HCV proteins might ultimately lose their effectiveness due to the emergence of resistance-associated mutations , an alternative strategy that explores the genetic stability of host factors indispensable for HCV replication may provide better therapeutic targets for anti-HCV medicine . Here , we employed a loss-of-function screen method to identify such potential targets and uncovered a potential novel anti-HCV mechanism by manipulating the biogenesis of 40S ribosomal subunit . We showed that inhibiting 40S ribosome biogenesis can selectively suppress HCV translation and thus effectively inhibit HCV replication . In contrast to the conventional thinking , the 40S ribosomal subunit can differentially affect different translational modes , and HCV translation is more sensitive to the amounts of 40S ribosomal subunit as compared to general translation in host cell . Since HCV is known to evade anti-viral effects including translational suppression elicited by interferon , our findings may help design a therapeutic strategy to supplement interferon-based therapy and minimize mutation-associated drug resistance problem .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"rna",
"interference",
"microbiology",
"internal",
"ribosome",
"entry",
"site",
"biology",
"viral",
"replication",
"molecular",
"biology",
"drug",
"discovery",
"biochemistry",
"rna",
"nucleic",
"acids",
"virology",
"antivirals",
"molecular",
"cell",
"biology"
] |
2012
|
Attenuation of 40S Ribosomal Subunit Abundance Differentially Affects Host and HCV Translation and Suppresses HCV Replication
|
Paroxysmal nocturnal hemoglobinuria ( PNH ) is an acquired clonal blood disorder characterized by hemolysis and a high risk of thrombosis , that is due to a deficiency in several cell surface proteins that prevent complement activation . Its origin has been traced to a somatic mutation in the PIG-A gene within hematopoietic stem cells ( HSC ) . However , to date the question of how this mutant clone expands in size to contribute significantly to hematopoiesis remains under debate . One hypothesis posits the existence of a selective advantage of PIG-A mutated cells due to an immune mediated attack on normal HSC , but the evidence supporting this hypothesis is inconclusive . An alternative ( and simpler ) explanation attributes clonal expansion to neutral drift , in which case selection neither favours nor inhibits expansion of PIG-A mutated HSC . Here we examine the implications of the neutral drift model by numerically evolving a Markov chain for the probabilities of all possible outcomes , and investigate the possible occurrence and evolution , within this framework , of multiple independently arising clones within the HSC pool . Predictions of the model agree well with the known incidence of the disease and average age at diagnosis . Notwithstanding the slight difference in clonal expansion rates between our results and those reported in the literature , our model results lead to a relative stability of clone size when averaging multiple cases , in accord with what has been observed in human trials . The probability of a patient harbouring a second clone in the HSC pool was found to be extremely low ( ~10-8 ) . Thus our results suggest that in clinical cases of PNH where two independent clones of mutant cells are observed , only one of those is likely to have originated in the HSC pool .
Paroxysmal nocturnal hemoglobinuria ( PNH ) is an acquired disorder of hematopoietic stem cells ( HSC ) due to a somatic mutation in the PIG-A gene [1 , 2] . Loss of function or hypofunction mutations in this gene result in loss of or reduced ability to synthesize the glycosylphosphatidylinositol ( GPI ) anchor . As a consequence , many cell surface proteins that need this anchor to attach to the plasma membrane are no longer available or only available in reduced numbers on the cell [3 , 4] . Some of these proteins such as CD55 and CD59 are essential for the protection of red blood cells from complement mediated lysis . As a consequence , erythrocytes that lack CD55 and CD59 undergo intravascular hemolysis leading to anemia , hemoglobinuria , iron deficiency and fatigue . Scavenging of nitric oxide by free plasma hemoglobin results in endothelial and platelet dysfunction leading to the high risk of venous and arterial thrombosis associated with this disease . Additional symptoms related to nitric oxide depletion include abdominal pain , esophageal pain , chronic kidney disease and erectile dysfunction . PNH is a rare condition and many practitioners ( mostly those working outside of large hospital facilities ) have yet to encounter a single patient with this disease . Although the discovery of somatic mutations in the PIG-A gene ( which resides on the X chromosome ) provided a very elegant explanation of how an acquired mutation in a single gene could lead to the disease phenotype [1] , this does not explain how the mutant clone expands . It has been shown that ( i ) the mutation rate in PIG-A deficient cells is normal [5] , ( ii ) PIG-A deficient cells are not more resistant to apoptosis than normal cells [6] , ( iii ) the replication rate of mutated cells is normal [7] and PIG-A deficient cells do not have a proliferative advantage over normal cells [8] . Perhaps it was natural for investigators in the field to assume from the outset that there must be a selective fitness advantage of PIG-A mutated cells and consequently to investigate possible causes for some benefit that enables clonal expansion . It has been proposed that the selective advantage of mutated cells is extrinsic to them and due to an immune mediated attack on normal cells [9] . Some evidence in support of this hypothesis exists [10–13] , but this hypothesis is unable to explain the following observations: ( i ) PIG-A is ubiquitously expressed in the body–why should the immune attack presumably against the GPI anchor be restricted to the normal HSC population ? ( ii ) Immunosuppressive therapy does not lead to the elimination of the mutant cells and expansion of the normal HSC with return to normal hematopoiesis . ( iii ) A significant fraction of patients with PNH undergo ‘spontaneous’ extinction of the clone [14] . A second hypothesis has been proposed , which postulates that additional mutations in one or more genes that confer a fitness advantage to the PIG-A mutant cells may occur . Indeed , several case reports are available including two patients with a mutation in HMGA2 [15] , one patient with a concomitant JAK2V617F mutation [16] , a mutant N-RAS [17] and more recently a patient with PNH and concomitant BCR-ABL in the same cell population was also reported [18] . However , these cases appear to be the exception and not the rule , and their description in our opinion requires specific explanations other than the one responsible for the general origin: clonal expansion and possible elimination of PNH phenotype clones . We have previously shown that given the normal mutation rate in PNH cells , it would be rare for a patient to have a second mutation in a PIG-A mutated stem cell that would provide the fitness advantage necessary for clonal expansion [19 , 20] . Moreover , to date there is no evidence of a fitness advantage for the PIG-A mutated cells even though recent data suggests that some could have additional mutations ( that are typically seen in myelodysplastic syndrome or leukemia ) present [8 , 21] . Finally , deep sequencing of patients with PNH clones reveals that a significant fraction do not have additional mutations present apart from that in PIG-A [21] . Given these observations , we proposed that perhaps clonal expansion in PNH is simply due to neutral drift , given i ) that cells with GPI deficiency do not seem to have any evolutionary advantage over normal cells and ii ) the limited data in support of any benefit with immunosuppressive therapy ( in the absence of concomitant aplastic anemia ) [22] . Interestingly , in some patients with PNH more than one clone is detected–one with complete deficiency of GPI anchored proteins ( so called PNH III cells ) and another with partial deficiency of GPI anchored proteins ( PNH II cells ) [3 , 23] and this has been confirmed by sequencing [21] ( for historical reasons , normal cells are referred to as PNH I cells ) . Clearly , these two mutant populations arise due to independent mutations in the PIG-A gene in different cells leading to these phenotypes . In this work , we use evolutionary principles and stochastic dynamics modelling of the HSC pool to determine the incidence of the disease in populations , estimate clone size and average age at diagnosis , and also address the question of multiple PIG-A mutated clones in patients with this disease . These results strengthen and extend the ones found by Dingli et al . [22] , who first tested this hypothesis , by the use of a Markov chain formulation rather than traditional simulations . We provide exact solutions for the probabilities in the state space and for the first time , an estimate of the rate of clonal expansion in this disease .
Depending on the total number of mutated stem cells in an individual we can assign different diagnoses: when at least 20% of the total stem cell pool has a PIG-A mutation the individual is defined as having clinical PNH . Cases where the PIG-A clone ( s ) consist of <20% of the cell population are considered to be subclinical ( or latent ) PNH [24] . The probability for an individual to develop clinical PNH increases with age according to the curves shown in Fig 1A , although this risk actually decreases briefly during the period of ontogenic growth due to the corresponding growth of the stem cell pool [25] . Indeed , the limited data available suggests that PNH is quite rare in children [26 , 27] and generally occurs in the context of bone marrow failure . While classical hemolytic PNH represents about 10% of pediatric patients with a PIG-A mutant population , data from the International PNH registry suggests that perhaps half of adults with PNH have classical hemolytic disease [28] . We find that the probability of a patient having clinical PNH with two independent clones arising from the HSC pool is approximately 103 times smaller than the same probability of diagnosis with a single clone ( Fig 1A ) . Furthermore , we estimate that patients who have 3 or more distinct PNH clones contributing to hematopoiesis occur with a probability that is another 2 orders of magnitude lower ( Fig 1A ) . This implies that approximately only 1 in 1000 cases of clinical PNH would host more than a single mutant clone that arose in the stem cell compartment . Note that these numbers result from a model dealing only with stem cell dynamics . Thus , this does not preclude the occurrence of mutations farther downstream among progenitor cells ( which are present in larger numbers than HSC and also divide faster [19 , 29] ) . Moreover , PIG-A mutations occurring in early progenitors will also remain contributing to hematopoiesis for years before any eventual wash-out [30 , 31] . Thus , divergent PIG-A mutations found in mature cells are more likely to have originated at later stages of differentiation [19] than to originate in independent mutations occurring in the active stem cell population . Using population age distribution data obtained in 2010 by the United States Census Bureau , we estimate the prevalence of clinical PNH ( weighted sum of census data and clonal existence probabilities ) for both mono- and multiclonal cases in the USA ( Fig 1B and 1C ) . We calculate an expected prevalence of 1 . 76 cases per 105 citizens for any diagnosis of clinical PNH ( mono- or multiclonal ) , which is similar to what has been reported in a well-defined population by Hill et al . [32] . The expected number of patients with biclonal disease arising at the level of the HSC is determined at 1 . 29 per 108 individuals . For the US population , this would amount to approximately 3000 patients with a single clone and 2 patients with biclonal disease , respectively . The number of individuals in the population with a subclinical ( <20% ) PIG-A mutated clone is estimated to be much higher , at 6 . 0 per 104 for monoclonal and 1 . 9 per 107 for biclonal cases , which amounts to respectively 184 , 495 and 60 individuals in the US . The first mutated cell in the HSC pool can occur quite early in an individual’s life , as shown in Fig 2A . The probability of harboring a mutant cell in the stem cell population grows one order of magnitude from age 20 ( ~2×10-3 ) to age 100 ( ~2×10-2 ) . Though these values may seem quite high , it is important to note that in the neutral drift hypothesis , the second line of defense against PNH is the significant low likelihood of clonal expansion , a fact that is illustrated well by comparing the probability of occurrence of a clone ( which is quite common in healthy people [33] ) with the probability of having clinical PNH . For example , in an individual of age 60 , the probability of having acquired a mutant clone is 1×10-2 , while the probability of having clinical PNH is 2×10-5 , three orders of magnitude smaller . The average ages of clonal occurrence are projected at 41 and 54 years for mono- and biclonal ( stem cell ) cases respectively ( Fig 2B ) . In general , it appears that , on average , most clones arrive only after adulthood is reached and the hematopoietic stem cell pool has reached its maximal size . The average age at diagnosis–in our model we take this as the time at which the total number of mutated HSCs reaches 20%–is found to be 49 years , and is quite similar to what has been reported from the International PNH registry [28 , 34] . Because some investigators define clinical PNH at a lower threshold , especially in the presence of aplastic anemia , we also calculated the average age when 10% of the HSC pool is composed of PIG-A mutant cells , and obtained a mean arrival time of 44 years . As mentioned above , the lack of a selective advantage makes it difficult for the mutated clone to expand , since at each replication event it is equally likely to decrease in size as it is to expand ( if one neglects the low probability of mutation ) [35–37] . Over time the size probability distribution widens , adding more emphasis on larger clones while smaller clones become less probable ( since they are more likely to go extinct ) . Thus , in cases where two separate clones are simultaneously present , the first that occurred is likely to be larger and therefore less likely to resolve than the second . From a mathematical perspective , this behavior can be ascribed to the fact that the all-normal state ( absence of mutants ) and all-mutant state ( complete takeover by mutants—fixation ) are ( in the absence of mutations ) absorbing states of the evolutionary dynamics . An important consequence of the all-normal absorbing state is that most clones which arise in a population go extinct before reaching a significant size . We find that approximately 83% of all clones that appeared in our in-silico population resolved , and in most of these cases clonal extinction would have occurred soon after the clone’s arrival , so that the individuals at stake would never have been diagnosed with PNH as their clone would have been very small . On the other hand , extrapolating these simulations to the entire hematopoietic tree clearly suggests that the massive cell turnover ( with mutation ) that occurs normally in hematopoiesis explains why finding a PIG-A mutant cell population with sensitive sequencing or flow cytometry in a healthy individual is not unexpected [33] . If a clone does manage to increase in size , the likelihood of spontaneous clonal extinction becomes less pronounced . In Fig 3A we show the probability distribution of clonal size measured at 3 different times after disease diagnosis . The variance of this distribution increases not only over time–as the clone has more time to expand or diminish–but also as the clone increases in size due to the frequency dependence of this “random walk” . In particular , the closer the clone size comes to comprising 50% of the SC pool , the larger this variance will be . One consequence of this behavior is that the distributions shown in Fig 3A are skewed to the right . Note , however , that despite the changing shape of the size distribution , the mean clone size ( the average of this distribution ) does not change over time . This result implies that in a cohort of diagnosed patients , we expect the average of their clone sizes to remain stable despite individual expansions or recessions . Using the census data , the average clone size m in individuals in the US population with at least one mutant HSC is estimated in our model to be at 3 . 4% of the total pool . However , the average clone size in those suffering from clinical PNH ( m≥20% ) is much larger , found to be 31 . 1% , or 19 . 6% if this threshold is instead taken at m≥10% . Of course , in individual patients , clone size can be quite high and be the predominant contributor to hematopoiesis . Whenever the number of mutant stem cells reaches the threshold of clinical PNH , it is nevertheless possible for the disease to disappear due to the stochastic nature of the clonal dynamics . We calculated the probability that a recently diagnosed case of clinical PNH ( assuming the SC pool is ≈20% mutated at diagnosis ) becomes subclinical again . The result ( Fig 3B ) shows that over time it is more likely for the disease to recede than to persist , although it would take at least 2–10 years for significantly smaller clone sizes ( <15% or <10% ) to be reached . The probability of the clone becoming truly extinct only becomes realistic after 20–50 years , and in reality clinically detectable extinction will depend on the assay that is used to determine the presence or absence of the clone . It should be evident that more sensitive flow cytometry based assays will be able to detect the clone , even when the ‘old’ standard Ham’s test becomes negative . Note that these results only represent the likelihood of disease reduction under neutral dynamics , and as such do not exclude the possibility of a more fit clone arising which can advantageously compete with the PIG-A clone and lead to disease loss [38] . We compared our predictions of average clone size and patient age at diagnosis with data from the International PNH Registry [28] . Our predicted mean clone size of 31 . 1% ( standard deviation: 32 . 6 ) and mean age at diagnosis of 48 . 6 ( standard deviation 52 . 8 ) seem to fit nicely with the registry data for patients with AA-PNH syndrome ( categorized as also suffering from aplastic anaemia ) which shows a mean clone size of 28 . 3% ( standard deviation 32 . 8% ) and 43 . 2% of patients diagnosed between 30 and 59 years of age . However , other categories presented much greater average clone sizes ( though similar ages at diagnosis ) . Our prediction of the number of patients with clinical PNH in the general US population is slightly lower than what Hill et al [36] reported . However , we note that our modeling takes into account the age structure of the population in the United States , whereas the patients observed by Hill et al . were from Great Britain , which may have a different age distribution . Furthermore , our prevalence estimates depend on a strict definition of clinical PNH ( clone size >20% ) , whereas in clinical practice , the transition from subclinical to clinical disease may be less abrupt . We also compared our findings on clonal expansion rate with measurements from Araten et al . [7] , who reported a ≥5% size increase per year in 12 out of 36 patients . Most of the other patients experienced either a reduction or no change at all , though the authors did not specify these amounts quantitatively . The study found no significant expansion or reduction ( ≈0% ) when calculating the mean over all patients , which nicely fits our neutral drift model . Our model projected the fraction of patients that would experience a ≥5% increase after 1 year to be between 5% and 10% depending on the size of the initial clone , which is lower than their observed 33% . This discrepancy could be due to several factors including the relatively small size of their patient cohort and the fact that our model does not include the contribution of progenitor compartments to the overall size of the PNH population . If the progenitor cell population is contributing significantly to hematopoiesis , then more fluctuations in clone size would be expected due to the shorter lifetime of these cells [30] . Our modeling predicts that after 10 years from diagnosis , the probability that the clone is small enough not to be associated with clinical PNH is upwards of 30% ( Fig 3B ) . This is also comparable to what Hillmen et al [14] reported in their cohort of patients who survived for more than 10 years from diagnosis ( 12 out of 35 patients ) .
The appearance of mutations in HSC and their fate over time is an important clinical problem , since many diseases such as myelodysplastic syndromes and several leukemias ( e . g . chronic myeloid leukemia , some subtypes of acute myeloid leukemia ) arise due to mutations within the HSC . Landmark studies in PNH have shown that it is an acquired clonal HSC disorder [39] with very interesting dynamic properties , including an uncanny probability of spontaneous clonal extinction [14] . The mechanism of clonal expansion in PNH has been a source of great debate and several hypotheses have been proposed to explain it , such as a selective advantage of the mutant cells due to an immune attack on normal HSC ( extrinsic advantage ) , or the presence of a second mutation that grants a fitness advantage ( intrinsic advantage ) . Some evidence for either hypothesis exists , but both also suffer from deficiencies as described in the Introduction . In particular , immunosuppressive therapy does not return hematopoiesis to normal and there is no reduction in the size of the PIG-A mutant clone once the presumed selective advantage is eliminated . It is also difficult to see how a cell can acquire multiple mutations sequentially in the absence of genomic instability , which has not been observed in PNH [5] . We have proposed that the PIG-A mutant cells generally possess no fitness advantage ( or disadvantage ) , and that clonal expansion is simply a consequence of neutral drift within the ( small ) active HSC pool that maintains hematopoiesis [22] . This hypothesis leads to the simplest of explanations of PNH , and our stochastic modeling suggests that this may be the case–at least in some patients–since we are able to predict the incidence and prevalence of the disease , average age at diagnosis , average clone size and the probability of clonal extinction purely from first principles with results quite similar to what has been reported in the literature . Although it is difficult to deliver conclusive proof of our hypothesis , the close parallel between our predictions and clinical reality provides considerable support for it . It has been reported that in at least one patient , that PNH clonal extinction was concomitant with the appearance of a new population of cells that harbored mutations in genes such as STK36 that can potentially provide a fitness advantage to the non-PIG-A mutant cells [38] . Our model makes no assumptions about the possibility of more fit clones arising , but merely gives the likelihood of the disease disappearing entirely on its own through neutral drift . Moreover , while the ongoing dynamics within the normal hematopoietic stem cell group put this population at continued risk of accumulating new mutations–some of which could lead to a fitness advantage as proposed by Babushok et al [38]–the presence of such mutations by itself does not imply that the PNH clone resolved due to takeover of hematopoiesis by a new population with mutations that are often found in patients with myeloid disorders . As discussed elsewhere [40] the presence of such mutations by itself does not necessarily mean that the mutant cells with a normal PIG-A gene have a fitness advantage since the mutant gene ( e . g . STK36 ) may or may not be a ‘driver’ mutation . This point is further highlighted by the fact that in the case reported by Bubashok et al [38] , PNH clonal reduction occurred over the course of 12 years while our estimate for the general population under neutral drift is that on average , 8–10 years are required for the PNH clone to reach ~15% . Neutral drift may come as a surprise for many in the field of hematology and oncology who are accustomed to associate malignant clonal expansions in cancer with some form of selective advantage . Nevertheless , it is not uncommon for mutations in populations to expand by neutral drift , as suggested by Kimura many years ago [41] . In fact , the recent cancer genomics data explosion suggests that most mutations found in tumours are examples of neutral drift and are labeled passenger mutations , since they provide no fitness advantage to the tumour [42] . Of course , mutations that increase the fitness of malignant clones clearly exist . Perhaps the main difference in PNH is that neutral drift could be the main mechanism of clonal expansion . Not only is it the simplest explanation , but it also provides a very elegant correlation of genotype with phenotype , a feat that is much more difficult to achieve in malignant tumours due to the complex mutational landscape that they harbor . In conclusion , we present a stochastic model of HSC dynamics which studies mutations in the PIG-A gene that leads to the PNH phenotype . Our predictions based on an in silico model of Markovian stochastic dynamics enable us to determine from first principles the incidence of the disease in a population , the average clone size and the probability of clonal extinction , with results similar to what is observed in clinical practice . We also find that in a neutral drift model the probability of multiple PNH clones arising separately in the HSC pool is exceptionally small , a result which suggests that in clinical cases where differing clones are observed , all but one of the clones are likely to have emerged in later stages of differentiation . We propose that PNH is perhaps the first disease where neutral drift alone may be responsible for clonal expansion leading to a clinical problem .
The model describes a system of non-interacting HSC that undergo cell division and differentiation . The size NSC of this stem cell pool changes during ontogenic growth from about 20 cells at birth , to approximately 400 cells by adulthood [43] , following the growth curve determined by Dingli et al [25 , 44] . A stochastic evolution of the system is modelled in discrete-time steps , where at each time step a division and subsequent differentiation event are performed ( so-called birth-death process ) . This is done by randomly selecting for replication a single cell from the stem cell pool , which generates two daughter cells . Since we consider only neutral dynamics , all cells possess the same fitness , meaning each cell ( whether it is normal or PIG-A mutated ) has the same probability of being selected . Following replication , a new cell is randomly selected from the total population ( now of size NSC+1 ) for differentiation , removing it from the stem cell pool . Thus NSC remains unchanged and this neutral evolution model can be considered a special case of the well-known Moran process [45] . When a normal cell replicates there is a probability μ that one of the daughter cells acquires a mutation , introducing a potential seed for the development of a mutated clone in the population . The cells in the active HSC compartment , replicate slowly , at a rate of approximately once per year [46] . The conversion between number of cell replications and biological time is done in the following way: When the number of replication events equals the cell population size , then one year has passed . Thus , in adulthood , if ≈400 replications within the HSC pool occurred , a year would have passed . The increase of NSC during ontogenic growth follows an predetermined growth curve derived before [25 , 44] and is implemented by including additional divisions without successive differentiations at fixed times ( 2 week intervals ) to match the expected growth rate . Thus , while the Moran-like dynamics are respected for all division-differentiation events , they are periodically interrupted during this phase to model the increase of the population . While this model of neutral dynamics has already been introduced and studied by Dingli et al . [22] , the current approach ( described below ) allows for a more robust probing of the system from which new insights can be obtained . The discrete time-evolution of this system is well described by a Markov chain in which the state space is represented by the number of mutants present in the population . To predict the stochastic evolution we numerically evolve the master equation Pm[x+1]=∑kpm , k × Pk[x] where Pmx is the probability of finding the HSC cell population in state m ( with m being the number of mutated cells in the population ) at time step x , and pm , k is the probability to go from state k to state m in a single time step . Starting from a mutant free population leads to the initial conditions P00 = 1 ( the probability of 0 mutants existing at time t = 0 is 1 ) and Pm>00 = 0 ( the probability of more than 0 mutants existing at time t = 0 is 0 ) . The transition probabilities are found by considering all changes that may occur when performing a division and differentiation event . For example , an event in which a normal cell divides without acquiring a mutation and a normal cell differentiates results in a transition into the same state ( one normal cell is added and one is removed ) , and occurs with a probability 1-mNSC1-μ1-mNSC+1 . The total probability of staying in the same state ( i . e . the transition m→m ) is found by summing this term with all other events that result in a same state transition; in particular , if a mutant cell is selected in both division and differentiation steps–probability mNSCm+1NSC+1 , and if a normal cell divides but acquires a mutation and a mutant cell differentiates–probability 1-mNSCμm+1NSC+1 . In a similar manner we can obtain all nonzero elements of the transition matrix to obtain: {pm , m−1=m−1NSC ( 1−mNSC+1 ) + ( 1−m−1NSC ) μ ( 1−mNSC+1 ) pm , m=mNSCm+1NSC+1+ ( 1−mNSC ) μm+1NSC+1+ ( 1−mNSC ) ( 1−μ ) ( 1−mNSC+1 ) pm , m+1= ( 1−m+1NSC ) ( 1−μ ) m+1NSC+1 Note that for the “division-only” events occurring sporadically during ontogenic growth a simpler transition matrix is used in which no differentiation takes place ( see S1 Text ) . It is clear that only transitions between “nearest-neighbours” in the state space are possible , so that pm , k = 0 if m-k>1 . An interesting property of this system can be seen from that fact that if we neglect the possibility of a new clone arising ( that is , neglecting the possibility of mutation so that μ = 0 ) , the transition probabilities to move up or down from a state m in a single time step are identical: pm+1 , m=pm−1 , m=NSCm−m2NSC ( NSC+1 ) This symmetry implies the system will perform a random walk reminiscent of Brownian motion , the main difference being that the probability to move away ( up or down ) from a current state is not independent of m , instead adhering to the quadratic function given above , with a maximum value at m = NSC/2 , reflecting the frequency dependence of the evolutionary dynamics . This means that the closer the mutant population gets to this maximum , the more “volatile” it becomes . It is also worth noting that while the above transition matrix holds–as mentioned earlier–only in the absence of fitness differences between cell types , it can easily be extended to cases where one type is more likely to divide or differentiate . The probability of a cell selected for division ( or differentiation ) being a mutant then not only depends on the size m of the mutant clone as m/NHSC , but also on the clone’s relative fitness rPIGA [47] P{dividing cell∈mutants}=rPIGAmrPIGAm+rhealthy ( NHSC−m ) Where rhealthy is the fitness factor of the non-mutated HSC . It is also immediately clear that taking rPIGA = rhealthy reduces the expression to the selection-free case treated in this work . The entire set of transition probabilities in the presence of selection are given in the appendix . Evolving the basic Markov chain described above provides no method for tracking the evolution of multiple clones that arise from separate mutational events , as only a single PIG-A mutant population is considered . Thus , in order to distinguish clonally unrelated subpopulations , we extend the model by expanding the state space to account for multiple mutations . In the following , we take advantage of previous estimates which show that one can safely ignore two mutations in the same stem cell [19 , 20] to limit our state space to account for a maximum of two different clones . This considerably simplifies the associated computations , as the state space scales as ~Ni , with N being the cell population size and i the maximum number of independent clones . As a result , the state space is divided into three separate histories , as shown in Fig 1 , corresponding to states with different pasts . For example , an evolutionary history in which a mutation occurred but the resulting clone eventually died should correspond to a different state than one where no mutation occurred in the first place . Thus , the master equation is now altered to include transitions to different histories: Pm1 , m2i[x+1]=∑m1' , m2' , jpm1 , m2 , m1' , m2'i , j× Pm1' , m2'j[x] where any state is now characterized by the clone sizes m1 and m2 , and the appropriate history i . The tensor elements pm1 , m2 , m1' , m2'i , j represent transitions from state m1' , m2' to ( m1 , m2 ) and history j to i , and are found in an identical manner as before , though now more transitions are possible ( Fig 4 ) . The probabilities obtained by evolving the master equation also allow us to calculate the average clone size m in individuals of a given population whose clone size is within a chosen range . For example , in “clinical PNH” patients ( defined as having a clone which involves at least 20% of the HSC pool ) we are only interested in individuals whose number of PIG-A mutated cells ranges between 20% and 100% . The expression to evaluate is then: m=∑y=1100∑m=m0NSCm wy , mqy∑y=1100∑m=m0NSCwy , mqy , where wy , m is the probability of an individual of age y having a clone of size m ( wy , m = ∑i , nPm , niy which results from evolving the Markov chain ) , and qy is the fraction of individuals of age y in the population based on the 2010 US census [United States Census Bureau . Single Years of Age and Sex: 2010 . https://www . census . gov . Accessed 13 June 2017 . ] , where we take individuals up to 100 years of age . The terms in the numerator form the average when all possible sizes 0 , … , NSC are considered , while the denominator normalizes the result if we are interested in a particular range , e . g . m0→Nsc ( it is easy to see that when m0 = 0 the denominator becomes 1 ) .
|
The mechanisms leading to expansion of HSC with mutations in the PIG-A gene that leads to the PNH phenotype remains unclear . Data so far suggests there is no intrinsic fitness advantage of the mutant cells compared to normal cells . Assuming neutral drift within the HSC compartment , we determined from first principles the incidence of the disease in a population , the average clone size in patients , the probability of clonal extinction , the likelihood of several separate clones coexisting in the HSC pool , and the expected expansion rate of a mutant clone . Our results are similar to what is observed in clinical practice . We also find that in such a model the probability of multiple PNH clones arising independently in the HSC pool is exceptionally small . This suggests that in clinical cases where more than one distinct clone is observed , all but one of the clones are likely to have emerged in cells that are downstream of the HSC population . We propose that PNH is perhaps the first disease where neutral drift alone may be responsible for clonal expansion leading to a clinical problem .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"united",
"states",
"medicine",
"and",
"health",
"sciences",
"markov",
"models",
"cell",
"cycle",
"and",
"cell",
"division",
"population",
"genetics",
"cell",
"processes",
"geographical",
"locations",
"cloning",
"gene",
"pool",
"cell",
"differentiation",
"north",
"america",
"physiological",
"processes",
"developmental",
"biology",
"mathematics",
"stem",
"cells",
"molecular",
"biology",
"techniques",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"molecular",
"biology",
"probability",
"theory",
"people",
"and",
"places",
"hematopoietic",
"stem",
"cells",
"cell",
"biology",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"evolutionary",
"biology",
"hematopoiesis"
] |
2018
|
Evolutionary dynamics of paroxysmal nocturnal hemoglobinuria
|
Plants have varying abilities to tolerate chilling ( low but not freezing temperatures ) , and it is largely unknown how plants such as Arabidopsis thaliana achieve chilling tolerance . Here , we describe a genome-wide screen for genes important for chilling tolerance by their putative knockout mutants in Arabidopsis thaliana . Out of 11 , 000 T-DNA insertion mutant lines representing half of the genome , 54 lines associated with disruption of 49 genes had a drastic chilling sensitive phenotype . Sixteen of these genes encode proteins with chloroplast localization , suggesting a critical role of chloroplast function in chilling tolerance . Study of one of these proteins RBD1 with an RNA binding domain further reveals the importance of chloroplast translation in chilling tolerance . RBD1 is expressed in the green tissues and is localized in the chloroplast nucleoid . It binds directly to 23S rRNA and the binding is stronger under chilling than at normal growth temperatures . The rbd1 mutants are defective in generating mature 23S rRNAs and deficient in chloroplast protein synthesis especially under chilling conditions . Together , our study identifies RBD1 as a regulator of 23S rRNA processing and reveals the importance of chloroplast function especially protein translation in chilling tolerance .
Low temperature inhibits plant growth in general and limits the geographical distribution of plants . Earlier studies have identified numerous physiological and cellular changes associated with chilling ( more than 0°C ) or freezing ( less than 0°C ) conditions , such as alterations in membrane composition , calcium signals , metabolite composition , photosynthesis , and protective molecules [1 , 2] . Most of these changes are thought to help plants to cope with low temperature stresses . Plants differ in their abilities to tolerate chilling stresses . Low temperature often inhibits photosynthesis and reduces carbon uptake and allocation to developing sink tissues [3 , 4] . Many tropical and subtropical plants including maize , rice , and tomato do not survive at 4°C because they cannot undergo photosynthesis and carbon metabolism under this condition . Arabidopsis , as well as some overwinter cereals , can grow at the low temperature due to their biochemical and physiological adaptations which may include acclimation of photosynthetic metabolism [5 , 6] . Translation in chloroplast appears to be especially sensitive to chilling stresses . Chilling slows down protein biosynthesis in plastids by eliciting frequent ribosome pausing in tomato [7] . The otherwise chilling tolerant Arabidopsis plants become chilling sensitive when they are defective in chloroplast ribosomal biogenesis and RNA processing [8–12] . For instances , loss of the translation elongation factor SVR3 , the rRNA maturation factor NUS1 , and chloroplast RNA binding proteins CP29A and CP31A , all lead to increased chilling sensitivity through affecting chloroplast biogenesis [9 , 10 , 13] . In addition , the loss of chloroplast ribosomal subunits reduces the ability of plants to recover from prolonged chilling periods [11 , 12] . Chloroplast function is carried out by genes coded mostly by the chloroplast genome [14] . Transcription of such genes relies on two plastid RNA polymerases: nucleus-encoded RNA polymerase ( NEP ) and plastid-encoded RNA polymerase ( PEP ) [15 , 16] . Chloroplast RNAs need to be processed to become functional rRNAs and mRNAs . Many of the processing factors for RNA cleavage , splicing , editing or stability are RNA-binding proteins [13 , 17–19] . They are all coded by the nuclear genome . One family has pentatricopeptide repeats ( PPR ) and it usually carries out specific RNA processing in chloroplasts [19] . Another family contains RNA recognition motif/RNA-binding domain/ribonucleoprotein ( RRM/RBD/RNP ) domain and these proteins , referred to as RNPs , are suspected to regulate larger sets of RNAs [20] . Among the chloroplast RNPs , CP31A and CP29A are associated with a large pool of chloroplast transcripts and influence their stability , processing , and splicing [13] . While chilling tolerance mechanism is not well understood , cold acclimation , an enhancement of freezing tolerance by a prior exposure to low non-freezing temperature , has been intensively studied in Arabidopsis thaliana [21–25] . Cold acclimation is mediated in part by C-repeat binding factors ( CBFs ) which regulate the expression of a large number of Cold Responsive ( COR ) genes , some of which are thought to confer freezing tolerance . The upregulation of the CBF genes by low temperature is critical for this acclimation and is mainly modulated by ICE1 ( Inducer of CBF expression 1 ) , ICE2 and the three closely related CAMTAs ( Calmodulin binding Transcription Activators ) [26–29] . Although genes regulated by CBF proteins play vital roles in freezing tolerance , they only represent a small percentage of the COR genes . CBF independent regulations of cold acclimation or freezing tolerance have been found in Arabidopsis [30 , 31] . It is yet to be determined whether or not chilling tolerance and cold acclimation have shared mechanisms . To have a better understanding of chilling tolerance in Arabidopsis , we carried out a chilling sensitive mutant screen on all available T-DNA insertion mutants that represent half of the total Arabidopsis genes . Interestingly , mutants defective in chloroplast localized proteins are overrepresented , indicating the importance of chloroplast function in chilling tolerance . Detailed characterization of one such mutant of a RNA binding protein supports a critical role of chloroplast translation in chilling tolerance .
In order to identify genes important for chilling tolerance in Arabidopsis at the genome scale , we analyzed 11 , 000 T-DNA insertion mutants covering half of the total Arabidopsis genes for chilling sensitive phenotypes . Each of them is putative homozygous T-DNA knockout or knockdown mutants in indexed genes generated by the SALK Institute [32] and available from the Arabidopsis Biological Resource Center ( http://abrc . osu . edu/ ) . The use of these indexed knockout or knockdown lines over chemical mutagenized population enabled the assessment and verification of phenotypes under different growth conditions and allowed uncovering of conditional lethal mutants and avoiding temperature-sensitive alleles of essential genes . In this screen ( S1 Fig ) , four mutant seeds of each line were germinated and grown side by side with the wild-type Col-0 on vertical plates under a 16 hours ( h ) light/day photoperiod first at 22°C for 8 days and then transferred to 4°C for two months . Lines showing abnormal growth phenotypes compared to the wild type only at 4°C but not 22°C were selected as chilling sensitive mutants . These phenotypes include albino , yellow or purple leaves , abnormal leaf shapes , smaller leaves , and shorter roots . Because not all of these T-DNA lines were homozygous for the T-DNA insertion mutations , we further analyzed those lines where not all four seedlings exhibited mutant phenotypes to exclude those whose mutant phenotypes are not correlated with the T-DNA insertion . From this screen , a total of 54 lines showed growth defects at 4°C but not 22°C and we defined these mutants as chilling sensitive ( Fig 1A ) . We then examined the genes indexed to be disrupted by T-DNA insertions in these lines . Those with insertion in the promoter region but not exon , intron , or UTRs were removed as they might not affect the function of the indexed gene and the phenotype is unlikely due to a defect in the gene . This leaves us with 49 chilling sensitive mutant lines where the function or expression of the indexed genes are disrupted . None of the 49 genes are in the characterized CBF cold acclimation pathway , and loss of function mutants of CBF1 , CBF2 , CBF3 , ICE1 , HOS1 and SIZ1 were not in the collection . Although causal genes for chilling sensitivity may not be the indexed gene in a small proportion of the lines , we decide to analyze functional categories of these candidate genes as a group to reveal potential important chilling tolerance mechanisms . Among proteins encoded by the 49 candidate genes , 16 were annotated as localized to the chloroplast ( Table 1 ) , 10 to the nucleus , 6 to the cytosol , 4 to the mitochondria , and the rest either to other locations or without localization information . There is thus a high representation of chloroplast related genes in these mutants , suggesting a critical role of chloroplast function in chilling tolerance . Among the 16 proteins with chloroplast localization , 4 were chloroplast RNP proteins and three were previously characterized: ORRM1 ( AT3G20930 ) , CP29A ( AT3G53460 ) and CP31A ( AT4G24770 ) . Mutants of CP29A and CP31A were chilling sensitive and defective in processing various RNAs in chloroplast [13] . ORMM1 is required for plastid RNA editing in Arabidopsis and maize [33] , but its role in chilling tolerance has not been analyzed . The other RRM/RBD/RNP coding gene , AT1G70200 , was not characterized before , and we named it RBD1 . Mutants of all of these 4 genes showed bleaching in newly emerging leaves at 4°C . A mutant of another RNP coding gene AML1 ( AT5G61960 ) also had a bleaching phenotype . AML1 is predicted to localize in the nucleus but it may also have a chloroplast localization signal from the e-FP data . It plays a role in meiosis as well as in vegetative growth in Arabidopsis thaliana [34] , but its role in chilling tolerance was not analyzed . We therefore further analyzed chilling sensitive phenotypes of mutants of these 5 RNP coding genes . When grown on soil at the normal temperature 22°C , the rbd1 , cp29a and cp31a mutants exhibited a wild-type phenotype , while orrm1 and aml1 had a smaller size than the wild type Col-0 ( Fig 1B ) . The aml1 had yellow cotyledons at germination but recovered in a few days , and it often had white strips on the leaves . Consistent with visual phenotypes , the rbd1 , cp29a and cp31a mutants has a fresh weight similar to that of the wild type at 22°C , but orrm1 and aml1 mutants displayed a 35% and 58% reduction compared to the wild type ( Fig 1C ) . When shifted to 4°C after grown at 22°C for three weeks , new leaves emerged from all of these mutants were yellow ( Fig 1B ) . When these mutants were grown constantly at 4°C in soil from germination , all their leaves were yellow and the plants were much smaller than the wild type with only 10% to 35% of the wild-type fresh weight ( Fig 1C ) . We chose RBD1 for further analysis because there were no prior reports on this gene . The RBD1 gene encodes a protein of 538 amino acids ( aa ) containing one RNA recognition motif/ RNA-binding domain ( Fig 2A ) . We confirmed that the chilling sensitive phenotype observed in the mutant line SALK_041100 is due to the loss of the RBD1 function through analysis of additional RBD1 mutants and complementation test . The SALK_041100 line ( named rbd-1 ) has the T-DNA inserted in the third exon of the RBD1 gene , while the T-DNA insertion line SALK_012657 ( named rbd1-2 ) has an insertion in the 5’-UTR of RBD1 ( Fig 2B ) . Both mutants were loss-of-function mutants of RBD1 as no full length transcripts could be amplified by Reverse Transcription ( RT ) -PCR ( Fig 2C ) . The rbd1-2 mutant , like rbd1-1 , produced yellow emerging leaves at 4°C ( Fig 2D ) . In addition , RNA interference ( RNAi ) was used to reduce the expression of RBD1 , and 13 of 15 RNAi transgenic lines produced yellow leaves under chilling conditions ( Fig 2D ) . Furthermore , when the full-length RBD1 cDNA driven by the cauliflower mosaic virus ( CaMV ) 35S promoter was transformed into the rbd1 mutants , the chilling sensitive phenotype was suppressed in 30 of 32 the transgenic lines ( Fig 2D ) . Therefore , the loss of RBD1 function does confer chilling sensitivity . We characterized the growth phenotypes of the rbd1-1 mutants in more detail . When plants were shifted from three weeks’ growth at 22°C to 4°C , the yellowing or bleaching became visible only after 2 weeks of chilling treatment and only happened in leaves emerged at 4°C . There is a gradient of yellowing decreasing from the youngest leaf to the oldest leaves ( Fig 2D ) . This yellowing phenotype is reversible , when plants were moved from 4°C back to 22°C , the pale green phenotype disappeared in 2 days ( Fig 2E ) . The rbd1-2 mutant also had a lighter green appearance compared to the wild type when grown constantly at 22°C . Using spectrophotometer , we found that the two rbd1 mutants had a 10–20% reduction of chlorophyll a , chlorophyll b and carotenoids compared to the wild-type Col-0 . This phenotype was not observed in the complemented lines ( Fig 2F ) . Therefore , the rbd1 mutants are compromised in chloroplast function especially under chilling conditions . Because the rbd1 mutants displayed chilling sensitivity , we analyzed whether chilling induction of the CBF and COR genes was altered in the rbd1 mutants . CBF1 , CBF2 , and CBF3 are induced by 4°C treatment at 6 hours to the same extent in Col-0 and rbd1-1 ( Fig 3A ) . Similarly , these genes fall to the basal level after 3 weeks of 4°C treatment in both the wild type and the mutant ( Fig 3A ) . COR15A , COR47 , and KIN1 genes are regulated by the CBF genes , and they were induced by 4°C treatment at 6 hours and stayed elevated during the subsequent chilling treatment for four weeks ( Fig 3B ) . The rbd1-1 mutant had the same induction kinetics and amplitudes of these three COR genes by 4°C treatment ( Fig 3B ) . Therefore , the rbd1-1 mutant is not impaired in its capacity to induce the CBF or COR expression by 4°C and the chilling sensitivity of the mutants is likely independent of the CBF pathway . The RBD1 gene is expressed in green tissues . Using RT-PCR analysis , RBD1 mRNA was detectable in leaves , stems , flowers and siliques but not in roots ( Fig 4A ) . The expression pattern of RBD1 was further investigated in transgenic lines containing a ß-glucuronidase ( GUS ) reporter gene fused to the RBD1 promoter . GUS expression was observed in all green tissues throughout development ( Fig 4B ) . In 3-day-old seedlings , GUS staining was primarily detected in emerging cotyledons ( Fig 4B ) . In 14-day-old seedlings , it was detected in most of the plant but not in roots , and was particularly strong in cotyledons ( Fig 4B ) . GUS staining was also observed in the stems and siliques , but not in the matured seeds ( Fig 4B ) . Within the flower , it was strong in the sepals and carpels , but not in the stamens and petals ( Fig 4B ) . In addition , older green tissues usually had higher expression levels than younger green tissues , suggesting that RBD1 might contribute to the function of mature green tissues . Altogether , these expression patterns show that RBD1 is highly expressed in the photosynthetic tissues , which was in agreement with its putative roles in chloroplast function . RBD1 is not induced by light according to the expression data from Arabidopsis eFP Browser ( http://bbc . botany . utoronto . ca ) . It is not induced by chilling treatment either . RT-PCR revealed that RBD1 is expressed at the same level during the chilling treatment from day 0 to day 21 ( Fig 3C ) . The RBD1 gene therefore might be constitutively expressed at the transcript level . RBD1 is annotated as a chloroplast targeted protein , which could explain its expression in green tissues . To verify the subcellular localization of the RBD1 protein , we expressed GFP ( Green fluorescent Protein ) fusion of RBD1 under the 35S promoter ( RBD1:GFP ) in protoplasts isolated from wild-type Arabidopsis leaves and monitored fluorescence by confocal microscopy . While the GFP protein alone was present in the cytoplasm and nucleus , the RBD1:GFP fusion protein was found exclusively in chloroplasts ( Fig 4C ) . In addition , the RBD1:GFP fusion protein was dispersed as small fluorescent particles within chloroplast ( Fig 4C ) , which is reminiscent of nucleoid localization [35] . This localization pattern does not appear to be temperature-dependent . In both 22°C and 4°C , RBD1:GFP is only present in the nucleoid structure in chloroplast ( S2 Fig ) . The nucleoid localization pattern and the yellowish leaf mutant phenotype promoted us to investigate the role of RBD1 gene in chloroplast RNA processing because nucleoid is the site of rRNA processing and ribosome assembly [36] . We first analyzed total RNAs separated on formaldehyde denaturing agarose gel stained by ethidium bromide . Plastid rRNAs ( 23S , 16S , 5S , and 4 . 5S ) and cytosolic rRNAs ( 25S , 18S , 5 . 8S , and 5S ) can be easily observed on such gels because they are very abundant [37] . When the same amount of total RNAs was loaded on gel , rbd1-1 and rbd1-2 mutants showed significant reduced accumulation of the 1 . 1 and 1 . 3-kb species of the 23S rRNA compared to the wild type under chilling stress conditions , but exhibited almost the same amount of rRNAs as wild type at 22°C ( Fig 5A ) . RNA gel blot hybridization was used to further analyze the defects in 23S rRNA processing . The 23S rRNA with a full length of 2 . 9-kb is cleaved internally at two sites , yielding fragments of 0 . 5 , 1 . 1 , and 1 . 3-kb in wild type [38] . With a probe detecting the 3’-end fragments , we found that 4°C treatment did not affect the processing of 23S rRNA transcripts in the wild type ( Fig 5B ) . In contrast , the chilling treated rbd1 mutants accumulated the partially processed 2 . 9-kb and the 2 . 4-kb of 23S rRNA at a much higher level than the wild type , while the fully processed product of 1 . 1-kb rRNA was greatly reduced ( Fig 5B ) . Consistent with a defect in processing of 23S rRNA which is a component of the chloroplast ribosome , the rbd1 mutants had a reduction in chloroplast translated protein . When total leaf proteins were analyzed on a SDS polyacrylamide gel , the chloroplast Rubisco Large Subunit ( RbcL ) protein , which is synthesized in chloroplast , was found to have a drastic reduction in rbd1-1 and rbd1-2 mutants compared to the wild type by 4°C treatment but not at 22°C ( Fig 5C ) . We further analyzed the chloroplast translation efficiency in the rbd1 mutants . Spectinomycin is a chemical inhibitor of chloroplast translation as it prevents translocation of the peptidyl-tRNA from the A site to the P site on the 30S subunit of 70S ribosomes [39] . When seeds were sowed and grown on 1/2 MS supplied with spectinomycin at a concentration of 3mg/L at 22°C , both the rbd1-1 and rbd1-2 mutants exhibited total yellowing while the wild type and the complemented rbd1 mutants maintained some green tissues ( Fig 5D ) . This suggests that translation in chloroplast is compromised in the rbd1 mutants even under non-chilling condition . We further analyzed the transcripts of 9 chloroplast RNAs in the rbd1 mutants under normal and chilling conditions to assess how broad a role the RBD1 gene might play in chloroplast RNA regulation . They are 16S rRNA , PEP transcribed ndhF , psaA , rbcL , psbB , psbF and petB , NEP transcribed ycf3 , and PEP and NEP transcribed rps4 . Among these , 16S rRNA , ndhF , psbB , petB , ycf3 and rps4 need to be processed to become mature transcripts . In contrast to that of 23S rRNA , the processing of 16S rRNA was only slightly affected in the rbd1 mutants compare to Col-0 at 4°C but not 22°C ( S3B Fig ) . The transcripts for ndhF , psaA and rbcL showed decreased levels under chilling stress compared to the wild-type plant , but displayed similar levels to wild type at 22°C ( S3A Fig ) . For psbB , psbF and petB , low temperature down regulated their transcripts levels , but the transcripts were at similar levels in the rbd1 mutants and wild type at 22°C and 4°C ( S3B Fig ) . The transcripts of ycf3-ex2 and rps4 showed an over-accumulation under chilling condition in the rbd1 mutants compared to the wild type , but exhibited the same levels at 22°C ( S3C Fig ) . In all , the loss of RBD1 mainly alters processing of 23S rRNAs but not other chloroplast RNAs we analyzed . It also reduces the expression level of some PEP transcribed genes , but increases the expression of some NEP transcribed genes at chilling temperature . The leaf yellowing phenotype in the rbd1 mutants likely results from an accumulative defect over time because it only became visible two weeks after plants were shifted from 22°C to 4°C . To identify earlier defects in the mutants , we monitored the molecular events after the plants were shifted from 22°C to 4°C at 0 hour ( h ) , 6 h , 1 day ( d ) , 3 d , 7 d , 14d , 21 d and 28 d . Among the 6 genes that showed altered processing and expression in the mutants , 23S rRNA was the first to show defects . Its processed 1 . 1-kb transcript was significantly lower than the wild type after 7 days of 4°C treatment ( S4A Fig ) . The transcript of psaA showed defects in the rbd1 mutants after 2 weeks of cold treatment ( S4B Fig ) . The transcripts of the rest ndhF , rbcL , ycf3-ex2 and rps4 genes began to show significant change after 3 or 4 weeks of chilling treatment ( S4 Fig ) . Therefore , the rbd1 mutants exhibit a defect in 23S rRNA processing very early on ( S4B , S4C and S4D Fig ) , which would lead to reduced chloroplast translation and bleaching phenotype under chilling conditions . To determine whether or not RBD1 protein is associated with the 23S rRNA and the association is temperature regulated , we performed RNA co-immunoprecipitation ( IP ) assay at both 22°C and 4°C . RBD1 was fused to the GFP ( RBD1:GFP ) and expressed under the strong 35S promoter in Arabidopsis protoplasts . As a control , the signal peptide of ORRM1 that targets the protein to chloroplast was fused with GFP ( sORRM1:GFP ) for protoplast expression as well . After transformation , the protoplasts were incubated at 22°C for 8 h before being split for further incubation at 22°C and 4°C respectively for 12 h to assess temperature dependency of binding . Total proteins from protoplasts were then IPed by the anti-GFP antibodies , and the co-IPed RNAs ( after DNase I treatment ) were reverse transcribed and detected by quantitative Real Time-PCR . We analyzed four transcripts: 23S rRNA , 16S rRNA , psbF , and rbcL and found binding only to 23S rRNA ( Fig 6 ) . Three regions of the 23S rRNA precursor were analyzed: 5’ end , middle part , and 3’ end , residing in three different processed RNAs . There was an 8 , 11 and 6 fold increase of PCR products for the three regions of 23S rRNA precursor in the RBD1:GFP sample compared to the sORRM:GFP sample at 22°C ( Fig 6A ) . Furthermore , the fold increase for the three regions were 23 , 39 and 19 fold in the RBD1:GFP sample compared to the sORRM:GFP at 4°C ( Fig 6B ) . Therefore , RBD1 binds to three regions of 23S rRNA precursor , and the binding is at least two times more at 4°C than at 22°C . In contrast , no difference was detected for the 16S rRNA , psbF and rbcL between RBD1:GFP sand sORRM:GFP at either 22°C or 4°C ( Fig 6A and 6B ) , suggesting a specificity of transcript binding for RBD1 . We further compared the molecular defects among the 5 RNP mutants after 4 weeks of 4°C treatment . The rbd1 mutants had the most drastic defect in 23S rRNA processing among the 5 RNP mutants . By RNA gel staining , reduction of processed 23S rRNAs compared to the wild type was the most pronounced in the rbd1-1 mutant ( Fig 7A ) . By RNA blotting , the processed 1 . 1-kb transcript was found to decrease in all of these 5 RNP mutants , with the most severe reduction happening in rbd1-1 . The partially processed 2 . 9- and 2 . 4-kb transcripts were over-accumulated in these mutants at 4°C except for orrm1 ( Fig 7B ) . Among the 5 RNP mutants , the rbd1-1 mutant exhibited the highest ratio of partially processed transcripts ( 2 . 4- and 2 . 9-kb ) over processed transcripts ( 1 . 1-kb ) . Consistent with the observed 23S rRNAs processing defect in the 5 RNP mutants , the accumulation of RbcL protein at 4°C was reduced to different extent , with an approximately 20% to 40% reduction compared to the wild type ( Fig 7C ) . In contrast to the RbcL protein , TOC75 , a cytosol synthesize chloroplast outer membrane protein [40–42] , showed an increased amount of 20% to 40% at 4°C in the 5 RNP mutants compared to the wild type ( Fig 7D ) . In addition , rbd1-1 showed the highest sensitivity to spectinomycin followed by aml1 among the 5 RNP mutants assayed ( Fig 7E ) . At 3mg/L concentration , all rbd1-1 seedlings turned white and the aml1 mutant was pale yellow , while the other three mutants stayed green similarly to the wild type . These results indicate that loss of RBD1 has a larger effect on 23S rRNA processing compared to other mutants , which may impact chloroplast translation more than the others . It also shows that the ORRM1 mutation has the least impact on rRNA processing .
In this genome-wide chilling sensitivity screen , we identified 54 T-DNA insertion lines that exhibited a chilling sensitive phenotype in Arabidopsis . Although we have not tested if all results from disruption of the gene that the T-DNA is inserted in , we expect that most of them are . Interestingly , a large proportion of these genes are related to chloroplast function , indicating that it is one of the weakest links in chilling tolerance . Earlier studies have implicated a strong association of compromised chloroplast function with chilling tolerance , and this study supports this notion at the genome wide scale . Wild-type Arabidopsis plants usually survive at low temperatures such as 4°C but mutants with compromised chloroplast function are sensitive to chilling similarly to the subtropical and tropical plants . To what extent difference in chilling tolerance among different plants is due to difference in chloroplast function at low temperatures is worth studying in the future . This screen also suggests that chilling tolerance and cold acclimation may not use identical mechanisms in Arabidopsis . Mutants of the known key regulators of the CBF pathway , which are unfortunately not present in the collection , are not known or reported to have chilling sensitive phenotypes . Mutants identified in the chilling sensitive screen here were not isolated as defective in the CBF pathways mutants . Indeed , the rbd1-1 mutant is not defective in cold induction of the CBF pathway , suggesting that the induction of the CBF pathway may not require optimal chloroplast function and chilling sensitivity can result from CBF independent defect . Nevertheless , chilling tolerance can be connected with the CBF pathway . In Arabidopsis , a chilling sensitive mutant crlk1 ( calmodulin receptor like kinase 1 ) is chilling sensitive and is also delayed in CBF induction [43] . In rice , a chilling sensitive mutant cold1 is defective in CBF induction [44] . It is possible that chilling sensitivity can result from defects in multiple processes and some might be shared with the CBF pathway . The chloroplast localized proteins critical for chilling tolerance are involved in multiple aspects of chloroplast function , including transcript regulation and maturation , chloroplast development , protein transportation and secretion , as well as photosynthesis . For instance , YS1 is required for editing of rpoB transcripts and chloroplast development during early growth [45] . PQL3 is required for NDH activity in photosynthetic electron transport chain [46] . ATTK1B plays an important role in plant growth and development through the nucleotide salvage pathway [47] . Thus , the high proportion of chloroplast related genes in these mutants indicated a key role of chloroplast function in chilling tolerance in Arabidopsis . We found in this study that RBD1 , one of the chloroplast localized proteins identified in the chilling sensitive screen , is involved in 23S rRNA processing . The highly conserved rrn operon on chloroplast genomes encodes the 23S , 16S , 5S and 4 . 5S rRNAs and three tRNAs [48] . The primary precursor is initially processed to generate tRNAs , precursors of 23S , 16S , 5S and 4 . 5S rRNA by a series of endo- and exonucleolytic cleavages [49 , 50] . The partially processed 23S rRNA is then excised at two sites named the hidden breaks [38] ( Fig 5B ) . Several molecules have been found to be involved in this final processing step . The CSP41b endonuclease and the RH39 RNA helicase are shown to be involved in processing the 23S rRNA at the 1st and 2nd hidden breaks respectively [36 , 38 , 51] . The RNase E endonuclease as well as the PNPase and RNase R exonucleases may also be involved in the final processing of 23S rRNA because their loss of function mutants accumulate incompletely processed 2 . 9-kb and 2 . 4-kb of the 23S rRNA , although their main function is to excise the 23S-4 . 5S species to generate 23S and 4 . 5S rRNAs [50 , 52 , 53] . Our study identifies RBD1 as a new regulator of 23S rRNA processing . Loss of RBD1 causes over-accumulation of the partially processed 2 . 9-kb and the 2 . 4-kb of 23S rRNA and reduction of the fully processed 1 . 1-kb species while the total amount of 23S rRNA does not change significantly ( Fig 5B ) . The RBD1 protein is localized to nucleoid where the processing happens ( Fig 4C ) , and it binds to the 23S rRNA but not other RNAs tested ( Fig 6 ) . This indicates that RBD1 functions in the last step of 23S rRNA processing , and it may facilitate the processing by guiding the endonucleases to the hidden breaks . Further refining its RNA biding sequences and determining its interaction with processing enzymes will reveal more its mode of action . Compared to the other 4 RNP mutants with similar chilling sensitive phenotypes , the rbd1-1 mutant exhibited the strongest 23S rRNA processing defect ( Fig 7B ) . It has additional chloroplast RNA defects , but they did not occur at normal growth temperature and were induced at a much later stage after plants were chilling treated ( S3 and S4 Figs ) . Because RBD1 does not appear to have enzymatic activities , it might interact with an RNase to bring it to the location for processing via its specific binding to the target RNA . Therefore , RBD1 is a positive facilitator of 23S rRNA processing . This function of RBD1 is more important at low temperatures than at normal temperatures . The rbd1 mutants show a slight defect at normal growth temperature but very strong defect at 4°C ( Fig 5B and S4A Fig ) . Interestingly , RBD1 exhibited a higher association with the 23S rRNA at 4°C than at 22°C , suggesting a regulation of its activity by temperature . The enhanced binding of RBD1 to the precursor could compensate for a less efficient binding of the processing enzyme RNase to 23S rRNA , so that chloroplast translation machinery can be efficiently produced at low temperatures as well . It is still yet to be determined how chloroplast function , especially translation , is the weak link in chilling tolerance . Likely , RNAs in chloroplast form more non- or less- functional secondary structures under chilling temperatures . Such structures of rRNAs may compromise the assembly or function of the translation machinery , while those of mRNAs may reduce their translation efficiency . RNA binding proteins such as RBD1 and other RNPs may reduce the formation of such non- or less-functional structures and thus become more critical under chilling conditions . Our study also indicates that various processes of RNA metabolism and protein translation in chloroplasts are inter connected . Different primary defects can have ripple effects leading to a similar chilling sensitive phenotype at later stages . For instance , RBD1 is closely associated with processing of the 23S rRNA transcripts . The primary defect of rbd1 mutants is likely the reduced processing and accumulation of mature 23S rRNA products . This will lead to defect of plastid ribosome especially under chilling conditions and subsequently the translation defect in the chloroplasts . Since the core subunits of PEP are synthesized on plastid ribosomes , the rbd1 mutants are expected to have reduced PEP level which will lead to a reduction in PEP-dependent mRNAs . As a compensation , the NEP dependent genes might become overexpressed [54] . Indeed , the chilling-treated rbd1 mutants had reduced transcript levels of the PEP-dependent genes tested such as ndhF , psaA and rbcL , while NEP-dependent genes such as ycf3 had higher expression . rps4 , transcribed by both PEP ( in green tissue ) and NEP ( in white tissue ) [55] , had an overexpression in rbd1 mutants under chilling condition ( S3 Fig ) . The slight deficiency in 16S rRNA processing might be an indirect effect of losing the RBD1 function because abnormal ribosome assembly have been shown to indirectly affect RNA processing [56 , 57] . Interestingly , the rbd1 mutants had a reduced level of chloroplast synthesized RbcL protein , but increased level of cytosol translated TOC75 protein ( Fig 7D ) , suggesting a compensation at the translation level as well . Together , this study reveals the importance of chloroplast RNA-binding proteins in chilling tolerance , and further studies will further enhance our understanding of molecular mechanisms of chilling tolerance and its variations in different plant species .
Arabidopsis T-DNA insertion lines were obtained from Arabidopsis Biological Resource Center with the stock numbers CS27941 , CS27942 and CS27943 . Four seeds of each line were sterilized and sowed on 0 . 5× MS ( Murashige and Skoog , Sigma ) solid medium with 1 . 2% agar and 2% sucrose . Plants were grown on vertical plates under a 16 h-light/d photoperiod for 8 days before being transferred to 4°C for two months . When using soil , plants were grown under a 12h-light photoperiod with light intensity at 100 μmol m-2 sec-1 and relative humidity at 50–70% . For spectinomycin treatment , seeds were sterilized and planted on 0 . 5×MS medium containing 0 . 8% agar and 2% sucrose with 0–3 mg/L spectinomycin under the conditions described above . For complementing the rbd1 mutants , a full-length At1g70200 cDNA was amplified and cloned into PMDC32 [58] to generate the construct of p35S::RBD1 . For the promoter reporter construct pRBD1::GUS , a 1 . 5-kb sequence upstream of the RBD1 translation start site was amplified and cloned into pCAMBIA1300 vector ( CAMBIA , http://www . cambia . org ) . To generate a GFP-tagged RBD1 fusion protein ( RBD1:GFP ) for transient expression in protoplasts , the coding region of the RBD1 was amplified from the cDNA and cloned into the pSAT6-EGFP-N1 vector [59] . Agrobacterium tumefaciens stains of GV3101 carrying the resulting constructs were used to transform plants by standard floral dipping [60] . Primary transformants were selected on 0 . 5×MS ( Murashige and Skoog , Sigma ) medium containing 0 . 8% agar and 2% sucrose with 25 mg/L hygromycin for selection . Protoplast isolation and transformation were carried out as previously described [61] . In brief , protoplasts were generated from 14-day-old wild-type Arabidopsis seedlings grown on plates with 12h light/ 12h darkness photoperiod and transformed with the plasmid DNAs . Protoplasts were then analyzed for GFP signals or for RNA co-IP from 12 hours to 48 hours after transformation . GUS staining was performed as described previously [62] . Briefly , tissues were stained with X-Gluc staining buffer ( 5 mM potassium ferrocyanide , 5 mM potassium ferricyanide , 100 mM sodium phosphate buffer , pH 7 . 0 , and 0 . 005% Triton X-100 and 2 mM 5-bromo-4-chloro-3-indolyl-beta-D-glucuronic acid ) for 1–2 hours at 37°C , followed by incubating in 70% ethanol to remove chlorophyll . Total RNA was extracted with TRIzol reagent ( Invitrogen ) according to the manufacturer’s protocol . cDNAs were synthesized from total RNA by using AffinityScript QPCR cDNA Synthesis Kit ( Agilent Technologies ) . Real-time quantitative PCR was performed on the BIO-RAD PCR System using iQSYBR GREEN SuperMix ( BIO-RAD ) . Actin was used as a control . Total RNA was extracted with TRIzol reagent ( Invitrogen ) according to the manufacturer’s instructions . For transcript analysis , five youngest visible leaves were harvested for RNA extraction . Ten micrograms of RNA per sample were separated on an agarose gel containing 1 . 2% formaldehyde and then transferred to uncharged nylon membranes ( HybondN; GE Healthcare ) . The blots were UV cross-linked ( 150 mJ/cm2 ) and hybridized with gene specific , 32P labeled , single strand DNA probes . Arabidopsis leaves were quick frozen in liquid nitrogen and homogenized in extraction buffer ( 50mM Tris-HCl , 1mM EDTA , 1mM EGTA , 150mM NaCl , 10% Glycerol , 5mM DTT , 0 . 25% Triton-X 100 , 2% PVPP , 1mM PMSF ) . After centrifugation at 14 , 000 rpm for 10 min twice , the supernatants were mixed with loading buffer , boiled and loaded onto 12% SDS polyacrylamide gel . The proteins were visualized with Coomassie Blue staining of the gel or Ponceau S staining of the transferred blot . After transformation , the protoplasts were incubated at 22°C for 8 h before half of each sample was transferred to 4°C and the other half to 22°C for 12 h . Protoplasts were then disrupted in 500 μL protoplast disruption buffer ( 0 . 3M sorbitol , 20mM Tricine-KOH ( pH 8 . 4 ) , 10 mM EDTA , 10 mM NaHCO3 and 0 . 1% BSA ) and incubated on the ice for 30 minutes with several inversion , then were centrifuged at 300×g for 2 minutes . Poured off the supernatant , chloroplasts pellet were disrupted in 200 μL chloroplast disruption buffer ( 2mM DTT , 200 mM KOAC , 30 mM HEPES , pH 8 . 0 , 10 mM MgOAc , and 2mg/ml proteinase inhibitor cocktail ) and incubated on the ice for 30 minutes with occasional roughly vortex , then were centrifuged at 16 , 000×g for 30 minutes at 4°C . The supernatant was diluted with one volume of Co-IP buffer ( 150 mM NaCl , 20 mM Tris-HCl , 1 mM EDTA , 5 mM MgCl2 , 1 . 1% Triton X-100 , 100U/ml RNase inhibitor , 1 mM PMSF and 2mg/ml proteinase inhibitor cocktail ) and incubated with 10 μg of GFP antibody ( mouse IgG2a , Invitrogen ) for 6 h with rotation , followed by 2 h rotation with 50 μL Dynabeads Protein G ( Invitrogen ) at 4°C . Beads containing the IPed protein and its bound RNAs were collected with a magnet , and supernatants were recovered and pellets were washed three times with Co-IP buffer . After the last wash , pellets were resuspended in 200μL Co-IP buffer . Resuspension was then extracted with TRIzol reagent ( Invitrogen ) according to the manufacturer’s protocol . Total RNA was treated by DNase I ( Promega ) to remove DNA contamination , and RT-minus control was performed to confirm complete removal of DNA in the following steps . cDNAs were synthesized from total RNA by using Superscript III reverse transcriptase ( Invitrogen ) . Real-time quantitative PCR was performed on the BIO-RAD PCR System using iQSYBR GREEN SuperMix ( BIO-RAD ) .
|
Compared to cold acclimation ( enhancement of freezing tolerance by a prior exposure to low non-freezing temperature ) , the tolerance mechanism to non-freezing chilling temperatures is not well understood . Here , we performed a genome-wide mutant screen for chilling sensitive phenotype and identified 49 candidate genes important for chilling tolerance in Arabidopsis . Among the proteins encoded by these 49 genes , 16 are annotated as having chloroplast localization , suggesting a critical role of chloroplast function in chilling tolerance . We further studied RBD1 , one of the four RNA-binding proteins localized to chloroplast . RBD1 is only expressed in the green photosynthetic tissues and is localized to nucleoid of chloroplasts . Furthermore , RBD1 is found to be a regulator of 23S rRNA processing likely through direct binding to the precursor of 23S rRNA in a temperature dependent manner . Our study thus reveals the importance of chloroplast function especially protein translation in chilling tolerance at genome-wide scale and suggests an adaptive mechanism involving low temperature enhanced activities from proteins such as RBD1 in chilling tolerance .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"plant",
"anatomy",
"rna-binding",
"proteins",
"plant",
"cell",
"biology",
"rna",
"extraction",
"brassica",
"chloroplasts",
"plant",
"science",
"model",
"organisms",
"plants",
"cellular",
"structures",
"and",
"organelles",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"proteins",
"leaves",
"ribosomes",
"biochemistry",
"rna",
"plant",
"cells",
"plant",
"and",
"algal",
"models",
"ribosomal",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"phenotypes",
"genetics",
"ribonucleoproteins",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"non-coding",
"rna",
"organisms"
] |
2016
|
Chloroplast RNA-Binding Protein RBD1 Promotes Chilling Tolerance through 23S rRNA Processing in Arabidopsis
|
Mutualistic associations between symbiotic bacteria and their hosts are common within insect systems . However , viruses are often considered as pathogens even though some have been reported to be beneficial to their hosts . Herein , we report a novel densovirus , Helicoverpa armigera densovirus-1 ( HaDNV-1 ) that appears to be beneficial to its host . HaDNV-1 was found to be widespread in wild populations of H . armigera adults ( >67% prevalence between 2008 and 2012 ) . In wild larval populations , there was a clear negative interaction between HaDNV-1 and H . armigera nucleopolyhedrovirus ( HaNPV ) , a baculovirus that is widely used as a biopesticide . Laboratory bioassays revealed that larvae hosting HaDNV-1 had significantly enhanced resistance to HaNPV ( and lower viral loads ) , and that resistance to Bacillus thuringiensis ( Bt ) toxin was also higher at low doses . Laboratory assays indicated that the virus was mainly distributed in the fat body , and could be both horizontally- and vertically-transmitted , though the former occurred only at large challenge doses . Densovirus-positive individuals developed more quickly and had higher fecundity than uninfected insects . We found no evidence for a negative effect of HaDNV-1 infection on H . armigera fitness-related traits , strongly suggesting a mutualistic interaction between the cotton bollworm and its densovirus .
The interactions between symbiotic species and their hosts are becoming increasingly understood within insect systems [1] , [2] , [3] . Symbionts form diverse evolutionary relationships that influence the life history of their host , from mutualistic , by protecting them from natural enemies or increasing their host's fitness though a variety of means [1] , [4] , [5] , [6] , [7] , [8] , to parasitic , either by decreasing their resistance to harmful microorganisms or their tolerance to environmentally harmful factors , or by killing them directly [9] , [10] , [11] . There is a growing literature on the mutualistic interactions between intracellular bacterial symbionts , such as Wolbachia and their insect hosts , in which the symbionts spread through the host population by increasing the fitness of infected hosts [1] , [6] , [12] , [13] . However , viral mutualistic symbioses have rarely been reported . This may be because , as obligate symbionts , viruses have long been considered harmful to their host and are usually isolated from cadavers killed by the virus . Moreover , until relatively recently , laboratory techniques only had the capacity to shed light on overtly pathogenic viruses , and not covert beneficial ones [14] , [15] , [16] . The development of molecular and sequencing technology facilitates the discovery and analysis of non-pathogenic virus species , using techniques such as suppression subtractive hybridization ( SSH ) and RNA-seq [17] , [18] . Generally , viruses isolated from healthy individuals may be conditionally beneficial to their hosts . Recently , these ‘good viruses’ have attracted more attention , largely due to the prospect of using them in applications such as gene therapy and as tools for gene manipulation [2] , [19] . As defined by Roossinck , there are few examples of viral mutualistic symbioses in insects ( identified as conveying benefit to the host without any detectable fitness costs ) [2] . The cotton bollworm moth , Helicoverpa armigera , is a major migratory pest of cotton and other economically-important crops throughout Asia , Africa , Europe and Australasia [20] , [21] , [22] . In China , the introduction of Bt-cotton in the 1990s has seen a dramatic decline in the H . armigera moth population . However , there are signs of Bt-resistance emerging [23] , [24] , fueling renewed interest in other forms of biological pest control , including the use of host-specific viral pesticides , derived from densoviruses [25] , small RNA viruses [26] and baculoviruses [27] , [28] , [29] , [30] , [31] . Previously , we reported a novel densovirus ( HaDNV-1 , from the family Parvoviridae ) in H . armigera moths that possesses a monosense genome that is 4926 nucleotides in length and clustered with the members of the genus Iteravirus in phylogenetic analysis [32] . This has allowed further investigation into the interactions between HaDNV-1 and its host H . armigera , which we report here . The main objective of this study was to establish the ecological significance of this virus within the migratory H . armigera system . Specifically , we undertook experiments to determine the transmission strategies of HaDNV-1 , the impact of HaDNV-1 infection on host fitness , including its capacity to modulate resistance to potentially lethal biopesticides , and the prevalence of HaDNV-1 in field populations of H . armigera . Our results show that HaDNV-1 can be both horizontally- and vertically-transmitted in H . armigera; that HaDNV-1 infection increases host-fitness by increasing larval/pupal development rate , female lifespan and egg/offspring production; and that it also enhances larval resistance to H . armigera nucleopolyhedrovirus ( HaNPV ) , a widely-used biopesticide . Resistance to Bt Cry1Ac protoxin was also enhanced , but only at relatively low toxin concentrations . Overall , we found no evidence for a negative effect of densovirus infection on H . armigera fitness-related traits , strongly suggesting a mutualistic interaction between the cotton bollworm and HaDNV-1 .
To establish the modes of transmission of the densovirus HaDNV-1 , we first produced an uninfected laboratory colony from a single breeding pair of H . armigera ( NONINF strain ) . An infected strain ( INF strain ) was subsequently produced using neonate larvae from the NONINF strain , dosing them with either purified HaDNV-1 ( 108/µl; method 1 , see Materials and Methods ) or filtered liquid from infected individuals ( 108/µl; method 2 , see Materials and Methods ) . Thus , our results indicated that HaDNV-1 could efficiently infect larvae by oral ingestion . The efficiency of infection with filtered liquid was higher than that of the purified virus ( Table 1 , Fig . S1A , S1B ) , suggesting that the purification process might have inactivated the virus in some way . We also found that individuals artificially infected with HaDNV-1 via peroral infection could efficiently transmit the viral infection to their offspring ( Fig . S1C ) , and the same was true for naturally infected individuals ( Fig . S1D ) , suggesting vertical transmission of the virus . HaDNV-1 was capable of being vertically-transmitted from both infected females and infected males , but transmission-efficiency was higher from infected females than males ( Table 1 , Fig . S1E , S1F , S1G ) . With qPCR , we tested whether vertical transmission of HaDNV-1 was due to virus contamination on the surface of the eggs ( transovum ) , or whether the virus was transmitted within the egg itself ( transovarial ) . HaDNV-1 titers were not significantly different between sodium hypochlorite-treated and non-treated eggs ( t = 1 . 296 , d . f . = 6 , P = 0 . 24 ) ( Fig . 1 ) , suggesting that transovarial transmission was occurring . To examine the possibility of horizontal transmission through ingestion of contaminated foodplant ( as would be a possibility in wild populations ) , we placed uninfected neonate larvae in diet cells that had previously housed infected insects ( n = 8 ) . Our results indicated that horizontal virus transmission did not occur in this manner , despite our previous experimental evidence that larvae could be orally infected . To examine this further , we used a range of HaDNV-1 concentrations to infect larvae and subsequently examined virus intensity in host frass ( faeces ) . As expected , larval infection rate was positively related to the magnitude of the HaDNV-1 oral challenge , with low infection rates at doses less than 106/µl ( Table 2 ) ; but even for larvae challenged with large viral doses , their frass contained only very low levels of HaDNV-1 , with only 3 out of 20 samples containing more than 1×105/mg and none with more than 5×105/mg . Therefore , while we cannot exclude the possibility that horizontal transmission of HaDNV-1 may occur via the oral-fecal route , the viral levels in frass were very low and may not be sufficient for oral infection . HaDNV-1 distribution was quantified within different host body tissues using qPCR . In both larvae and adults , HaDNV-1 titers were significantly higher in the fat body than in all other tissues: larvae: F = 11 . 098 , d . f . = 5 , 36 , P<0 . 0001 ( Fig . 2A ) ; adult females: F = 26 . 601 , d . f . = 5 , 30 , P<0 . 0001 ( Fig . 2B ) ; adult males: F = 44 . 560 , d . f . = 5 , 30 , P<0 . 0001 ( Fig . 2C ) . Using H . armigera as a control , we tested four other species of lepidopterans for their potential to act as alternative hosts for HaDNV-1 , by attempting oral inoculation in Spodoptera exigua , Spodoptera litura , Agrotis segetum and Agrotis ipsilon . Results indicated that while oral inoculation with HaDNV-1 could successfully infect H . armigera , none of the four other species tested positive ( Fig . S2A ) . We also tested field-captured adults of the closely-related species H . assulta but failed to find any HaDNV-1 positive individuals ( n = 9; Fig . S2B ) . Based on these available data , it appears that infection with HaDNV-1 is host-specific to H . armigera . To quantify the impact of HaDNV-1 infection on H . armigera development , a number of bioassays were performed using neonate larvae orally inoculated with filtered liquid from either HaDNV-1 infected ( DNV+ ) or non-infected ( DNV− ) individuals ( Fig . S3A , S3B ) . Both male and female DNV+ individuals developed significantly more quickly than the control individuals in both the larval ( female: t = 2 . 732 , d . f . = 312 , P = 0 . 0067 , male: t = 4 . 147 , d . f . = 379 , P<0 . 001 ) ( Fig . 3A ) and pupal stages ( female: t = 5 . 100 , d . f . = 312 , P<0 . 001 , male: t = 4 . 057 , d . f . = 379 , P<0 . 001 ) ( Fig . 4A ) . Between 7–11 days post-hatch ( approximately 3rd–5th instar ) DNV+ larvae weighed significantly more than DNV- larvae by an average of ∼20% ( GLMM with larval identity as a random term and log10-transformed larval weight as the dependent variable: Age ( days ) : F = 2386 . 8 , d . f . = 1 , 127 , P<0 . 0001; HaDNV-1 infection status ( +ve or −ve ) : F = 27 . 25 , d . f . = 1 , 36 , P<0 . 0001 ) ( Fig . 3B , Fig . S4 ) . However , their growth rates over this period did not differ , suggesting that densovirus effects on larval growth rate occurred prior to day 7 post-hatch ( GLMM: interaction between infection status and age: F = 0 . 01 , d . f . = 1 , 126 , P = 0 . 91 ) ( Fig . S4 ) . A chloroform-wash assay indicated that at 9 days old , DNV+ larvae contained more lipid than DNV− individuals , measured as either lipid mass ( t = 2 . 045 , d . f . = 50 , P = 0 . 046 ) or as a percentage of the whole body ( t = 2 . 342 , d . f . = 50 , P = 0 . 023 ) ( Fig . 3C , 4D ) . Larval mortality of DNV+ was significantly lower than DNV− ( Table 3 ) . However , there was no significant difference in pupal weight between DNV+ and DNV− insects ( GLM: densovirus infection status: F = 0 . 99 , d . f . = 1 , 692 , P = 0 . 329; Sex: F = 41 . 08 , d . f . = 1 , 693 , P<0 . 0001; interaction term: F = 0 . 064 , d . f . = 1 , 691 , P = 0 . 80; female: t = 0 . 96 , d . f . = 312 , P = 0 . 34 , male: t = 0 . 481 , d . f . = 379 , P = 0 . 63 ) ( Fig . 4B ) , or pupation rate or eclosion rate between HaDNV-1 positive and HaDNV-1 negative insects ( Table 3 ) . To determine the effect of HaDNV-1 infection on adult life-history traits , we used individuals from the non-infected ( NONINF ) and infected ( INF ) strains; and their infection status was confirmed by PCR ( Fig . S3C , S3D ) . Infected INF strain moths produced significantly more eggs ( t = 2 . 172 , d . f . = 93 , P = 0 . 032; Fig . 4C ) and more neonates ( t = 3 . 026 , d . f . = 93 , P = 0 . 0032; Fig . 4D ) than individuals from the uninfected NONINF strain . Egg viability ( hatch-rate ) was significantly higher in the INF strain than in the NONINF strain ( Table 3 ) . The life-span of densovirus-infected females was significantly longer than that of females that were virus-free ( χ21 = 13 . 5 , d . f . = 1 , P = 0 . 0002; Fig . 4E ) , but the longevity of males was not significantly different between the two strains ( χ2 = 1 . 64 , d . f . = 1 , P = 0 . 2; Fig . 4E ) . In larval field-collections , there was a non-random association between the two viruses ( Chi-square test with Yates' correction: χ2 = 35 . 63 , d . f . = 1 , P<0 . 0001 ) . Thus , there were relatively fewer larvae infected with both HaDNV-1 and HaNPV than would be expected by chance alone ( 14% versus 20% ) . When split by year , this effect was significant in 2012 , when the overall HaNPV prevalence was 61% ( χ2 = 19 . 75 , d . f . = 1 , P<0 . 0001; proportion infected with both viruses = 20% observed versus 26% expected ) , but not in 2013 , when HaNPV prevalence was just 4% ( χ2 = 0 . 82 , d . f . = 1 , P = 0 . 36; 2% observed versus 2% expected ) ( Table S1 ) . In adult field-collections , the prevalence of HaDNV-1 infection was uniformly high each year between 2008 and 2012 - 87% , 81% , 77% , 68% and 67% , respectively ( Fig . S5 ) . However , there was evidence for a significant decline in densovirus prevalence over the five years ( GLMM with location as a random effect: χ21 = 39 . 06 , P<0 . 0001 ) . Despite high levels of baculovirus being observed in the larval field populations , we failed to detect any HaNPV-positive individuals in a random selection adult moths collected from four geographically diverse sites ( n = 361 samples ) . To determine the interaction between the densovirus HaDNV-1 and the baculovirus HaNPV , we first confirmed individuals from NONINF strain were NPV-free using PCR with specific primers . Then , NONINF strain neonates were inoculated with either HaDNV-1 ( DNV+ ) or water ( DNV− controls ) , and infections verified using PCR . Survival to pupation in larvae not exposed to HaNPV did not differ between DNV+ ( 95% ) and DNV− ( 92% ) larvae ( χ2 = 0 . 27 , d . f . = 1 , P = 0 . 60 ) . However , for those larvae exposed to the baculovirus , there was a significant difference between DNV+ and DNV− larvae in their susceptibility to HaNPV ( GLM: HaDNV-1 infection-status: χ2 = 4 . 04 , d . f . = 1 , P = 0 . 044 , parameter estimate ± standard error = 0 . 4645±0 . 2319 ) , with densovirus-infected larvae suffering lower mortality rates for a given virus dose ( GLM: log10 virus dose: χ21 = 98 . 56 , P<0 . 0001; LC50s = 3 . 13×107 versus 9 . 10×107 OB per ml , for DNV− and DNV+ larvae , respectively; Fig . 5A ) ; the interaction between viral dose and infection status was marginally non-significant ( dose*status: χ2 = 3 . 72 , d . f . = 1 , P = 0 . 054 ) . We tested the differences of HaNPV replication between HaDNV-1 positive and negative individuals by repeating the HaNPV bioassay with 108 OBs/ml . The baculovirus bioassay indicated that there was no HaNPV-induced mortality in the control larvae that were exposed to water only , and that most mortality in the HaNPV-challenged larvae started at day 5 ( 120 h post-inoculation ) ( Fig . 5B ) . In NPV-challenged larvae , those carrying HaDNV-1 suffered significantly lower mortality overall than HaDNV-1 negative insects ( Likelihood-ratio test: χ2 = 23 . 24 , d . f . = 1 , P<0 . 0001; linear coefficient ( 95% confidence interval ) = 0 . 248 ( 0 . 134 , 0 . 457 ) ) . Therefore , we collected samples before day 5 post-challenge to estimate HaNPV viral loads using qPCR . As would be expected , HaNPV titers ( log-transformed ) increased over time post-challenge and the rate of HaNPV titer increase was lower for HaDNV-1 positive larvae than in larvae lacking HaDNV-1 , as indicated by a significant interaction term ( linear model: Time post-challenge: F = 27 . 02 , d . f . = 1 , 112 , P<0 . 0001; DNV infection status: F = 5 . 69 , d . f . = 1 , 112 , P = 0 . 019; Time* DNV status interaction: F = 8 . 69 , d . f . = 1 , 112 , P = 0 . 0038; Fig . 5C ) . However , HaNPV titers were not directly correlated with HaDNV-1 titers in HaDNV-1 positive individuals ( r = 0 . 066 , n = 58 , P = 0 . 623 ) . These results suggest that HaDNV-1 protected H . armigera from HaNPV , possibly by slowing the accumulation of HaNPV . A similar bioassay using the Bt toxin Cry1Ac instead of the baculovirus generated consistent results . As expected , larval development score increased over time and declined with increasing Bt dose ( linear mixed-effects model with larval identity as a random term: Day: F = 18147 . 38 , d . f . = 1 , 4172 , P<0 . 0001; Log2Btdose: F = 1335 . 48 , d . f . = 1 , 4172 , P<0 . 0001 ) . However , development was also influenced by the interaction between DNV infection status and the dose of Bt administered ( DNV status: F = 120 . 21 , d . f . = 1 , 4172 , P<0 . 0001; DNV status * Bt dose interaction: F = 111 . 81 , d . f . = 1 , 4172 , P<0 . 0001 ) , with the enhanced development of HaDNV-1 positive larvae at low Bt concentrations declining as Bt dose increased , such that mean development rate was independent of DNV infection status as Bt concentrations above 1 . 6 µg/g ( Fig . 6 ) . We also performed the bioassay with Bt cotton . As expected , there was a significant effect of Bt cotton on larval development rate , with development being significantly stunted in larvae exposed to the Bt plants ( linear model: Diet: F = 63 . 74 , d . f . = 1 , 476 , P<0 . 001; mean score ± s . e . : Bt cotton = 1 . 717±0 . 153; non-Bt cotton = 3 . 529±0 . 167 ) . However , whilst DNV positive larvae tended to have slightly higher development scores than DNV negative larvae ( 2 . 754±0 . 176 versus 2 . 492±0 . 164 ) , this difference was non-significant and the interaction between DNV status and Bt exposure was also non-significant ( DNV status: F = 1 . 336 , d . f . = 1 , 476 , P = 0 . 24; DNV status * Diet interaction: F = 0 . 0084 , d . f . = 1 , 476 , P = 0 . 93 ) .
In conclusion , our studies to date suggest a mutualistic relationship between the cotton bollworm and HaDNV-1 , in which the cotton bollworm appears to benefit from HaDNV-1 infection , with all host fitness parameters so far tested ( larval growth rate , larval and pupal development rate , fertility , adult female lifespan , and resistance to baculovirus and low doses of Bt toxin ) enhanced at no detectable cost . The study of beneficial viruses in both vertebrate and invertebrate systems has only relatively recently attracted researchers' attention [2] , predominantly due to the explosion of new technologies that now make the detection of such organisms possible . It should be noted that the coevolution between viral mutualistic symbionts and their hosts could be an important factor to consider when studying the adaptability of insect host species . Illuminating the function of such viral symbionts may offer novel insights for future pest management strategies .
Cotton bollworms ( H . armigera ) were reared using artificial diet [53] at 25±1°C with a 14:10 , light:dark photoperiod . Adult moths were provided with 10% sugar and 2% vitamin complex . The colony was established from thirty breeding pairs captured at Langfang ( Hebei province , China ) in 2005 . Individuals successfully producing offspring were tested for the presence of HaDNV-1 , using the methods described below . Offspring from a single uninfected breeding pair were reared to produce the NONINF strain ( uninfected ) laboratory culture . HaDNV-1 virus was isolated from migrating H . armigera adults captured in 2010 and 2011 using a vertical-pointing trap , and stored in liquid nitrogen [20] . Briefly , DNA was extracted from host tissues ( except for the abdomens ) of each individual , and PCR undertaken to detect the presence of HaDNV-1 . Subsequently , the abdomens of positive individuals were divided into two groups: one group was used to purify the HaDNV-1 using the method described by La Fauce et al . ( method 1 ) [54]; the other group was used to prepare a filtered liquid , containing an unpurified form of virus ( method 2 ) . Briefly , this second method involved grinding four abdomens under liquid nitrogen and transferring to 1 ml PBS buffer ( 0 . 01M , pH 7 . 4 ) . The homogenate was centrifuged at 6500×g for 15 min at 4°C , and the liquid supernatant subsequently filtered with Sartorius Minisart 0 . 2 µm PES ( Invitrogen , Grand Island , USA ) . The abdomens of negative individuals were filtered using the same method . Quantification of the viruses was performed using the qPCR method described below . All the samples were stored at −20°C . To detect the existence of HaDNV-1 in H . armigera , specific primers amplifying a 496 bp fragment , DVVPF/DVVPR ( Table S2 ) were designed according to the genomic sequence of HaDNV-1 . The PCR program was as follows: 30 s at 94°C , 30 s at 55°C , and 30 s at 72°C for 40 cycles . For detection of H . armigera nucleopolyhedrovirus ( HaNPV ) , a pair of specific primers amplifying a fragment of 445 bp , NPVF/NPVR , were designed according to the open reading frame 14 ( ORF14 ) of the genomic sequence of HaNPV . The PCR program was as follows: 30 s at 94°C , 30 s at 57°C , and 30 s at 72°C for 40 cycles . For quantifying the copy numbers of HaDNV-1 and HaNPV , an absolute quantification qPCR methodology using a standard curve was performed [55] . Fragments containing the primers and probes of HaDNV-1 and HaNPV were amplified with our de novo primers ( PF/PR for HaDNV-1 , NPVF/NPVR for HaNPV ) using the program: 30 s at 94°C , 30 s at 53°C , and 60 s at 72°C for 40 cycles , and cloned into the pEASY-T Cloning Vector ( TransGen , Beijing , China ) . These plasmids were subsequently used for the quantification standard curve assay . qPCR was carried out with the TaqMan method in 20 µl reaction agent comprised of 1 µl of template DNA , 2×Premix Ex Taq ( Takara , Japan ) , 0 . 2 µM of each primer and 0 . 4 µM probe , using a 7500 Fast Real-time PCR System ( Applied Biosystems ) . Thermal cycling conditions were: 45 cycles of 95°C for 15 s , 60°C for 34 s . The DNA sample of each group was replicated three times . All primers used in this study were shown in Table S2 . The equation of y = −1 . 052x+42 . 327 ( y = the logarithm of plasmid copy number to base 2 , x = Ct value , R2 = 0 . 9997 ) and y = −0 . 9861x+44 . 647 ( y = the logarithm of plasmid copy number to base 2 , x = Ct value , R2 = 0 . 9999 ) were used to calculate the copy number of HaDNV-1 and HaNPV , respectively . We constructed an infected line ( INF strain ) of H . armigera by orally infecting NONINF strain larvae with HaDNV-1 ( from filtered liquid , method 2 - see above ) and maintained them by vertical transmission of the virus , using the primers DVVPF/DVVPR to confirm successful establishment of HaDNV-1 infection . Subsequently , individuals from both NONINF strain and INF strain were used to determine the transmission modes of HaDNV-1 . For vertical transmission , ♀+/♂− , ♀−/♂+ , ♀+/♂+ and ♀−/♂− pairs were crossed and DNA from 3rd instar offspring larvae used to probe for HaDNV-1 . For the diet contamination assay , ( to determine horizontal transmission efficiency ) , infected individuals from the INF strain were reared in diet cells until the start of the 3rd instar and then removed . Uninfected NONINF strain neonates were then placed in the vacated cells and reared to the pupal stage . DNA was extracted from the adults and probed for HaDNV-1 infection using PCR . Horizontal transmission of HaDNV-1 was determined using PCR with adult DNA as temples and different concentrations of the densovirus: 108 , 107 , 106 , 105 , 104/µl . The frass of larvae from HaDNV-1 positive individuals were also quantified by qPCR , as described above . To examine virus infection in different body tissues , DNA was extracted from body parts of infected individuals ( both larval and adult stages ) and the copy numbers of HaDNV-1 were quantified by qPCR . To account for individual variation , we first calculated the copy numbers per milligram of tissue and then summed all the copy numbers from different tissues from the same individual and the percentage of each tissue was statistically analyzed ( larvae: n = 7; adult males: n = 6; adult females: n = 6 ) . To further establish the role of vertical transmission in the life-cycle of the densovirus , we quantified HaDNV-1 infections in H . armigera eggs , primarily to distinguish between transovarial and transovum infection routes . Eggs from INF strain breeding pairs , which both of females and males were infected by HaDNV-1 , were submerged in 1% sodium hypochlorite for 10 minutes . They were then filtered through a damp cloth , thoroughly rinsed , and allowed to dry . Four groups of hypochlorite-treated eggs ( n = 50 eggs per group ) were tested against non-treated eggs ( control ) and HaDNV-1 infections tested by qPCR . To test the impact of HaDNV-1 infection on the life table parameters of its host , neonate NONINF strain larvae were first orally inoculated with either filtered-liquid containing HaDNV-1 , or filtered-liquid from uninfected individuals ( control ) . One hundred NONINF strain neonates were placed in each treatment Petri-dish for 2 days to ensure that larvae ingested the treated diet . They were then transferred to a 24-well plate ( one individual per well: diameter = 1 . 5 cm; height = 2 cm ) until the 5th larval instar; larvae were then individually reared in glass tubes until eclosion ( diameter = 2 cm; height = 7 . 5 cm ) ( Fig . S6 ) . The status of individuals was checked every day at 9:00 am . The weight of larvae from the 7th to 11th day post hatch , and the pupa on the 3rd day were recorded . Fifth-instar larvae were randomly selected to estimate the infection rate of HaDNV-1 during the experiment . This bioassay was replicated twice ( n = 288 and n = 168 individuals , respectively ) . Individuals dying within 24 hours of the experimental set up were considered handling deaths , and excluded from the analysis . In addition , newly eclosed adults from both the HaDNV-1 negative NONINF strain and HaDNV-1 positive INF strain were used to determine longevity , egg production and hatch rate . Three pairs of adults were put in each plastic cup ( diameter = 8 . 5 cm; height = 10 cm ) ( Fig . S6 ) . The experimental replicates were 3×77 for NONINF strain and 3×60 for INF strain , respectively . We recorded the number of eggs and newly hatched larvae every day . After death , individuals were used to detect HaDNV-1 via PCR . Data from failed matings were excluded . To quantify the impact of HaDNV-1 infection on host growth , we measured relative lipid mass within larvae of H . armigera . Larvae 9 days post-hatch were chosen to compare the lipid content between HaDNV-1 positive ( n = 19 ) and HaDNV-1 negative ( n = 33 ) individuals . The protocol was undertaken as Clissold et al . [56] . Briefly , the larval samples were freeze-dried , weighed , chloroform-extracted 3 times , dried again and weighed . The lipid mass was calculated by subtracting the post-chloroform-wash mass from the pre-chloroform-wash mass . To assess the capacity of HaDNV-1 to act as a beneficial symbiont , we quantified the interaction between HaDNV-1 and the common baculovirus pathogen HaNPV , via a series of laboratory bioassay studies . As previously described , neonate larvae were first treated with HaDNV-1 filtered liquid ( either from HaDNV-1 infected or HaDNV-1 negative individuals ) . Two-day old larvae were then transferred to a 24-well plate and maintained on diet until the 9th day after hatching . Individuals weighing between 5–11 mg ( early third-instar stage ) were chosen for the HaNPV bioassay . Purified powder of HaNPV at a concentration of 5×1011 occlusion bodies ( OBs ) per g was generously provided by Dr . Qilian Qin in the Institute of Zoology , Chinese Academy of Science , Beijing , China . Larvae were orally dosed with 4 treatments of HaNPV ( 30 larvae per treatment at: 0 ( control ) , 1×106 , 1×107 , 1×108 , and 1×109 OBs/ml ) . Only larvae that ingested all the NPV within a 24 h period were used for the bioassay . Larvae were subsequently monitored daily for NPV mortality until pupation , and all viral deaths stored at −20°C . PCR with specific primers was used to test for NPV in dead larvae with non-obvious symptoms . To assess HaNPV infection levels in HaDNV-1 positive and negative individuals , we performed a separate HaNPV bioassay with 108 OBs/ml . There were 24 individuals in each replicate and three replicates per treatment . Only larvae that ingested all the NPV within a 24 h period were used for the bioassay . The absolute quantification qPCR methodology was used to quantifying the copy numbers of HaNPV as described above . Survival analysis was conducted using Cox's proportional hazards model . For the Bacillus thuringiensis bioassays , various concentrations of the Bt Cry1Ac protoxin were added and thoroughly mixed with standard artificial diet to obtain the desired concentrations ( 0 ( control ) , 0 . 4 µg/g , 0 . 8 µg/g , 1 . 6 µg/g and 3 . 2 µg/g ) . After mixing , the diet solidified and solid 1 mg pieces were placed into each well of a 24-well plate and two-day old larvae infected or uninfected by HaDNV-1 were then transferred to each well ( Fig . S6 ) . There were 24 individuals in each replicate and three replicates per treatment . We graded the larvae from day 4 to day 9 after hatching according to the development rate: death = 0 , early first instar stage = 1 , middle first instar stage = 2 , last first instar stage = 3 , early second instar stage = 4 , middle second instar stage = 5 , last second instar stage = 6 , early third instar stage = 7 , middle third instar stage = 8 , last third instar stage = 9 , early fourth instar stage = 10 , middle fourth instar stage = 11 [57] . At seedling stage with 5 leaves , we chose the new cotton 33B with Cry1Ac ( Monsanto Company , Bt cotton ) using Shi Yuan 321 ( Shijiazhuang Acadamy of Agricultural Sciences , NonBt cotton ) as control to perform the bioassay . Two-day old larvae infected or uninfected by HaDNV-1 were transferred to a 24-well plate with Bt-cotton or NonBt-cotton . There were 40 individuals in each replicate and three replicates per treatment . We graded the larvae after 7 days according to the development rate . Samples of larvae were collected at 7 locations in 2012 ( Jinan , Dezhou and Taian , Shandong province; Cangzhou , Heibei province; Tianmen and Qianjing , Hubei province; Maanshan , Anhui province ) and 6 locations in 2013 ( Luohe , Luoyang , Yuanyang and Nanyang , Henan province; Langfang and Cangzhou , Hebei province ) . The infection rate of HaDNV-1 and HaNPV was determined using the PCR method described as above . Samples of adults were collected at fifteen locations from 2008 to 2012: A = Xinxiang , Henan province; B = Dezhou; C = Langfang; D = Yantai Shandong province; E = Yancheng , Jiangsu province; F = Handan , Shandong province; G = Changde , Hunan province; H = Tianmen , I = Qianjiang; J = Maanshan; K = Taian; L = Luohe; M = Weinan , Shanxi province; N = Shihezi , O = Kashi , Xinjiang province . We also randomly selected four places to detect HaNPV in the populations , including site 1 in 2010 ( 54 samples ) , site 2 in 2010 ( 103 samples ) , site 4 in 2012 ( 104 samples ) and site 13 in 2011 ( 100 samples ) . Using the same oral inoculation method as previously described ( section 2 . 5 ) , we chose four species of Lepidoptera ( Spodoptera exigua , Spodoptera litura , Agrotis segetum , Agrotis ipsilon ) to determine the host range of HaDNV-1 infection . We also collected nine adults of H . assulta from field populations , and PCR was used to detect HaDNV-1 infection . Statistical analyses were conducted using STATA v . 9 . 0 and R v3 . 0 . 1 [58] . Student's t-test or ANOVA with Tukey were used to determine the level of significance in the relative levels of HaDNV-1 . Egg hatch rates and larval/pupal mortality , pupation and eclosion rates were determined using generalized linear models ( GLMs ) with binomial errors . Analysis of the NPV and Bt bioassay data was also conducted using GLMs with binomial errors . A generalised linear mixed effects model ( GLMM ) with binomial errors was used to determine temporal variation in HaDNV-1 infection rates . A GLMM with Gaussian errors was used to quantify variation in larval growth rates with larval identity included as a random term . Development following exposure to Bt toxin in artificial diet was analyzed using linear mixed effects models using the lme function in R , with larval identity as a random term to account for the repeated measures data structure . The GenBank accession number of genomic sequence of HaDNV-1 and HaNPV were HQ613271 and AF303045 , respectively .
|
The old world cotton bollworm , Helicoverpa armigera , is one of the most significant pests of crops throughout Asia , Europe , Africa and Australia . Herein , we report a novel densovirus ( HaDNV-1 ) which was widely distributed in wild populations of H . armigera and was beneficial to its host by increasing larval and pupal development rates , female lifespan and fecundity , suggesting a mutualistic interaction between the cotton bollworm and HaDNV-1 . The cotton bollworm is currently widely controlled by the biopesticides Bacillus thuringiensis ( Bt ) toxin and the baculovirus HaNPV . It is therefore important to estimate the risk that the symbiotic virus will negatively impact on the efficiency of these biopesticides . Field and laboratory results suggest that HaDNV-1 infection significantly increases larval resistance to HaNPV and Bt toxin . These results have important implications for the selection of biopesticides for this species , and highlight the need for greater research into the elegant microbial interactions that may impact host individual and population dynamics .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"ecology",
"virology",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"microbiology",
"microbial",
"ecology",
"agriculture"
] |
2014
|
Densovirus Is a Mutualistic Symbiont of a Global Crop Pest (Helicoverpa armigera) and Protects against a Baculovirus and Bt Biopesticide
|
Centromeres are sites for assembly of the chromosomal structures that mediate faithful segregation at mitosis and meiosis . Plant and animal centromeres are typically located in megabase-sized arrays of tandem satellite repeats , making their precise mapping difficult . However , some rice centromeres are largely embedded in nonsatellite DNA , providing an excellent model to study centromere structure and evolution . We used chromatin immunoprecipitation and 454 sequencing to define the boundaries of nine of the 12 centromeres of rice . Centromere regions from chromosomes 8 and 9 were found to share synteny , most likely reflecting an ancient genome duplication . For four centromeres , we mapped discrete subdomains of binding by the centromeric histone variant CENH3 . These subdomains were depleted in both intact and nonfunctional genes relative to interspersed subdomains lacking CENH3 . The intergenic location of rice centromeric chromatin resembles the situation for human neocentromeres and supports a model of the evolution of centromeres from gene-poor regions .
Centromeres are the essential sites on eukaryotic chromosomes that assemble kinetochores for attachment to spindle microtubules in mitosis and meiosis . Multicellular eukaryotes have “regional” centromeres encompassing hundreds of kilobases that are often located in megabase-sized arrays of 150–180-bp AT-rich tandem repeats known as satellite DNA . Centromeric satellite repeats evolve rapidly and show little sequence conservation even between related species [1] . Despite the frequent occurrence of regional centromeres in satellite arrays , in many eukaryotes centromere location appears to be specified epigenetically [2] by the deposition of specialized nucleosomes containing a centromere-specific variant of histone H3 , often known as CENH3 ( Centromeric H3 ) or as CENP-A , after the mammalian centromeric histone [3] . Evidence for the epigenetic nature of centromeres comes from neocentromeres , the rare sites of spontaneous CENP-A nucleosome deposition described in humans , flies , and barley , that assemble functional kinetochores on DNA sequences that lack any sequence similarity to the normal centromeres [4 , 5] . Experimental overexpression of CENH3 in the fruit fly Drosophila melanogaster likewise produced functional kinetochores on normally noncentromeric sequences [6] . These observations suggest that CENH3 nucleosomes and kinetochores can assemble on any DNA sequence , but in nature , they normally prefer specific satellite sequences . In animal centromeres , CENH3/CENP-A binding is discontinuous , with regions of CENH3/CENP-A nucleosomes interspersed with regions of H3-containing nucleosomes [7] . Similar discontinuous regions of CENP-A binding have been observed in human neocentromeres [8 , 9] and artificial chromosomes [10] . In rice ( Oryza sativa ) , the binding domains for CENH3 ( the National Center for Biotechnology Information [NCBI] Protein accession number AAR85315 ) have been mapped for two centromeres , Cen3 and Cen8 [11–13] . These two domains show evidence of containing both CENH3 and H3 nucleosomes , but previous methods were unable to map distinct CENH3-binding and CENH3-lacking subdomains . Precise mapping of CENH3 nucleosome subdomains is an essential first step in understanding both the structure of centromeres and why CENH3 nucleosomes are assembled and maintained at some locations and not others . Such precise mapping is not possible for most animal and plant centromeres because of highly repetitive satellite sequences , but some rice centromeres have relatively little of the rice-specific centromeric satellite sequence CentO [14] . Here , we use chromatin immunoprecipitation ( ChIP ) with 454 sequencing technology to map nine out of 12 rice centromeres , and to map CENH3-binding subdomains in the four centromeres that have the most complete sequence . We show that CENH3-binding subdomains are depleted in genes relative to adjacent CENH3-lacking subdomains , suggesting a model in which low transcriptional activity is important for the establishment of new centromere sequences and their evolution into mature centromeres .
We previously defined the crossover-suppressed regions for two rice centromeres , Cen3 [13] and Cen8 [12] , which span 3 , 113 kb and 2 , 312 kb , respectively . The CENH3-binding domains in these two centromeres were estimated to span approximately 1 , 810 kb and 750 kb , respectively , and reside within the crossover-suppressed regions . By using the same approach , we have defined the crossover-suppressed regions for the remaining ten centromeres and placed them on physical maps of the centromere regions ( Figure S1 ) . There were 14 physical gaps in the 12 rice centromeric regions of the current rice chromosome pseudomolecules ( http://rice . plantbiology . msu . edu/ ) . The gap in centromere 3 ( Cen3 ) was previously estimated as approximately 450 kb based on fluorescent in situ hybridization ( FISH ) on extended DNA fibers [13] . The other 13 gaps had a combined size of approximately 5 . 9 Mb ( Figure S1 ) . Among them , the size for four gaps ( ∼0 . 54 Mb in total ) was also estimated by FISH and fiber-FISH , including the 111-kb gap in Cen4 [15] , the 69-kb gap in Cen10 [16] , the 310-kb gap in Cen7 , and the 50-kb in Cen11 [17] . The remaining nine gaps , together approximately 5 . 35 Mb , were each sized by optical mapping [18] . Including these physical gaps , the sizes of the crossover-suppressed domains vary between 1 , 447 kb ( Cen10 ) and 5 , 449 kb ( Cen6 ) . Eight of the 12 centromeres have a CentO-containing sequence gap of at least 300 kb . In contrast , the remaining four centromeres contain only a limited amount of the CentO repeat ( ∼60–250 kb ) , including Cen4 [15] and Cen8 [11] , whose CentO arrays have been fully sequenced , and Cen5 and Cen7 , which each have a <100-kb CentO-containing sequence gap . The DNA sequences of these four centromeres provide the foundation for using a high-throughput approach to profile the CENH3 occupancy in these centromeres . A rice anti-CENH3 antibody was used for ChIP , and the immunoprecipitated DNA was sequenced ( ChIP-Seq ) using the 454 sequencing technology . As expected , CentO-related reads were highly overrepresented in the 325 , 298 reads in this ChIP-Seq dataset , with 34 . 7% of the reads matching the CentO consensus for at least 20 bp ( Figure S2 ) . This represents over 17-fold enrichment genome-wide , considering that the approximately 7 Mb of CentO satellite repeats , as estimated by FISH and by optical mapping , nearly all of them present in the centromeric regions , account for less than 2% of the approximately 382-Mb rice genome [14 , 18] . We used the BLAST program to map the remaining 212 , 086 CentO-less reads to the rice genome . To test our expectation that these mapped reads are highly enriched in the centromeric regions , we divided each chromosome into 20-kb windows , positioned every 10 kb , and plotted the read count within each window along its chromosome position . This analysis gave background reads on individual chromosome arms that appeared to be highly uniform , ranging from 7 . 4 to eight reads per 20 kb . These background reads most likely represent technical “noise” in the assay . However , we cannot exclude the possibility that they represent real noncentromeric signal , because CENH3 has been observed to localize at low levels throughout pachytene chromosomes in Arabidopsis thaliana [19] and is known to localize in euchromatin when overexpressed in animals [20] . The analysis also revealed conspicuous peaks within the four best-sequenced centromeres , spanning approximately 820 kb in Cen4 , approximately 630 kb in Cen5 , approximately 420 kb in Cen7 , and approximately 815 kb in Cen8 ( Figure 1A ) . Cen10 had a lower level of read enrichment; however , a region of approximately 610 kb that showed clear CENH3 binding was readily identified . In all five cases , the defined CENH3 domain spanned the entire CentO-containing sequence gap and the assembled major CentO arrays in the pseudomolecules . We did not detect clear peaks in the other seven centromeres ( Figure 1A ) . The presumed repetitiveness of the sequences in the remaining seven centromeres may prevent detection of the expected sequence read enrichment , since reads of repeated sequences cannot be mapped to specific locations . Four of these seven centromeres ( Cen1 , Cen2 , Cen6 , and Cen11 ) had >600-kb CentO repeats , and the other three ( Cen3 , Cen9 , and Cen12 ) had >400-kb CentO repeats . In addition , over 70% of the sequence reads that were not mapped to specific locations due to their high copy number or multiple identical matches had significant hits ( above the cutoff ) in the centromeres . To estimate the read densities of “unmappable” repetitive subregions for these seven centromeres , we made the simplifying assumption that their densities would be the same as immediately flanking “mappable” subregions . Within each 20-kb window , we identified subregions that are either highly repetitive ( ≥50 copies ) or represent identical duplicates in the genome . We then calculated the size of mappable sequences for each 20-kb window by subtracting the length of the above unmappable subregions from 20 kb . For each window , we extrapolated the read count of the mappable sequence over the whole 20-kb window . As a control , we went through the same procedure for approximately 2 Mb of sequences flanking the crossover-suppressed domain on each chromosome arm . This read-density plotting revealed the boundaries for four additional centromeres , including Cen1 ( ∼880 kb ) , Cen3 ( ∼1 , 210 kb ) , Cen9 ( ∼500 kb ) , and Cen12 ( ∼390 kb ) ( Figure 1B ) . However , we were not able to define the boundaries of the CENH3-binding domains for Cen2 ( ∼630 kb CentO ) , Cen6 ( ∼880 kb CentO ) , and Cen11 ( ∼2 Mb CentO ) . Because the CENH3-binding domains in most rice centromeres span 500–1 , 000 kb and all rice centromeres are known to contain CentO [14] , we predict that the CENH3-binding domains of Cen2 , Cen6 , and Cen11 contain mainly unmappable CentO repeats . Finally , we used real-time PCR to verify the CENH3 binding at five of the centromeres , including Cen1 , Cen3 , Cen9 , Cen10 , and Cen12 . Three to five primer pairs ( primer numbers 72–109 , Table S1 ) were designed to target the read-density peaks in the vicinity of each defined border of the CENH3-binding domains ( indicated by arrows in Figure 1B ) . These primer pairs showed 3 . 7-fold to 57 . 4-fold enrichment of ChIPed DNA over the mock control ( p = 0 . 03 – 7 . 8 × 10−7 , one-tailed Student t-test; Figure S3 ) . These data supported our defining of the CENH3 boundary by using the normalized ChIP-Seq profiles . A genome-wide duplication event is thought to have occurred during rice genome evolution because extensive regions of syntenic paralogous genes exist between several chromosome pairs [21 , 22] . If rice centromeres have ancient origins and have stable locations , they may therefore also exist as pairs of paralogs . Synteny of paralogous centromeric regions might be obscured , though , because centromeres are frequent sites for rearrangements [23] . For example , Cen8 has been subject to a rearrangement and positional shift compared with relatives that diverged about 5 and 10 million years ago ( Mya ) [24 , 25] ( Figure 2 ) . To investigate the divergence of paralogous centromeric regions , we sought to identify any possible homologous segments of Cen8 in the rice genome by searching paralogous genes that share significant sequence similarity to the 29 active genes ( 21 with expressed sequence tag [EST]/cDNA and eight supported by reverse-transcriptase PCR ( RT-PCR ) ) located in the 1-Mb sequence spanning Cen8 ( Table 1 ) . We identified apparently functional paralogs for 17 of the 29 genes ( Table S2 ) . Twelve of these 17 Cen8 genes ( Cen8 . t00715 . 1 and Cen8 . t00724 . 1 to Cen8 . t01259 . 1 ) showed 49%–97% protein identity with their respective paralogs ( Figure 2 ) . These 12 pairs of genes are highly conserved in the number of exons and the size of internal exons . Furthermore , five of these 12 Cen8 genes , together spanning a region of 827 kb that covers over 94% of the functional core , best matched another set of five genes scattered over a 544-kb region on the short arm of chromosome 9 ( Loc_Os09g02270 . 1 to Loc_Os09g03090 . 1 , Figure 2 ) . The major portion of this segment on chromosome 9 ( ∼471 kb ) is located within the crossover-suppressed domain of Cen9 and is approximately 1 , 280 kb away from the CentO array of Cen9 . The structural changes between the two duplicated segments are profound , including rearrangements of the conserved five gene pairs , as well as a lack of conservation for the majority of the active genes: 12 out of the 17 genes from chromosome 9 , and 24 out of the 29 genes from Cen8 . To define the boundaries of this duplicated region , we searched for nontransposable element ( non-TE ) genic matches to the rice genome within the 2 , 481 kb of DNA that flanks the short arm ( 1 , 259 kb ) and long arm ( 1 , 222 kb ) sides of the 1-Mb Cen8-spanning region [12] . The active genes from the 1 , 259-kb short-arm side had no matches in the vicinity of the 544-kb syntenic region on chromosome 9 . However , nine genes ( Cen8 . t01272 . 1 to Cen8 . t01718 . 1 ) from the 1 , 222-kb long-arm side had best matched paralogs in a 2 , 394-kb block on the long arm of chromosome 9 , while another 11 Cen8 genes had best-matched paralogs elsewhere ( Loc_Os08g38400 . 1 to Loc_Os01g38100 . 1 , Figure 2 ) . The two syntenic blocks on chromosome 9 are separated by a region of approximately 2 , 487 kb ( including the 302-kb gap ) that contains the functional core of Cen9 . Thus , Cen8 structurally resembles a deletion product derived from the Cen9 region ( Figure 2 ) . These syntenic blocks suggest that centromere function has remained within a region of approximately 3 Mb since the divergence of Cen8 and chromosome 9 , although the actual sequences of the functional centromere core have changed dramatically . Except for paralog Loc_Os09g10620 . 1 , which was not assayed by reverse-transcriptase PCR ( RT-PCR ) , all of the other 19 ( 9 + 11 − 1 ) paralogs for genes from the long-arm side are active genes with cognate ESTs/cDNAs or positive RT-PCR ( primers 22–29 in Table S1 ) . We evaluated the divergence of expression between eight of the 14 syntenic gene pairs by real-time RT-PCR using locus-specific primers ( primers 40–71 in Table S1 ) . RT-PCR results confirmed that these genes are transcribed in all five cDNA samples , including leaves , roots , etiolated leaves/shoots , calli , and panicle ( unpublished data ) , suggesting no difference in tissue specificity for these gene pairs . Nevertheless , real-time RT-PCR assays revealed significant alteration in levels of expression between these gene pairs in one to five cDNA samples ( Figure S4 ) . Genes with relatively higher or lower levels of expression are not biased towards the Cen8 or chromosome 9 segment . For example , gene Cen8 . t00793 . 1 had a significantly higher level of expression than its paralog Loc_Os09g02440 . 1 in three of the tissues; in contrast , gene Cen8 . t00960 . 1 had a markedly reduced level of expression compared with Loc_Os09g03090 . 1 across all the five tissues . To determine whether this duplication is confined to Oryza sativa ( 2n = 24 , AA genome ) or shared by other Oryza species , we compared protein sequences from these 14 pairs of genes to approximately 146 Mb of genome survey sequences in GenBank representing 11%–13% genome coverage from two non-AA species: O . brachyantha ( 2n = 24 , FF genome ) and O . granulata ( 2n = 24 , GG genome ) . For the Cen8 genes , we have identified five putative orthologs in O . brachyantha and three in O . granulata; for the chromosome 9 genes , we have identified five putative orthologs in O . brachyantha and four in O . granulata ( Table S3 ) . Therefore , this duplication predated the approximately 10 Mya of divergence between O . sativa and the most distantly related Oryza species O . granulata [25] . We used the YN00 program available in PAML 4 [26] to obtain an approximate estimation of age for the divergence of Cen8 genes and their chromosome 9 paralogs , using a substitution rate of 6 . 5 × 10−9 per synonymous site per year in rice [27] . These 14 pairs of duplicated genes had an average Ks of 0 . 7717 ± 0 . 1927 ( mean ± standard deviation [SD] ) , corresponding to a divergence time of 59 . 4 ( ±14 . 8 ) Mya ( Table 2 ) . This date is similar to the estimated age of 53–76 Mya for the divergence time of the duplicate chromosomes in a whole-genome duplication event that is hypothesized to have occurred prior to the radiation of grasses [22] . Synteny of this age between chromosome 8 and chromosome 9 was previously described as part of this whole-genome duplication , but these reports did not include the centromere region [21 , 22] . Our previous real-time PCR- and hybridization-based assays revealed that the occupancy of CENH3 in the functional domain of rice Cen8 [11] and Cen3 [13] is not continuous . Nevertheless , the extent of CENH3 binding and its positioning relative to that of the canonical histone H3 remained elusive . In the present study , we observed marked enrichment of ChIP sequence reads within Cen4 , Cen5 , Cen7 , and Cen8 , thus enabling us to profile the CENH3 binding at these four centromeres at a much higher resolution using 1-kb sliding windows ( Figure 3 ) . This plotting revealed two to six major subdomains of CENH3 binding within the four centromeres , ranging in size from a 6-kb subdomain in Cen8 up to a subdomain of approximately 365 kb in Cen7 . Cen4 , Cen5 , and Cen8 shared a similar pattern of CENH3 occupancy: each had five to six major CENH3 subdomains that together occupied 38%–49% of the entire functional core . In contrast , Cen7 had a distinctive CENH3 organization in which its 420-kb functional core was predominantly occupied by the two major CENH3 subdomains of 45 kb and approximately 365 kb , although we can not determine whether CENH3 binding is continuous over the central CentO array . We were unable to identify any sequence motifs that appeared to be putative boundary elements or that otherwise could predict the boundaries of these subdomains . Our previous analyses showed that active genes occur in the functional domains of Cen3 and Cen8 , but that gene density is reduced relative to flanking pericentric regions [12 , 13] . Low gene density might reflect a preferential association of CENH3 nucleosomes with nongenic DNA sequence , or it might simply reflect the random accumulation of transposons and satellites in a crossover-free region; in which case , we would expect CENH3 subdomains to include genes in proportion to the fraction of the total functional domain that these subdomains occupy . However , our earlier data provided only limited information about the distribution of CENH3 nucleosomes relative to genes . We previously identified 16 active genes from the functional core of Cen8 that were enriched for H3K4me2 [12] . CENH3 is sufficiently dissimilar from H3 so that the H3K4me2 antibody should only recognize H3 , but not CENH3 , thus linking these active genes to regions of H3 nucleosomes . To verify this inference , we conducted a comprehensive annotation for transcribed regions from the set of four 1-Mb sequences that span the functional core of Cen4 , Cen5 , Cen7 , and Cen8 . By using over 1 . 2 million publicly available rice ESTs and full-length cDNAs ( fl-cDNAs ) , we identified a total of 108 regions as being transcribed , including 70 regions harboring non–TE-related protein-coding genes ( Tables 1 and S4 ) . Strikingly , only five of these regions were located in the CENH3 subdomains ( Figures 4 and S5; Table 1 ) , including three TE-related genes in Cen8 ( Cen8 . t13171 . 1 , Cen8 . t13349 . 1 , and Cen8 . t13363 . 1 ) with unspliced ESTs . Real-time PCR analysis confirmed that these five genes indeed resided within the CENH3 subdomains ( primers 1 , 3–4 , and 6–12 in Table S1; Figure S6 ) . Excluding gene Cen8 . t01075 . 1 [12] , which overlapped a CENH3 peak in its intron , we used RT-PCR to verify the expression of the remaining four genes across a panel of four different tissues ( primers 1–2 , 4–5 , and 7–10 in Table S1 ) . Three of the four genes were confirmed to be transcribed in at least three different cDNA samples ( Figure S7 ) , suggesting that the CENH3-containing nucleosomes are likely to be compatible with transcription . We further mapped a total of 222 massively parallel signature sequencing ( MPSS ) mRNA tags ( 17 or 20 bp ) to the four centromeres . Most of the tags were mapped to regions corresponding to annotated active genes in the CENH3-lacking subdomains . Only two tags were derived from regions with CENH3 binding , one in Cen4 ( Figure S5 ) and the second one in Cen7 ( Figure 4 ) . Likewise , hypothetical genes solely predicted by gene finders were also significantly biased away from the CENH3-binding subdomains . Out of the 65 genes predicted from these 4-Mb sequences , only six were found to reside within the CENH3 subdomains ( Table 1 ) . Primers were designed for three of these hypothetical genes ( primers 15–19 and 21 in Table S1 ) , and real-time PCR analysis verified their association with CENH3 binding ( Figure S6 ) . Of the six genes , Cen8 . t00761 . 1 was previously known to be inactive in all of the four tissues [12]; we performed RT-PCR on three other genes ( primers 13–14 and 17–20 in Table S1 ) , confirming Loc_Os08g22389 . 1 as the only active gene ( Figure S7 ) . A close examination of these active genes ( supported by ESTs and cDNAs ) within the four 1-Mb centromeric regions revealed two distinct patterns of organization . Cen4 , Cen5 , and Cen8 each had six to 13 active genes within the core , and 11 out of the 14 intervening regions between major CENH3 subdomains each had at least one active gene . However , we did not identify any active gene in the approximately 420-kb core of Cen7 , which showed a much higher proportion of CENH3 occupancy than the other three centromeres . The 18 active genes from the 1-Mb region spanning this centromere were all present in the flanking regions . In addition , of the 287 MPSS small RNA signatures mapped to the four 1-Mb sequences , over 96% matched sequences from the CENH3-lacking subdomains , reflecting a pattern highly consistent with what we have seen for the centromere-derived mRNA and cDNA sequences . Collectively , these data show that genes are almost entirely confined to the CENH3-lacking subdomains in rice centromeres ( Figures 4 and S5 ) . A centromere might evolve from an ordinary genomic region via a neocentromeric intermediate [28] . The lack of genes and transcription in the CENH3 subdomains is consistent with two alternative hypotheses regarding the origin and evolution of rice centromeres . The CENH3 subdomains originally might have contained a similar density of active genes as the flanking CENH3-lacking subdomains , but most of these genes have lost their function and degenerated during centromere evolution . According to this hypothesis , for the genes that were functional prior to the centromere formation , we would be able to identify a nonfunctional gene copy or gene fragment in the CENH3 subdomains of the present-day centromeres . Alternatively , the original CENH3 subdomains might have been quite small and thus contained few active genes . During centromere evolution , individual CENH3 subdomains might have expanded as a result of TE insertion , satellite amplification , or both . To distinguish these possibilities , we looked for instances where active genes from outside of centromeres have homologs in the centromeric regions , which we examined for evidence of deterioration . We first divided the 372-Mb finished sequence of the rice genome into two portions: the 4-Mb sequences corresponding to the four centromeres as discussed above and the remaining 368-Mb DNA . We then compared all the non–TE-related active genes ( with cognate ESTs or cDNAs ) from the 368-Mb DNA to the 4-Mb centromere sequences by Tblastn . We identified eight regions in the CENH3 subdomains that showed sequence similarity to active genes located elsewhere . We manually checked the sequence alignments between these eight centromere regions and their respective homologous genes outside of the centromeres , and found that the centromeric homologs were associated with a premature stop codon ( three genes ) , frame shift ( one gene ) , or truncation of coding sequence ( three genes ) . The presence of a premature stop codon or frame shift is strong evidence for a pseudogene status; for the genes whose parental genes have only a single coding exon , truncation of the CDS also suggests a pseudogene status [29] . Based on this analysis , seven genes can be classified as pseudogenes ( Table S5 ) . Finally , for five of the eight centromeric genes , including Cen7 . t12284 . 1ψ for which no sign of deterioration was detected , we designed primers ( primers 30–39 in Table S1 ) from the putative exons for RT-PCR analysis . None of the primer pairs yielded amplification in any of the four tissues ( unpublished data ) . Therefore , we conclude that all these eight genes from the CENH3 subdomains are pseudogenes . For comparison , we similarly searched for sequence similarity between active genes located outside of the four centromeres and the CENH3-lacking subdomains of Cen8 , and found 40 such regions ( 43 Cen8 regions – 3 in CENH3-binding subdomains; Table S2 ) . Twenty-three of them matched: intergenic regions ( 15 − 3 ) , introns of active genes ( 2 ) , inactive genes ( 6 ) , or noncoding regions ( 3 ) , and all but two of them displayed features of deterioration , resembling those from the CENH3 subdomains . The fact that CENH3 subdomains have far fewer active genes , predicted genes , and pseudogenes than CENH3-lacking subdomains suggests that the regions to which the CENH3 was deposited during centromere formation originally had few genes .
Our genome-wide determination of CENH3 binding has mapped the boundaries of nine of the 12 rice centromeres despite their location predominantly in highly repetitive CentO satellite sequences . Within four of these centromeres , we mapped discontinuous CENH3-binding subdomains ranging in size from 6 kb to approximately 365 kb . Previously , mapping of discontinuous CENH3-binding at the sequence level has only been possible in two human neocentromeres [8 , 9] because of their lack of highly repetitive sequences . The CENH3 subdomains mapped here from four natural centromeres resemble those from neocentromeres in that they vary in size and spacing . Remarkably , these regions are deficient in both genes and pseudogenes compared with adjacent CENH3-lacking subdomains , strongly suggesting that they were gene poor at the time CENH3 became established in these subdomains . DNA sequences in rice such as retrotransposons are typically removed in a few million years through unequal homologous recombination and illegitimate recombination [30 , 31] . We therefore cannot formally exclude the possibility that genes and pseudogenes have been preferentially removed from CENH3-binding subdomains , but we know of no reason to expect that these processes would selectively remove pseudogenes from the CENH3-binding subdomains but not from the interspersed CENH3-lacking subdomains of Cen8 . We therefore think it is more likely that CENH3-binding subdomains are preferentially established in regions depleted of genes . New centromeres have been proposed to arise from neocentromeres that subsequently accumulate satellite sequences [28] . Cen8 is similar to neocentromeres in having a substantial amount of unique sequences , with only approximately 70 kb of the rice centromeric satellite CentO in a 750-kb CENH3-binding region that includes active genes [11 , 12] . Cen8 thus resembles a transition stage between a neocentromere and a centromere composed of satellite repeats [11] . Recently , this view has been challenged based on evidence that the orthologous regions to Cen8 in the related wild rice species O . brachyantha and O . officinalis have an inversion of marker order [24] . These authors suggested that Cen8 was instead derived by recent rearrangement from an ancestral centromere with a typical large satellite array . An ancient satellite array can also be inferred from comparison with the paralogous region that includes Cen9 ( Figure 2 ) and with Cen8 from the closely related rice species O . punctata , in which CENH3 binds to a CentO array estimated to be 1 . 4 Mb [32] . These observations suggest that a conventional satellite array served as the centromere on chromosome 8 from prior to the divergence of grasses 50–70 Mya until after the divergence of closely related rice species approximately 5 Mya . However , the existence of an ancient satellite array that was subsequently lost can be viewed as supporting a neocentromere-like origin of the present rice Cen8 , which occupies largely nonsatellite sequences that presumably did not function as a centromere prior to rearrangement sometime within the last approximately 5 Mya . Cen8 differs from mammalian neocentromeres , however , in being located within a few megabases of the ancestral satellite array and perhaps retaining a small piece of that array , whereas mammalian neocentromeres form with no satellite sequences at many locations that are cytologically distant from the previous centromere . Despite these differences , the similarities between Cen8 and neocentromeres suggest that common processes underlie the formation of both . Like Cen8 , neocentromeres have appeared in gene-poor regions following chromosome rearrangements that delete or disrupt native centromeres . For example , rare human centromere “shifts” are associated with simultaneous deletion of centromeric alpha-satellite [4] . Two neocentromeres in 15q24–26 were found near rearrangement breakpoints [33] , suggesting a possible direct role for rearrangements in both neocentromere and Cen8 formation , perhaps through chromatin remodeling associated with DNA repair . In maize , an inversion with one break in centromeric satellite generated a new centromere adjacent to but not including one portion of the split satellite array [34] . Other neocentromeres have been found in large ( 0 . 8–3 . 9 Mb ) gene-free or gene-poor regions [35 , 36] , and the discontinuous binding of CENP-A in one neocentromere was found in intergenic subregions of an otherwise relatively gene-rich region [9] . These observations echo the gene depletion we see in the CENH3-binding subdomains of four native rice centromeres , and strongly suggest that gene activity , although not completely incompatible with centromere function , is detrimental to neocentromere establishment and centromere maintenance . This is consistent with the facts that regional centromeres in plants and animals typically occupy nongenic satellites , and that the short centromeres of several unicellular eukaryotes also reside in smaller ( 4–18 kb ) gene-free regions [37–39] . Gene transcription results in nucleosome replacement [40] , and transcription disrupts the single centromeric nucleosome of a budding yeast “point” centromere [41] . In Drosophila , CENH3 nucleosomes have been shown to be heterotypic tetramers that wrap only a single turn of DNA , in contrast to octameric H3 nucleosomes [42] , and this may make them more susceptible to displacement by transcription . CENH3 nucleosomes can be normally incorporated into euchromatic regions [19] , and transcription may be important for evicting them so that they can be degraded by proteasomes [43] . Many satellites are also transcribed , but most such transcripts are processed into short interfering RNAs ( siRNAs ) that result in H3 methylation and silencing , keeping satellite transcription at very low levels [32 , 44–46] . Regions of low gene activity may therefore favor retention of enough CENH3 nucleosomes to establish and maintain kinetochore function . Neocentromeres are smaller and have less CENP-A than normal human centromeres [47] , suggesting that they must expand if they are to become successful mature centromeres . The Cen8 region has expanded by segmental duplication [48] and by insertion of retrotransposons , which comprise nearly a third of Cen8 [12] and which are enriched relative to the orthologous region of O . brachyantha [24] . New retrotransposons are likely to become silenced and methylated on histone H3K9 , which is incompatible with kinetochore function [10] . However , once retrotransposons mutate to nonfunctionality , silencing may be lost and degenerating retrotransposons may contribute to the growth of nongenic CENH3-binding regions , and hence to expanded centromere variants that become rapidly fixed through centromere competition in female meiosis [1] . Regional centromeres that consist of unique sequences like neocentromeres and Cen8 appear to be short-lived , since most regional centromeres are composed of satellite repeats . Eventual acquisition of the 150–180-bp satellites that underlie both CENH3 and H3 nucleosomes in most regional centromeres may be favored because of their ability to order nucleosomes into regular arrays [49] . The occurrence of 180-bp satellites in classical maize neocentromeres ( knobs ) , which lack CENH3 [50] , strongly suggests that satellites are favored for their ability to organize H3 octameric nucleosomes . A densely packed rigid structure of H3 nucleosomes imposed by satellites [51] may help to orient and distribute the anaphase forces of the 10–100 microtubules that are bound to centromeric chromatin in a typical regional centromere [52] . Since satellite arrays also have low transcriptional activity and are readily expandable , they may provide the same advantages for centromere function as gene-depleted unique sequences while also ordering nucleosomes . The exact mechanisms by which satellite arrays can expand and become homogenized are not well understood , but they probably include unequal crossover and gene conversion events that yield new centromere variants at an accelerated frequency . Competition between these variants in female meiosis [1] may underlie rapid replacement of one type of satellite with another , as has been observed between wild rice species diverged from O . sativa by approximately 5 and 10 Mya [53] . Satellites acquired by neocentromeres or other unique-sequence centromeres like Cen8 may eventually come to dominate these centromeres , thereby converting them into mature regional centromeres .
ChIP was performed as previously described [11] . Approximately 50 g of rice leaf tissue were used in ChIP to obtain approximately 3 μg of immunoprecipitated DNA for sequencing . The ChIP DNA was subjected to modified sequencing pipeline on Roche's GS 20 sequencing platform [54] after the concentration and the size distribution of the ChIP DNA were accurately measured by Agilent's Bioanalyzer and Invitrogen's Qubit fluorometer . DNA fragmentation step was omitted for Roche's library construction based on the size of the ChIP DNA . DNA sample was end-repaired and phosphorylated before being ligated to Roche's genomic sequencing adaptors . Single-stranded library fragments were collected by following the manufacturer's instruction . The concentration and the size distribution of the single-stranded library was measured by Agilent's Bioanalyzer and appropriate amount of the library was mixed and hybridized with Roche's DNA capture beads . DNA bound beads were amplified and enriched and then were loaded onto GS 20 platform according to modified manufacturer's instruction . We generated a total of 325 , 298 high quality reads . The majority of the sequence reads ( 94% ) had a length between 80 and 130 bp . We first used Blastn to find CentO-containing sequences from all ChIP 454 sequence reads . The rice variety Nipponbare pseudomolecules ( http://rice . plantbiology . msu . edu/ , version 5 ) contained 6 , 003 hits that aligned with CentO consensus over at least 20 bp , only 24 of which were present outside of the genetically defined centromeres . Therefore , we defined the CentO-containing reads as those that had a minimum of 20-bp alignment with CentO . We then used Megablast to map the remaining CentO-less reads to the version 5 pseudomolecules , at the cutoff of ≥97% identity and ≥90% coverage of the read length . We found that 4 . 3% of the reads had no hits or had hits below the cutoff , and were excluded from further analysis . Of the remainder , we did not assign map locations for 46 , 436 reads that either had two or more hits with the same identity and coverage , or had 50 or more copies but lacked a perfect match . By retaining only the best hit for reads with fewer than 50 copies and the perfect match for those with a higher copy number , we were able to assign map locations for 151 , 810 of the reads , accounting for 46 . 7% of the total dataset ( Figure S2 ) . To plot the distribution of 454 reads , we split each chromosome into 20-kb windows , spaced every 10 kb , and plotted the number of reads mapped to each window along its chromosome coordinate . Centromeres have a much denser distribution of repetitive DNA than chromosome arms , so reads derived from centromeric regions are less likely to be mapped to distinct locations . To address this mapping bias , we screened for less repetitive , mappable sequences for each window and used the read count over these mappable sequences to calculate a read density representing the relative enrichment . For each chromosome , we extracted a portion of the CentO-masked pseudomolecule that covers the entire genetically defined centromere and extends approximately 2 Mb into the flanking region on each side . The extracted sequences were split into 60-bp bins , spaced every 30 bp , which were blasted against the whole rice genome . For each bin , we retained hits with alignment of ≥30 bp and calculated its copy number . Overlapping bins with less than 50 copies were concatenated . We further tried to identify regions that have at least one duplicate of 100% identity elsewhere in the genome . We parsed out bins that had alignments of ≥30 bp of 100% identity to more than one location , and overlapping bins were concatenated . We aligned them back to the genome to identify those that have at least one perfect duplicate match of ≥130 bp , since over 98% of the reads are <130 bp in length . Regions that had fewer than 50 genomic hits but lacked a perfect duplicate were merged with the locations anchored by the mapped ChIP-Seq to form the mappable regions . We estimated the read density , which represents the number of reads in a 20-kb mappable region , using the above 20-kb windows according to the following formula: Read density = ( [Number of reads − 10]/mappable sequences in kilobases ) × 20 kb , where 10 is the 75th percentile for the number of reads in the 20-kb windows from both chromosome arms , which are the same for all chromosomes . We also plotted the read count using 1-kb window size for chromosome 4 , chromosome 5 , chromosome 7 , and chromosome 8 . To identify a subset of 1-kb windows that are enriched for ChIP-Seq reads , for each chromosome , we generated expectation distributions of read counts under the null hypothesis of no enrichment . This was accomplished by randomly assigning each mapped ChIP-Seq read to a unique chromosome position ten times and calculating the nominal p-values for each 1-kb window . Enriched 1-kb windows were defined as those that had a p-value of ≤1 × 10−4 ( chromosome 5 and chromosome 7 ) or 2 × 10−4 ( chromosome 4 and chromosome 8 ) , corresponding to five read counts . We used a total of 32 , 129 full-length cDNAs ( http://cdna01 . dna . affrc . go . jp/cDNA ) and over 1 . 2 million ESTs ( GenBank dbEST ) from rice to identify regions of transcription from four 1-Mb centromeric regions according to the published procedure [12] . These four 1-Mb regions span the cores of Cen4 , Cen5 , Cen7 , and Cen8 as revealed by the current study . Transcribed regions were further broken down into non–TE-related protein-coding genes , TE-related genes , and noncoding regions . Additional regions of transcription were identified by using rice MPSS mRNA and small RNA signatures [55] as described [13] , where only signatures with a single perfect match in the genome ( to a given centromere ) were retained . Finally , predicted genes without EST/cDNA evidence for transcription were downloaded from the rice genome annotation project database ( http://rice . plantbiology . msu . edu ) , and the expression of those present in Cen8 was tested by RT-PCR , either previously [12] or as part of this study . We used Tblastn to compare the above four 1-Mb centromere sequences to a total of 32 , 816 putative non–TE-related protein sequences representing 23 , 570 active genes from the rest of the genome . By this comparison , we tried to identify whether any of the active genes in the centromeres has a functional copy elsewhere in the genome . We also wanted to understand whether there are major differences between the CENH3 subdomains and the CENH3-lacking subdomains regarding the distribution of decayed genes or gene fragments ( pseudogene ) whose parent genes are outside of centromeres . For a match to be considered significant , we required an identity of ≥35% , a coverage of ≥50% of the parent gene , and an alignment length of 50 or more amino acids ( aa ) , or a coverage of <50% but an alignment length of ≥100 aa . The alignment output between the genomic sequences from the centromeres and protein sequences from active genes elsewhere were manually checked for instances of gene deterioration . By comparison to the parent gene , the homologous region from the centromere was classified as representing a pseudogene if any of the following features of deterioration were detected: a premature stop codon , a frame shift caused by insertion or deletion , or a 3′ or 5′ truncation of the coding sequence . In addition , regions confirmed to be not expressed by RT-PCR were also annotated as pseudogenes . If the homologous region from the centromere was already annotated as a gene , a further comparison was made to reveal any changes in gene structure . We used RT-PCR to test the expression of genes present in the CENH3 ChIP-Seq peaks , including four genes with unspliced ESTs ( primers 1–2 , 4–5 , and 7–10 ) , three hypothetical genes ( primers 13–14 and 17–20 ) , and five pseudogenes ( primers 30–39 ) , as well as another four hypothetical genes ( primers 22–29 ) that best matched Cen8 active genes . Primers were designed to have a length between 21 and 29 bp , with an annealing temperature from 60 . 1 to 66 . 7 °C ( Table S1 ) . We isolated total RNA from four different rice tissues or treatments , including leaves , roots , etiolated leaves/shoots , and calli , and performed RT-PCR as described [12] . We designed ChIP-PCR primers to verify the CENH3 binding in 19 predicted ChIP-Seq peaks from five centromeres ( Figure S3 ) and eight genes also in the ChIP-Seq peaks ( Figure S6 ) . Real-time PCR analysis was used to determine the relative enrichment of CENH3-associated sequences in the bound fraction over the mock control . PCR reactions were carried out in triplicates using the DyNAmo HS SYBR Green qPCR kit ( MJ Research ) and run at 95 °C for 15 min , followed by 45 cycles of 95 °C for 10 s , 62–65 °C for 30 s , and 72 °C for 30 s . We used three active genes as negative controls that yielded similar results , including Cen8 . t00421 . 1 and Cen8 . t00479 . 1 [12] , as well as LOC_Os01g57730 . 1 . For each primer pair , we calculated the relative fold enrichment ( RFE ) as described [12] , using Cen8 . t00421 . 1 as the reference gene . We used real-time RT-PCR to compare the transcription level between eight pairs of duplicated genes from Cen8 and chromosome 9 . For each gene pair , locus-specific primers were designed based on the alignment between the coding sequences from both copies , ensuring that each Cen8-locus primer must have at least five base pairs difference from the corresponding chromosome 9-locus primer , both with similar annealing temperatures ( Table S1 ) . Real-time RT-PCR was conducted following the above procedure , where the previously described four cDNA samples were tested , plus an additional cDNA sample from 3-d-old panicles . We used real-time PCR on genomic DNA to evaluate whether there is a difference in the amplification efficiency between the primer pair targeting the Cen8 copy and that targeting the chromosome 9 copy . For each gene pair , we obtained the PCR cycle threshold ( CT ) difference from the genomic DNA template , expressed as ΔCT-gDNA = CT ( Cen8 copy ) − CT ( chromosome 9 copy ) , and used this difference to normalize individual ΔCT-cDNA ( CT difference from each cDNA sample ) . A relative fold change of a Cen8 copy over a chromosome 9 copy in expression was calculated as 2–ΔΔCT , where ΔΔCT = ΔCT-cDNA − ΔCT-gDNA . We performed a two-tailed Student t-test of CT values at the significant level of α = 0 . 05 . For each pair of duplicated genes between Cen8 and chromosome 9 , protein sequences were aligned using the National Center for Biotechnology Information bl2seq , and the resulting alignment was manually checked . The corresponding coding sequences were then aligned using ClustalW ( http://www . ebi . ac . uk/Tools/clustalw2/index . html ) , with the protein alignment as the guide . We used the YN00 program from the PAML package [26] to estimate the synonymous ( Ks ) and nonsynonymous ( Ka ) substitution rates . We averaged the Ks and Ka values from three different counting methods , including LPB93 , LWL85 , and LWL85m that all account for transition–transversion rate difference [26] . This mean Ks was used to estimate the divergence time between each pair of duplicated genes , using the formula T = Ks/2 × 6 . 5 × 10−9 , where 6 . 5 × 10−9 represents the substitution rate per synonymous site per year in grass [27] . Finally , the divergence time between the Cen8 and chromosome 9 syntenic block was estimated using the average Ks from all 14 pairs of genes .
|
Before a cell divides , its chromosomes must be duplicated and then separated to provide each daughter cell with an identical genome copy . To accomplish this separation , the cell-division apparatus attaches to structures on the chromosomes called centromeres . Most plant and animal centromeres contain highly repetitive DNA sequences and specific proteins such as CENH3; however , it is not known which of the many repeats bind CENH3 . Some rice centromeres , however , consist largely of single-copy DNA , providing a tractable model for investigating CENH3-binding patterns . Using modern DNA sequencing technology and an antibody to CENH3 , we were able to find which sequences in the rice genome are bound by CENH3 . We uncovered evidence that one centromere , Cen8 , which has lost much of its repetitive content through a rearrangement within the last approximately 5 million years , is derived from a highly repetitive centromeric region that was duplicated along with the rest of the genome 50–70 million years ago . We also found that CENH3 is bound discontinuously in centromeric subdomains that have fewer genes than subdomains lacking CENH3 . These results suggest , not only that centromeres evolve in gene-poor regions , but also how centromeres might evolve from single-copy to repetitive sequences .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"plant",
"biology",
"evolutionary",
"biology",
"molecular",
"biology",
"genetics",
"and",
"genomics"
] |
2008
|
Intergenic Locations of Rice Centromeric Chromatin
|
Apolipoprotein B ( ApoB ) and ApoE have been shown to participate in the particle formation and the tissue tropism of hepatitis C virus ( HCV ) , but their precise roles remain uncertain . Here we show that amphipathic α-helices in the apolipoproteins participate in the HCV particle formation by using zinc finger nucleases-mediated apolipoprotein B ( ApoB ) and/or ApoE gene knockout Huh7 cells . Although Huh7 cells deficient in either ApoB or ApoE gene exhibited slight reduction of particles formation , knockout of both ApoB and ApoE genes in Huh7 ( DKO ) cells severely impaired the formation of infectious HCV particles , suggesting that ApoB and ApoE have redundant roles in the formation of infectious HCV particles . cDNA microarray analyses revealed that ApoB and ApoE are dominantly expressed in Huh7 cells , in contrast to the high level expression of all of the exchangeable apolipoproteins , including ApoA1 , ApoA2 , ApoC1 , ApoC2 and ApoC3 in human liver tissues . The exogenous expression of not only ApoE , but also other exchangeable apolipoproteins rescued the infectious particle formation of HCV in DKO cells . In addition , expression of these apolipoproteins facilitated the formation of infectious particles of genotype 1b and 3a chimeric viruses . Furthermore , expression of amphipathic α-helices in the exchangeable apolipoproteins facilitated the particle formation in DKO cells through an interaction with viral particles . These results suggest that amphipathic α-helices in the exchangeable apolipoproteins play crucial roles in the infectious particle formation of HCV and provide clues to the understanding of life cycle of HCV and the development of novel anti-HCV therapeutics targeting for viral assembly .
More than 160 million individuals worldwide are infected with hepatitis C virus ( HCV ) , and cirrhosis and hepatocellular carcinoma induced by HCV infection are life-threatening diseases [1] . Current standard therapy combining peg-interferon ( IFN ) , ribavirin ( RBV ) and a protease inhibitor has achieved a sustained virological response ( SVR ) in over 80% of individuals infected with HCV genotype 1 [2] . In addition , many antiviral agents targeting non-structural proteins and host factors involved in HCV replication have been applied in clinical trials [3] , [4] . In vitro systems have been developed for the study of HCV infection and have revealed many details of the life cycle of HCV . By using pseudotype particles bearing HCV envelope proteins and RNA replicon systems , many host factors required for entry and RNA replication have been identified , respectively [5] , [6] . In addition , development of a robust in vitro propagation system of HCV based on the genotype 2a JFH1 strain ( HCVcc ) has gradually clarified the mechanism of assembly of HCV particles [7] , [8] . It has been shown that the interaction of NS2 protein with structural and non-structural proteins facilitates assembly of the viral capsid and formation of infectious particles at the connection site between the ER membrane and the surface of lipid droplets ( LD ) [9] . On the other hand , very low density lipoprotein ( VLDL ) associated proteins , including apolipoprotein B ( ApoB ) , ApoE , and microsomal triglyceride transfer protein ( MTTP ) , have been shown to play crucial roles in the formation of infectious HCV particles [10]–[12] . Generally , ApoA , ApoB , ApoC and ApoE bind the surface of lipoprotein through the interaction between amphipathic α-helices and ER-derived membrane [13] , [14] . This binding of apolipoproteins enhances the stability and hydrophilicity of lipoprotein . However , the specific roles played by the apolipoproteins in HCV particle formation are controversial . Gastaminza et al . demonstrated that ApoB and MTTP are cellular factors essential for an efficient assembly of infectious HCV particles [10] . However , studies by other groups demonstrated that ApoE is a major determinant of the infectivity and particle formation of HCV , and the ApoE fraction is highly enriched with infectious particles [11] . In addition , Mancone et al . showed that ApoA1 is required for production of infectious particles of HCV [15] . However , the evidence of the involvement of apolipoproteins in HCV particle formation is dependent on knockdown data and exogenous expression of the apolipoproteins , and thus the precise mechanisms of participation of the apolipoproteins in HCV assembly have not been elucidated [10] , [11] , [16] . Recently , several novel genome editing techniques have been developed , including methods using zinc finger nucleases ( ZFN ) , transcription activator like-effector nucleases ( TALEN ) and CRISPR/Cas9 systems [17]–[19] . DNA double strand breaks ( DSBs ) induced by these artificial nucleases can be repaired by error-prone non-homologous end joining ( NHEJ ) , resulting in mutant mice or cell lines carrying deletions , insertions , or substitutions at the cut site . To clarify the detailed function of gene family with redundant functions , the generation of animals or cell lines carrying multiple mutated genes may be essential . In this study , Huh7 cell lines deficient in both ApoB and ApoE genes were established by using ZFNs and revealed that ApoB and ApoE redundantly participate in the formation of infectious HCV particles . Interestingly , the expression of other exchangeable apolipoproteins , i . e . , ApoA1 , ApoA2 , ApoC1 , ApoC2 and ApoC3 , facilitated HCV assembly in ApoB and ApoE double-knockout cells . In addition , the expression of amphipathic α-helices in the exchangeable apolipoproteins restored the production of infectious particles in the double-knockout cells through an interaction with viral particles .
First , we compared expression levels of apolipoproteins between hepatocyte and hepatic cancer cell lines including Huh7 and HepG2 cells ( Fig . 1A and B ) . The web-based search engine NextBio ( NextBio , Santa Clara , CA ) revealed that ApoB , ApoH and the exchangeable apolipoproteins ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 , and ApoE are highly expressed in human liver tissues ( Fig . 1A ) . On the other hand , the expressions of ApoA1 , ApoC1 , ApoC2 , ApoC3 and ApoH in hepatic cancer cell lines were suppressed compared to those in hepatocytes ( Fig . 1B ) . To examine the roles of apolipoproteins in the formation of infectious HCV particles , the effects of knockdown of ApoA2 , ApoB and ApoE on the infectious particle production in the supernatants were determined in Huh7 cells by focus forming assay ( Fig . 1C ) . The transfection of siRNAs targeting to ApoA2 , ApoB and ApoE significantly suppressed the production of infectious HCV particles . This inhibitory effect is well consistent with the high level of expression of these apolipoproteins in the hepatic cancer cell lines , suggesting that the apolipoproteins involved in HCV assembly are dependent on the expression pattern in hepatic cancer cell lines , including Huh7 cells [20] . Therefore , we examined the effects of exogenous expression of the apolipoproteins highly expressed in the liver tissues on the infection of HCV in the stable ApoE-knockdown Huh7 cells ( Fig . 1D ) . In contrast to the control-knockdown cells , expression of not only ApoE but also ApoA1 , ApoA2 , and ApoC1 rescued the infectious particle formation in the ApoE-knockdown cells ( Fig . 1E ) , suggesting that various exchangeable apolipoproteins participate in the efficient production of infectious HCV particles . To obtain more convincing data on the involvement of apolipoproteins in the production of infectious HCV particles , we established knockout ( KO ) Huh7 cells deficient in either ApoB ( B-KO1 and B-KO2 ) or ApoE ( E-KO1 and E-KO2 ) by using ZFN ( Figure S1 ) . Deficiencies of ApoB or ApoE expression in these cell lines were confirmed by ELISA and immunoblotting analyses ( Figure S1 ) . First , we examined the roles of ApoB and ApoE on the entry and RNA replication of HCV by using HCV pseudotype particles ( HCVpp ) and subgenomic replicon ( SGR ) of the JFH1 strain , respectively . The B-KO and E-KO cell lines exhibited no significant effect on the infectivity of HCVpp and the colony formation of SGR ( Figure S2A and Figure S2B ) , suggesting that ApoB and ApoE are not involved in the entry and replication processes of HCV . To examine the role of ApoB and ApoE in the propagation of HCV , HCVcc was inoculated into parental , B-KO and E-KO cell lines at an MOI of 1 , and intracellular viral RNA and infectious titers in the supernatants were determined ( Figure S2C and Figure S2D ) . Although RNA replication and infectious particle formation in B-KO cells upon infection with HCV were comparable with those in parental Huh7 cells , E-KO cells exhibited slight reduction of particle formation , and the expression of ApoE in E-KO cells rescued infectious particle formation ( Figure S2C , Figure S2D , Figure S2E ) . Next , to examine the redundant role of ApoB , the effect of knockdown of ApoB on HCV assembly was determined in parental and E-KO Huh7 cell lines ( Fig . 2A ) . Knockdown of ApoB in E-KO cells resulted in a more efficient reduction of infectious particle production than that in parental Huh7 cells , suggesting that ApoB and ApoE have a redundant role in the formation of infectious HCV particles . To further confirm the redundant role of ApoB and ApoE in the HCV life cycle , especially in the particle formation , 2 clones of ApoB and ApoE double-knockout ( BE-KO1 and BE-KO2 ) Huh7 cells were established by ZFNs ( Figure S3A and Figure S3B ) . The lack of ApoB and ApoE expressions was confirmed by immunoblotting and ELISA analyses ( Figure S3C , Figure S3D , Figure S3E ) . The BE-KO cell lines also exhibited no significant effect on the infectivity of HCVpp ( Fig . 2B ) and the colony formation of SGR ( Fig . 2C ) . Next , we examined the redundant role of ApoB and ApoE on the propagation of HCVcc . Upon infection with HCVcc at an MOI of 1 , infectious titers in the supernatants of BE-KO1 and BE-KO2 cells were 50 to 100 times lower than those of parental Huh7 cells at 72 h post-infection , while the level of intracellular RNA replication was comparable ( Fig . 2D and E ) . In addition , exogenous expression of ApoE in BE-KO ( ApoE-res ) cells rescued the production of infectious viral particles to levels comparable to those in parental Huh7 cells ( Fig . 2F and G ) , suggesting that ApoB and ApoE redundantly participate in the particle formation of HCV . It is difficult to determine the roles of ApoB in the particle formation of HCV , because ApoB is too large ( 550 kDa ) to obtain cDNA for expression . However , previous reports have shown that expression of MTTP facilitates the secretion of ApoB [21] . To further clarify the roles of ApoB in the life cycle of HCV , we established knockout Huh7 cell lines deficient in MTTP ( M-KO1 and M-KO2 ) and in both ApoE and MTTP ( EM-KO1 and EM-KO2 ) by using the ZFN and CRISPR/Cas9 system ( Figure S4A and Figure S4E ) . The lack of MTTP , ApoB and ApoE expressions was confirmed by immunoblotting and ELISA analyses ( Figure Figure S4B , Figure S4C , Figure S4D , Figure S4F , Figure S4G , Figure S4H ) . As previously reported , the secretion of ApoB was completely abrogated in M-KO and EM-KO cells , while the mRNA levels of ApoB were comparable among Huh7 , M-KO and EM-KO cells ( Figure S4I ) . To examine the roles of MTTP in the assembly of HCV through the secretion of ApoB , HCVcc was inoculated into the Huh7 , B-KO , M-KO , E-KO , BE-KO and EM-KO cell lines at an MOI of 1 , and intracellular HCV genomes and infectious titers in the supernatants were determined ( Fig . 3A–C ) . Although intracellular RNA replication in M-KO and EM-KO cells was comparable with that in Huh7 , B-KO , E-KO and BE-KO cells ( Fig . 3B ) , infectious titers in the supernatants of EM-KO cells were severely impaired as seen in BE-KO cells , while those of M-KO cells were comparable to those of parental Huh7cells ( Fig . 3C ) , suggesting that MTTP participates in the HCV assembly through the regulation of ApoB secretion . To further confirm the roles of MTTP in HCV assembly through ApoB secretion , the effects of exogenous expression of MTTP in EM-KO cells on the infectious particle formation of HCV were determined . Immunoblotting and ELISA analyses revealed that exogenous expression of MTTP rescued the secretion of ApoB into the supernatants of EM-KO cells ( Fig . 3D and E ) , while expression of ApoE or MTTP in both BE-KO and EM-KO cells exhibited no effect on the intracellular RNA replication ( Fig . 3F ) . Although exogenous expression of ApoE rescued the infectious particle formation of HCV in both BE-KO and EM-KO cells , expression of MTTP rescued the particle formation in EM-KO cells but not in BE-KO cells ( Fig . 3G ) , supporting the notion that MTTP plays a crucial role in the HCV assembly through the maturation of ApoB . Next , to examine the roles played in HCV particles formation by other apolipoproteins highly expressed in the liver ( Fig . 1A ) , the expressions of ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 and ApoH in BE-KO1 cells were suppressed by siRNAs ( Fig . 4A and Figure S5 ) . While knockdown of ApoA1 , ApoC3 and ApoH exhibited no effect , that of ApoA2 , ApoC1 and ApoC2 significantly inhibited the release of infectious particles , which was consistent with the expression pattern of endogenous apolipoproteins except for ApoH in Huh7 cells ( Fig . 1B ) , suggesting that not only ApoB and ApoE but also other exchangeable apolipoproteins participate in HCV particle formation . To confirm the redundant role of these apolipoproteins on the infectious particle formation , the effects of exogenous expression of these apolipoproteins on the propagation of HCVcc in BE-KO1 cells were determined . ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 , ApoE and ApoH were expressed by lentiviral vector in BE-KO1 cells ( Fig . 4B upper panel ) . The expressions of ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 and ApoE but not of ApoH enhanced extracellular HCV RNA , while they exhibited no effect on intracellular HCV RNA ( Fig . 4C ) . In addition , the expressions of these exchangeable apolipoproteins enhanced the infectious particle formation in the supernatants of BE-KO1 cells ( Fig . 4B lower panel ) . On the other hand , the expression of nonhepatic apolipoproteins , including ApoD , ApoL1 , and ApoO , exhibited no effect on HCV particle formation in BE-KO1 cells ( Figure S6 ) . These results suggest that exogenous expression of not only the ApoE but also the ApoA and ApoC families can compensate for the impairment of HCV particle formation in BE-KO1 cells . Interestingly , specific infectivity ( infectious titers/viral RNA levels in supernatants ) was also enhanced by the expression of ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 and ApoE , suggesting that these apolipoproteins participate in the infectious but not non-infectious particle formation of HCV ( Fig . 4D ) . Previous reports have suggested that the expressions of Claudin1 ( CLDN1 ) , miR-122 and ApoE facilitate the production of infectious particles in nonhepatic 293T cells [16] . Therefore , the effects of exogenous expression of exchangeable apolipoproteins on particle formation were examined in 293T cells expressing CLDN1 and miR-122 ( 293T-CLDN/miR-122 cells ) . Exogenous expression of ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 and ApoE , but not of ApoH by lentiviral vector facilitated the production of infectious HCV particles in 293T-CLDN/miR-122 cells ( Fig . 4E ) . On the other hand , the expression of ApoE exhibited no effect on the propagation of Japanese encephalitis virus ( JEV ) and dengue virus ( DENV ) ( Figure S7 ) in BE-KO1 cells . These results suggest that the exchangeable apolipoproteins and ApoB redundantly and specifically participate in the formation of HCV particles . To examine the role of exchangeable apolipoproteins in the formation of other genotypes of HCV , the effect of exogenous expression of these apolipoproteins on the propagation of genotype 1b and 3a chimeric HCVcc , TH/JFH1 and S310/JFH1 viruses in BE-KO1 cells was determined ( Fig . 5 ) [22] , [23] . As seen in infection with HCVcc ( JFH1 ) , expression of ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 and ApoE enhanced the formation of infectious particles of TH/JFH1 and S310/JFH1 chimeric viruses . These results suggest that ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 and ApoE redundantly participate in the efficient formation of infectious HCV particles of genotypes 1b , 2a and 3a . To determine the details of the assembly of infectious HCV particles in the BE-KO1 cells , intracellular infectious titers were determined in Huh7 , BE-KO1 and ApoE-res cells by using the freeze and thaw method . Not only intracellular but also extracellular infection titers were impaired in BE-KO1 cells compared with those in parental and ApoE-res cells ( Fig . 6A ) , suggesting that intracellular particle formation is impaired by deficiencies in the expression of ApoB and ApoE . Previous reports have shown that the recruitment of viral proteins around LD and redistribution of LD are essential for HCV assembly [24] . To clarify the roles of the exchangeable apolipoproteins on HCV assembly in more detail , we examined the intracellular localization of viral proteins , LD and ER in BE-KO1 and ApoE-res cells . The localization of core proteins around LD and the membranous-web structure forming the replication complex were observed in BE-KO1 cells upon infection with HCVcc , as reported in parental Huh7 cells ( Fig . 6B , 6C and Figure S8 ) . However , greater accumulation of core proteins and LD around the perinuclear region was detected in BE-KO1 cells in comparison with ApoE-res cells ( Fig . 6C and 6D ) , supporting the notion that apolipoproteins participate in the infectious particle formation in HCV rather than viral RNA replication . Previous studies revealed that core proteins were mainly localized on the ER membrane upon infection with the genotype 2a Jc1 strain-based HCVcc ( HCVcc/Jc1 ) , and inhibition of capsid assembly and envelopment caused accumulation of core proteins on the surface of LD [25]–[27] . In ApoE-res cells , core proteins of HCVcc/Jc1 were mainly localized on the ER membrane , in contrast to the co-localization of core proteins of HCVcc ( JFH1 ) with LD ( Fig . 6E upper ) . However , core proteins were accumulated around LD in BE-KO1 cells infected with HCVcc/Jc1 , as seen in those infected with HCVcc ( JFH1 ) ( Fig . 6E lower ) . These results suggest that apolipoproteins participate in the steps of HCV particle formation occurring after HCV protein assembly on the LD . To further examine the involvement of apolipoproteins in the infectious particle formation of HCV , culture supernatants and cell lysates of BE-KO1 and ApoE-res cells infected with HCVcc were analyzed by buoyant density ultracentrifugation ( Fig . 7A–B ) [28] . Secretion of viral capsids in the supernatants was severely impaired in BE-KO1 cells in comparison with that in ApoE-res cells ( Fig . 7A upper ) , in contrast to the detection of abundant intracellular capsids in both cell lines ( Fig . 7B upper ) . Although peak levels of the core proteins and infectious titers were detected around 1 . 08 g/ml in both cell lines , the infectious titers in all fractions of BE-KO1 cells were significantly lower than those in ApoE-res cells , supporting the notion that apolipoproteins participate in the post-assembly process of HCV capsids which is required to confer infectivity . Next , to examine the involvement of apolipoproteins in the envelopment of HCV particles , lysates of BE-KO1 and ApoE-res cells infected with HCVcc were treated with proteinase K in the presence or absence of Triton X [26] . Protection of HCV core proteins from the protease digestion was observed in both cell lysates ( Fig . 7C ) , suggesting that apolipoproteins are not involved in the envelopment of HCV particles . Collectively , these results suggest that exchangeable apolipoproteins participate in the post-envelopment step of HCV particle formation . To determine the structural relevance of apolipoproteins involved in the HCV assembly , the secondary structures of the apolipoproteins were deduced by using a CLC Genomics Workbench and previous reports ( Fig . 8A ) [29]–[34] . Tandem repeats of amphipathic α-helices were observed in the apolipoproteins capable of rescuing HCV assembly in BE-KO1 cells , but not in those lacking this activity , suggesting that amphipathic α-helices in the apolipoproteins participate in the assembly of HCV . To examine the involvement of the amphipathic α-helices of the exchangeable apolipoproteins in the particle formation of HCV , we constructed expression plasmids encoding deletion mutants of ApoE and ApoC1 , and then these deletion mutants were exogenously expressed in BE-KO1 cells by lentiviral vectors ( Fig . 8B and C upper panels ) . The expression of all of the deletion mutants of ApoE and ApoC1 containing either N-terminal or C-terminal amphipathic α-helices rescued the particle formation of HCV in BE-KO1 cells ( Fig . 8B and C lower panels ) , suggesting that amphipathic α-helices in the apolipoproteins play crucial roles in the production of infectious HCV particles . In addition , more abundant full-length and truncated ApoE were detected in the precipitates of the culture supernatants of cells infected with HCVcc than those of mock-infected cells concentrated by ultracentrifugation , suggesting that the amphipathic α-helices of apolipoproteins are directly associated with HCV particles ( Fig . 8D and E ) . Taken together , the data in this study strongly suggest that exchangeable apolipoproteins redundantly participate in the infectious particle formation of HCV through the interaction between amphipathic α-helices and viral particles .
In this study , we demonstrated the redundant roles of ApoB and the exchangeable apolipoproteins ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 and ApoE in the assembly of infectious HCV particles . The deficiencies of both ApoB and ApoE inhibited the production of infectious HCV particles in Huh7 cells , and exogenous expression of exchangeable apolipoproteins rescued the particle formation . cDNA microarray revealed that the expression patterns of exchangeable apolipoproteins in hepatic cancer cell lines are widely different from those in liver tissue . In previous reports , ApoE and ApoB were identified as important host factors for the assembly of infectious HCV particles [10] , [11] , and knockdown of ApoE and ApoB expression also inhibited the production of infectious particles in this study . Because ApoB and ApoE are major apolipoproteins in VLDL , several reports have suggested that the VLDL production machinery participates in the production of HCV particles . Furthermore , density gradient analyses revealed co-fractionation of HCV RNA with lipoproteins , with the resulting complexes being termed lipoviroparticles ( LVP ) [12] , [35] . However , it has been reported that there is no correlation between secretion of VLDL and production of LVP [36] . In addition , exogenous expression of ApoE facilitated the infectious particle formation of HCV in 293T cells stably expressing CLDN1 and miR-122 [16] , suggesting that ApoE-mediated particle formation is independent from VLDL production . Furthermore , exogenous expression of ApoA1 , a major apolipoprotein of HDL , also facilitated the production of HCV particles as shown in Fig . 4E . These data suggest that the roles of the exchangeable apolipoproteins in HCV assembly are independent from the production of VLDL . MTTP plays crucial roles in the lipoprotein formation through the incorporation of triglyceride into growing lipoprotein and secretion of ApoB [21] . Although it has been shown that treatment with an MTTP inhibitor impairs the production of HCV particles [11] , in this study , we found that knockout of MTTP abrogated the secretion of ApoB but not the production of infectious HCV particles . Collectively , these data suggest that exchangeable apolipoproteins redundantly participate in the infectious particle formation of HCV independently from lipoprotein secretion machinery . Production of HCV capsids in the culture supernatants is impaired in 293T cells expressing miR-122 due to lack of ApoE expression , but envelopment of viral capsids is observed [37] , suggesting that ApoE is involved in the post-envelopment step . Coller et al . suggested that ApoE is associated with de novo formation of HCV particles during secretory pathway based on an experiment using HCV possessing a tetracysteine-tag in the core protein [38] . In this study , ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 and ApoE enhanced the formation of HCV particles in the post-envelopment step . These results suggest that a direct interaction between exchangeable apolipoproteins and enveloped particles in the ER lumen facilitates an efficient secretion of infectious HCV particles . Ultrastructural analysis of HCV particles has shown that large amounts of apolipoproteins , including ApoA1 , ApoB and ApoE , bind to the surface of viral particles [39] . Interestingly , ApoE-specific antibodies were more efficient in capturing viral particles than α-E1/E2 antibodies , and significantly large numbers of gold particles reacting with ApoE were observed per virion than those with E2 , suggesting that viral envelope proteins are masked by a large amount of apolipoproteins . The unique characteristics of interaction between apolipoproteins and HCV particles might be applied for visualization of entry and purification of HCV particles by using GFP- or affinity-tagged amphipathic α-helices of apolipoproteins . In the previous report , virocidal amphipathic helical peptides impaired the infectivity of viral particles [40] . There is a possibility that such peptide influences on the interaction between apolipoproteins and viral particles , and might be a new therapeutic approach . In previous reports , the importance of the interaction between lipoprotein receptors and ApoE in the entry of HCV has been well established . Lipoprotein receptors including scavenger receptor class B type 1 ( SR-B1 ) and low-density lipoprotein receptor ( LDLR ) are involved in HCV entry into the target cells [41] , [42] . LDLR is thought to mediate cell attachment of HCV through an interaction with virus associated ApoE [43] , [44] . SR-B1 also interacts with ApoE and hypervariable region 1 ( HVR1 ) in the envelope protein of HCV [43] . In this study we have shown that exchangeable apolipoproteins including not only ApoE but also ApoA and ApoC facilitate the production of infectious HCV particles , and that exchangeable apolipoproteins directly associate with viral particles . Meunier et al . reported that ApoC1 associates intracellularly with viral particles during particle morphogenesis and enhances the entry of HCV through an interaction of the C-terminal region of ApoC1 with heparan sulfate [45] . Another group also showed that ApoC1 enhances HCV infection through the triple interplay among HVR1 , ApoC1 , and SR-B1 [46] . These results suggest that the interaction of HCV particles with apolipoproteins also participates in the entry through the binding of lipoprotein receptors including SR-B1 and LDLR . Although the gene-knockout technique is essential to obtain reproducible and reliable data , and many knockout mice have been produced in various research areas , the development of experimental tools for HCV study has also been hampered by the narrow cell tropism [47] , [48] . A humanized mouse model in which human liver cells were xenotransplanted into immunodeficient mouse was developed and provided an important platform for the analysis of pathogenesis and the development of antivirals for HCV [49] . However , the exogenous expression of human receptor molecules required for HCV entry and impairment of innate immunity are required for the complete propagation of HCV in mice [50] . Gene-knockout techniques using a CRISPR/Cas9 system composed of guide RNA and Cas9 nuclease that form RNA-protein complexes to cleave the target sequences [19] have allowed quick and easy establishment of gene-knockout mice and cancer cell lines [51] , [52] , and indeed , such MTTP-knockout cell lines were established also in this study . Recently , the high-throughput screening of host factors involved in several conditions was reported by using a CRISPR/Cas9 system [53] . Together , these novel genome-editing techniques are expected to reveal the precise roles of host factors involved in the HCV life cycle . In summary , we have shown that apolipoproteins , including ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 , ApoE and ApoB , possess redundant roles in the assembly of HCV through the interaction of the amphipathic α-helices in the apolipoproteins with viral particles in the post-envelopment step . It is hoped that these findings will provide clues to the life cycle of HCV and assist in the development of novel antivirals targeting the assembly process of HCV .
The NextBio Body Atlas application presents an aggregated analysis of gene expression across various normal tissues , normal cell types , and cancer cell lines [20] . It enables us to investigate the expression of individual genes as well as gene sets . Samples for Body Atlas data are obtained from publicly available studies that are internally curated , annotated , and processed . Body Atlas measurements are generated from all available RNA expression studies that used Affymetrix U133 Plus or U133A Genechip Arrays for human studies . The results from 128 human tissue samples were incorporated from 1 , 067 arrays; 157 human cell types from 1 , 474 arrays; and 359 human cancer cell lines from 376 arrays . Gene queries return a list of relevant tissues or cell types rank-ordered by absolute gene expression and grouped by body systems or across all body systems . In the current analysis , we determined the expression levels of the apolipoproteins ApoA1 , ApoA2 , ApoB , ApoC1 , ApoC2 , ApoC3 , ApoD , ApoE , ApoH , ApoL1 , ApoL2 and ApoO in liver tissue . We used an analysis protocol developed by NextBio , the details of which have been described previously [20] . Expression profiling was generated using the 4 x 44 K whole human genome oligo-microarray ver . 2 . 0 G4845A ( Agilent Technologies ) as previously described [54] . Raw data were imported into Subio platform ver . 1 . 12 ( Subio ) for database management and quality control . Raw intensity data were normalized against GAP-DH expression levels for further analysis . These raw data have been accepted by GEO ( a public repository for microarray data , aimed at storing MIAME ) . Access to data concerning this study may be found under GEO experiment accession number GSE32886 . All cell lines were cultured at 37°C under the conditions of a humidified atmosphere and 5% CO2 . The human hepatocellular carcinoma-derived Huh7 and human embryonic kidney-derived 293T cells were obtained from Japanese Collection of Research Bioresources ( JCRB ) Cell Bank ( JCRB0403 and JCRB9068 ) , and maintained in DMEM ( Sigma ) supplemented with 100 U/ml penicillin , 100 µg/ml streptomycin , and 10% fetal calf serum ( FCS ) . The Huh7-derived cell line Huh7 . 5 . 1 was kindly provided by F . Chisari . Huh7 cells harboring JFH1-based HCV-SGR were prepared according to the method of a previous report [54] and maintained in DMEM containing 10% FCS and 1 mg/ml G418 ( Nakalai Tesque ) . The cDNA clones of pri-miR-122 , ApoA1 , ApoA2 , ApoC1 , ApoC2 , ApoC3 , ApoE , ApoH , and AcGFP were inserted between the XhoI and XbaI sites of lentiviral vector pCSII-EF-RfA , which was kindly provided by M . Hijikata , and the resulting plasmids were designated pCSII-EF-miR-122 , pCSII-EF-MT-apolipoproteins , and pCSII-EF-AcGFP , respectively . The deletion mutants of ApoC1 and ApoE were amplified by PCR and introduced into pCSII-EF . pHH-JFH1-E2p7NS2mt contains three adaptive mutations in pHH-JFH1 [55] . The pFL-J6/JFH1 plasmid that encodes the entire viral genome of the chimeric strain of HCV-2a , J6/JFH1 , was kindly provided by Charles M . Rice [8] . pTH/JFH1 ( genotype 1b ) and pS310/JFH1 ( genotype 3a ) were used for the production of chimeric viruses [22] , [23] . The plasmid pX330 , which encodes hCas9 and sgRNA , was obtained from Addgene ( Addgene plasmid 42230 ) . The fragments of guided RNA targeting the MTTP gene were inserted into the Bbs1 site of pX330 and designated pX330-MTTP . The plasmids used in this study were confirmed by sequencing with an ABI 3130 genetic analyzer ( Life Technologies ) . Mouse monoclonal antibodies to HCV core , β-actin and Calnexin were purchased from Thermo Scientific and Sigma Aldrich , respectively . Mouse anti-ApoA1 , ApoB , ApoC1 , ApoE and ApoH antibodies were purchased from Cell Signaling , ALerCHEK Inc . , Abnova , NOVUS Biologicals , and Santa Cruz Biotechnology , respectively . Rat anti-ApoA2 and Sheep anti-ApoC2 antibodies were purchased from R&D systems . Rabbit anti-NS5A antibody was prepared as described previously [54] . Alexa Fluor ( AF ) 488-conjugated anti-rabbit or mouse IgG antibodies , and AF594-conjugated anti-mouse IgG2a antibodies were purchased from Life Technologies . A small interfering RNA ( siRNA ) pool targeting various apolipoproteins ( siGENOME SMARTpool ) and control nontargeting siRNA were purchased from Dharmacon , and transfected into cells using Lipofectamine RNAi MAX ( Life Technologies ) according to the manufacturer's protocol . A human shRNA library was purchased from Takara Bio Inc . Upon transfection of pHH-JFH1-E2p7NS2mt or in vitro transcribed TH/JFH1 , J6/JFH1 and S310/JFH1 RNA into Huh7 . 5 . 1 cells , HCV in the supernatant was collected after serial passages , and infectious titers were determined by a focus-forming assay and expressed in focus-forming units ( FFU ) [22] , [23] , [54] . To compare the localization of core protein , J6/JFH1 was used in Fig . 6E . Pseudoparticles expressing HCV envelope glycoprotein were generated in 293T cells as previously reported [5] , and infectivity was assessed by luciferase expression using the Bright-Glo Luciferase assay system ( Promega ) and expressed in relative light units ( RLU ) . The lentiviral vectors and ViraPower Lentiviral Packaging Mix ( Life Technologies ) were co-transfected into 293T cells by Trans IT LT-1 ( Mirus ) , and the supernatants were recovered at 48 h post-transfection . The lentivirus titer was determined by the Lenti-XTM qRT-PCR Titration Kit ( Clontech ) , and the expression levels and AcGFP were determined at 48 h post-inoculation . Cells lysed on ice in lysis buffer ( 20 mM Tris-HCl [pH 7 . 4] , 135 mM NaCl , 1% Triton-X 100 , 10% glycerol ) supplemented with a protease inhibitor mix ( Nacalai Tesque ) were boiled in loading buffer and subjected to 5–20% gradient SDS-PAGE . The proteins were transferred to polyvinylidene difluoride membranes ( Millipore ) and reacted with the appropriate antibodies . The immune complexes were visualized with SuperSignal West Femto Substrate ( Pierce ) and detected by the LAS-3000 image analyzer system ( Fujifilm ) . Custom ZFN plasmids were designed to bind and cleave the ApoB , ApoE and MTTP genes and were obtained from Sigma Aldrich . Huh7 cells were transfected with in vitro transcribed ZFNs mRNA or pX330-MTTP by Lipofectamine 2000 ( Life Technologies ) , and single cell clones were established by the single cell isolation technique . To screen for gene-knockout Huh7 cell clones , mutations in target loci were determined by using a Surveyor assay as previously described [56] . Frameshift of the genes and deficiencies of protein expression were confirmed by direct sequencing and immunoblotting analysis , respectively . Protein concentrations of ApoB or ApoE in the culture supernatants were determined by using ELISA immunoassay kits ( Alercheck Inc . ) according to the manufacturer's protocol . Total RNA was extracted from cells by using an RNeasy minikit ( Qiagen ) and the first-strand cDNA synthesis and qRT-PCR were performed with TaqMan EZ RT-PCR core reagents and a ViiA7 system ( Life Technologies ) , respectively , according to the manufacturer's protocol . The primers for TaqMan PCR targeted to the noncoding region of HCV RNA were synthesized as previously reported [54] . Taqman Gene expression assays were used as the primers and probes targeting to apolipoproteins ( Life Technologies ) . Fluorescent signals were analyzed with the ViiA7 system . Cells cultured on glass slides were fixed with 4% paraformaldehyde ( PFA ) in phosphate buffered saline ( PBS ) at room temperature for 30 min , permeabilized for 20 min at room temperature with PBS containing 0 . 2% Triton after being washed three times with PBS , and blocked with PBS containing 2% FCS for 1 h at room temperature . The cells were incubated with PBS containing the appropriate primary antibodies at room temperature for 1 h , washed three times with PBS , and incubated with PBS containing AF488- or AF594-conjugated secondary antibodies at room temperature for 1 h . For lipid-droplet staining , cells incubated in medium containing 20 µg/ml BODIPY for 20 min at 37°C were washed with pre-warmed fresh medium and incubated for 20 min at 37°C . Cell nuclei were stained with DAPI . Cells were observed with a FluoView FV1000 laser scanning confocal microscope ( Olympus ) . The plasmid pSGR-JFH1 was linearized with XbaI , and treated with mung bean exonuclease . The linearized DNA was transcribed in vitro by using the MEGAscript T7 kit ( Life Technologies ) according to the manufacturer's protocol . The in vitro transcribed RNA ( 10 µg ) was electroporated into Huh7 cells at 107 cells/0 . 4 ml under conditions of 190 V and 975 µF using a Gene Pulser ( Bio-Rad ) and plated on DMEM containing 10% FCS . The medium was replaced with fresh DMEM containing 10% FCS and 1 mg/ml G418 at 24 h post-transfection . The remaining colonies were cloned by using a cloning ring ( Asahi Glass ) or fixed with 4% PFA and stained with crystal violet at 4 weeks post-electroporation . Intracellular viral titers were determined according to a method previously reported [10] . Briefly , cells were extensively washed with PBS , scraped , and centrifuged for 5 min at 1000× g . Cell pellets were resuspended in 500 µl of DMEM containing 10% FCS and subjected to three cycles of freezing and thawing using liquid nitrogen and a thermo block set to 37°C . Cell lysates were centrifuged at 10 , 000× g for 10 min at 4°C to remove cell debris . Cell-associated infectivity was determined by a focus-forming assay . Correlative fluorescence microscopy-electron microscopy ( FM-EM ) allows individual cells to be examined both in an overview with fluorescence microscopy and in a detailed subcellular-structure view with electron microscopy . Cells infected with HCVcc were examined by the correlative FM-EM method as described previously [57] . Culture supernatants of cells infected with HCVcc were concentrated 50 times by using Spin-X UF concentrators ( Corning ) , and the intracellular proteins collected after freeze-and-thaw were applied to the top of a linear gradient formed from 10–40% OptiPrep ( Axis-Shield ) in PBS and spun at 32 , 000 rpm for 16 h at 4°C by using an SW41 Ti rotor ( Beckman Coulter ) . Aliquots of 10 consecutive fractions were collected , and the infectious titer and density were determined . The proteinase K digestion protection assay was performed as described previously [37] . Briefly , cells were extensively washed with PBS , scraped , and centrifuged for 5 min at 1000× g . The cell pellets were resuspended in 500 µl of PBS and subjected to three cycles of freezing and thawing using liquid nitrogen and a thermo block set to 37°C . The cell lysates were centrifuged at 10 , 000× g for 10 min at 4°C to remove cell debris . The cell lysates were then incubated with 50 µg/ml proteinase K ( Life Technologies ) in the presence or absence of 5% Triton-X for 1 h on ice , and the digestion was terminated by addition of PMSF ( Wako Chemical Industries ) . The data for statistical analyses are the average of three independent experiments . Results were expressed as the means ± standard deviation . The significance of differences in the means was determined by Student's t-test .
|
In vitro systems have been developed for the study of hepatitis C virus ( HCV ) infection and have revealed many details of the life cycle of HCV . Apolipoprotein B ( ApoB ) and ApoE have been shown to play crucial roles in the particle formation of HCV , based on data obtained by siRNA-mediated gene knockdown and overexpression of the proteins . However , precise roles of the apolipoproteins in HCV assembly have not been elucidated yet . In this study , we show that infectious particle formation of HCV in Huh7 cells was severely impaired by the knockout of both ApoB and ApoE genes by artificial nucleases , and this reduction was cancelled by the expression of not only ApoE , but also other exchangeable apolipoproteins , including ApoA1 , ApoA2 , ApoC1 , ApoC2 and ApoC3 . In addition , expression of amphipathic α-helices in the exchangeable apolipoproteins restored the infectious particle formation in the double-knockout cells through an interaction with viral particles . These results provide clues to the understanding of life cycle of HCV and the development of novel antivirals to HCV .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"hepatitis",
"c",
"infectious",
"diseases",
"infectious",
"hepatitis",
"medicine",
"and",
"health",
"sciences",
"hepatitis",
"gastroenterology",
"and",
"hepatology",
"viral",
"diseases",
"liver",
"diseases"
] |
2014
|
Amphipathic α-Helices in Apolipoproteins Are Crucial to the Formation of Infectious Hepatitis C Virus Particles
|
Acquisition of malaria immunity in low transmission areas usually occurs after relatively few exposures to the parasite . A recent Plasmodium vivax experimental challenge trial in malaria naïve and semi-immune volunteers from Colombia showed that all naïve individuals developed malaria symptoms , whereas semi-immune subjects were asymptomatic or displayed attenuated symptoms . Sera from these individuals were analyzed by protein microarray to identify antibodies associated with clinical protection . Serum samples from naïve ( n = 7 ) and semi-immune ( n = 9 ) volunteers exposed to P . vivax sporozoite-infected mosquito bites were probed against a custom protein microarray displaying 515 P . vivax antigens . The array revealed higher serological responses in semi-immune individuals before the challenge , although malaria naïve individuals also had pre-existing antibodies , which were higher in Colombians than US adults ( control group ) . In both experimental groups the response to the P . vivax challenge peaked at day 45 and returned to near baseline at day 145 . Additional analysis indicated that semi-immune volunteers without fever displayed a lower response to the challenge , but recognized new antigens afterwards . Clinical protection against experimental challenge in volunteers with previous P . vivax exposure was associated with elevated pre-existing antibodies , an attenuated serological response to the challenge and reactivity to new antigens .
Malaria remains an important public health problem worldwide , affecting mainly developing countries in Africa , Asia and Latin America . The World Health Organization estimated that 214 million cases of malaria occurred worldwide in 2015 [1] . Of these cases , 13 . 8 million cases were calculated to be caused by Plasmodium vivax , a parasite species that predominates in South-East Asia and the American continent where it accounts for more than 50% of malaria cases [1] . In areas of high malaria transmission , individuals continuously exposed to Plasmodium develop partial protection against severe symptoms at an early age and a significant number of asymptomatic infections are recorded [2] . This clinical protection is mediated by both innate and acquired mechanisms that are not well understood [2–4] . Under conditions of hypo- or meso-endemic transmission , both clinical and subclinical infections are seen in all age groups and , despite the lower frequency of malaria exposure , significant protection against the disease is induced [5] . A high prevalence of uncomplicated and asymptomatic P . vivax and P . falciparum malaria infections are reported in both hyperendemic and unstable malaria transmission regions , indicating that a significant level of clinical immunity is induced by repeated exposure to the parasite [2 , 6–9] . Specific antibodies against P . vivax and P . falciparum proteins have been reported to be associated with clinical immunity [2 , 4 , 10] . However , only a few antigens have been made available through traditional cloning methods or peptide synthesis . Sequenced P . vivax and P . falciparum malaria parasite genomes , along with high-throughput proteomic techniques and bioinformatics are powerful tools currently available for systematic analyses of humoral immune responses associated with naturally and experimentally induced malaria . These analyses provide a better understanding of malaria parasite-host interaction , disease pathogenesis , host immune response and the identification of potential vaccine candidate antigens [11–13] . Despite the epidemiological importance of P . vivax , the immune mechanisms and their potential for vaccine development have been studied less than in P . falciparum . Currently , only two parasite antigens , PvCSP and Pvs25 have been assessed in early clinical development [14–16] as vaccine candidates , although several others are in preclinical development [17–19] . In recent years , the Malaria Vaccine and Drug Development Center ( MVDC ) in Cali ( Colombia ) has standardized a safe and reproducible method for P . vivax sporozoite challenge by Anopheles albimanus mosquito bites [20 , 21] . This method enables the evaluation of the protective efficacy of P . vivax vaccine candidates under controlled conditions , accelerating their clinical development both by facilitating efficacy studies and antigen discovery . In this context , a challenge study was recently conducted in malaria-naïve and semi-immune volunteers , who were exposed to controlled P . vivax infected mosquito bites [22] . Although all study subjects became parasitemic at the same time point after P . vivax challenge , all naïve volunteers developed symptomatic infections while semi-immune volunteers had either only mild symptoms or no symptoms . Antibody responses against two immune-dominant P . vivax antigens , PvCSP and PvMSP1 , showed no differences in the frequency of responders , although naïve volunteers exhibited significantly higher antibody responses to these antigens [22] . In order to fully characterize the natural protective antibody responses and to better understand the responses induced by P . vivax infection in both study groups , a protein microarray displaying 515 P . vivax antigens was probed with serum samples from these volunteers .
This trial was conducted according to ICH E-6 Guidelines for Good Clinical Practices [23] and the protocol was approved by Institutional Review Boards ( IRB ) of the MVDC and Centro Médico Imbanaco in Cali . Written informed consent was obtained from each volunteer at enrollment . The clinical trial was registered on clinicaltrials . gov , registry number NCT01585077 . The protocol for this trial is available as supporting information ( S1 Protocol ) . Blood samples were collected from malaria-naïve ( n = 7 ) and semi-immune ( n = 9 ) adult volunteers that participated in a clinical trial carried out at the MVDC [22] . Malaria-naïve volunteers were recruited in Cali ( a non-endemic city ) and declared not having suffered malaria and lack of previous malaria exposure was ascertained by negative indirect fluorescent antibody test ( IFAT ) . Semi-immune volunteers were recruited in Buenaventura ( malaria-endemic area ) and previous malaria exposure was confirmed by clinical history as well as by the presence of antibodies against P . vivax blood stages and sporozoites detected by IFAT . All volunteers were challenged by exposure to bites of two to four mosquitoes previously fed with P . vivax-infected blood obtained from a malaria patient ( field strain ) as reported before [22] . Volunteers were followed-up for malaria signs and symptoms and were treated orally with curative doses of chloroquine ( 25 mg/kg ) split in three doses and primaquine ( 0 . 5 mg/kg daily ) for 14 days , as recommended by the official Colombian guidelines , as soon as parasites were detected by microscopy [24] . Serum samples were collected before the challenge ( baseline ) and five , 11–13 ( here day 11 ) , 45 and 145 days after the challenge ( Fig 1 ) . Detailed information about demographic characteristics of the study participants , challenge infective dose , pre-patent period , parasite density after challenge , and clinical and laboratory evaluations was previously reported [22] . A custom protein microarray ( Pf/Pv500 ) displaying 515 P . vivax ( Pv ) and 500 P . falciparum proteins expressed on pre-erythrocytic and asexual parasite blood stages and printed as in vitro transcription/translation ( IVTT ) system was purchased from Antigen Discovery Inc . , ( Irvine , CA ) . Arrays content was down-selected from the Pv 4 , 506-protein microarray based on seroreactivity as detailed previously [25] . Although volunteers’ samples were hybridized to the whole array , data for P . vivax antigens only are presented in this paper . Microarray information is publicly available on the NCBI Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo/ ) and is accessible through accession number GPL18316 . Annotation of proteins presented in this study follows gene accession numbers published on PlasmoDB ( www . plasmodb . org ) . Of 515 P . vivax features on the array , 444 mapped to unique P . vivax proteins , of which the majority ( 247; 56% ) were classified as hypothetical proteins or hypothetical conserved proteins . Each array contained 24 negative “IVTT-control” reaction spots lacking plasmid template expression , which provide a donor-specific ‘background’ signal that was used to normalize data between individuals . For probing , serum samples were diluted 1:100 in protein array blocking buffer ( Maine Manufacturing , Sanford , ME ) supplemented with E . coli lysate ( GenScript , Piscataway , NJ ) to reach a final concentration of 10mg/ml , and pre-incubated at room temperature ( RT ) for 30 min . Concurrently , arrays were rehydrated in blocking buffer ( without lysate ) for 30 min . Arrays were probed with pre-incubated serum samples overnight at 4°C with gentle agitation , and then washed at RT five times with TBS-0 . 05% Tween 20 ( T-TBS ) , followed by incubation with biotin-conjugated goat anti-human IgG ( Jackson ImmunoResearch , West Grove , PA ) diluted 1:200 in blocking buffer for one hour at RT . After incubation with secondary antibodies , arrays were washed three times in T-TBS and bound IgG was visualized using streptavidin-conjugated SureLight P-3 ( Columbia Biosciences , Frederick , MD ) diluted 1:1000 in blocking buffer for 45 min at RT in the dark . Arrays were washed three times with T-TBS , and once with water . Chips were air-dried by brief centrifugation and scanned in a GenePix 4200AL laser scanner ( Molecular Devices , Sunnyvale , CA ) . All samples in this study were probed at the same time on the same batch of arrays . Analysis of the protein microarray data was accomplished following our previously published computational methods [3 , 11] . Briefly , microarray spot intensities ( median fluorescence intensity , MFI ) were quantified using ScanArray Express software ( Perkin Elmer , Waltham , MA ) and IVTT spot intensities were normalized by subtraction of the sample-specific median of the IVTT control spots . Antigens were considered seroreactive if the spot intensity of an individual ( or the average for a group of individuals ) was greater than a cutoff defined as the average plus two standard deviations of the reactivity to all P . vivax antigens in a US control population . Antibody breadth was used as defined for P . falciparum [26] as the number of seroreactive antigens per individual or group . Venn diagrams of group antibody breadths were produced using the BioVenn web application ( http://www . cmbi . ru . nl/cdd/biovenn/index . php ) [27] . Statistical analyses were performed on data normalized by dividing the IVTT signal by the sample-specific median of the IVTT control spots division ( fold-over control , FOC ) and taking the base 2 logarithm of the ratio ( Log2 FOC ) . Differentially reactive proteins between both groups were determined using Wilcoxon rank-sum test , and those with Log2 FOC > 1 considered seropositive ( Prism v6 . 0 , GraphPad Software Inc . , La Jolla CA ) . A p value < 0 . 05 was considered statistically significant .
Volunteers were adults aged between 19 and 38 years . Briefly , all volunteers developed infections , which were confirmed by microscopy and RT-qPCR , with similar median parasitemias between naïve and semi-immune volunteers ( 36 parasites/μL; IQR 9 . 0–98 . 8 vs 55 parasites/μL; IQR 29 . 5–163 . 5; p = 0 . 288 ) . All naïve volunteers presented with classical malaria signs and symptoms , while semi-immune volunteers displayed minor or no symptoms on the day of diagnosis [22] . Fig 2A shows a heat map of ‘subtracted’ array data ( IVTT values minus sample-specific IVTT controls signals ) for each naïve and semi-immune individual , and for US controls . The analysis revealed higher reactive responses in semi-immune than naïve individuals before the challenge , and both groups’ responses were higher than those in the US controls . This differential reactivity is seen more clearly from the slopes of the linear regression lines when average signal intensities from each group are plotted against the average of all three groups ( Fig 2B ) . The steeper slope of the semi-immune individuals relative to the naïve individuals confirms an overall higher reactivity in this group . The breadth of the baseline antibody profile , defined as the sum of reactive P . vivax antigens per individual , ranged from three to 71 reactive antigens for naïve individuals and three to 89 for semi-immune individuals . While the average group antibody breadth was broader for the semi-immune group ( 179 antigens ) in comparison to naïve volunteers ( 113 antigens ) , both groups shared reactivity for 98 of the antigens . Only a single seropositive antigen ( PVX_003775 , MSP4 ) was significant when naïve and semi-immune groups were compared ( p < 0 . 05 ) . To test whether this small number of differences was influenced by serum dilution , arrays were probed at 1:200 and 1:400 dilutions . A dilution of 1:200 yielded eight differentially reactive antigens with a Log2 FOC >1 ( Fig 2C and Table 1 ) . The majority of these were merozoite surface antigens , consistent with previous exposure to blood-stage parasites , with only one non-annotated antigen represented . At 1:400 dilution the number of reactive antigens fell to only two ( PVX_003775 , MSP4 and PVX_003770 , MSP5 ) indicating that the 1:200 dilution is optimal to maximize differences seen between Colombian naïve and semi-immune individuals . In conclusion , IgG antibodies to several P . vivax antigens were more elevated in semi-immune Colombian individuals than naïve Colombian individuals , although both groups had elevated antibodies compared to naïve US controls . To normalize differences in background reactivity seen between both study groups and to reveal only the signals induced in response to the P . vivax challenge , pre-existing background reactivity at baseline ( day 0 ) for each antigen was subtracted from the later time points data . In the semi-immune volunteers , the reactivity after challenge corresponded to a boosting of antibodies already present at baseline as well as appearance of new ones . At day five , reactivity of a few proteins was significantly higher in semi-immune than in naïve volunteers: serine-repeat antigen 5 ( SERA5; PVX_003830 ) and three hypothetical proteins with unknown function ( PVX_094690 , PVX_084120 , PVX_113590 ) . However , at diagnosis day ( day 11 ) the antibody response to P . vivax remained similarly low in both groups ( Fig 3A–3C ) . Notably , reactivity rose abruptly on day 45 in both groups , followed by a decline to near baseline by day 145 ( Fig 3A ) . The profile in one naïve volunteer ( indicated by † in Fig 3A ) who presented with a new P . vivax infection on day 130 ( indicated by ‡ in Fig 3A ) did not decline by the final time point . Indeed , the profile remained strong at a follow-up time point of 145 days . Since the serological dynamic of this individual was different to the others in the group , these data were removed from subsequent analyses . The expansion of the profile as measured by the group antibody breadth ( Fig 3B ) , was marginally more rapid in the semi-immune group , although at response peak ( day 45 ) the breadths were roughly equivalent in both groups ( naive = 188; semi-immune = 181; total reactivity = 236 ) . Both group profiles declined thereafter with roughly equivalent breadths at day 145 ( naïve = 103; semi-immune = 88; total reactivity = 138 ) . The response dynamics are shown by the dot plots of antibody breadth ( Fig 3C ) . These data were not subtracted from baseline signals to more clearly show the challenge-induced increase in the breadth relative to the pre-challenge baseline . To determine how long remain the antibodies elicited against P . vivax antigens without parasite re-exposure , the individuals were followed-up for 145 days . On this day several antigens were identified as significant when naïve and semi-immune were compared , although only six were considered seropositive ( Log2 FOC > 1 ) ; two of them were higher in semi-immune volunteers ( SERA5 and a hypothetical protein , PVX_094690 ) . In contrast , naïve volunteers had higher response to MSP1 , MSP8 , ETRAMP and a hypothetical protein with unknown function ( PVX_083560; S1 Table ) . As described [22] , naïve individuals all developed classical malaria symptoms such as headache , fever , nausea , chills , and malaise associated with P . vivax challenge at the time of parasite patency . In contrast , semi-immune volunteers reported either no symptoms or only minor symptoms associated with the P . vivax appearance in blood; only 33% presented fever ( body temperature ≥ 38°C ) . Therefore , semi-immune volunteers were segregated into those that developed symptoms ( or “non-protected” ) or did not develop symptoms ( or “protected” ) . The group means of the top 40 individual antigens subtracted from baseline signals showed that those semi-immune individuals that developed fever after challenge had a robust ( naïve-like ) response that peaked on day 45 , while the asymptomatic individuals showed an attenuated response at this time ( Fig 4A ) . Both returned to near baseline by day 145 . Segregation using headache as a symptom was also analyzed with similar results ( S1 Fig ) . Comparison of the protected and unprotected semi-immune profiles on day 45 identified several antigens as significant when data were segregated by fever ( Fig 4B and Table 2 ) , although only 12 were considered seropositive ( Log2 FOC >1; indicated by the bracket in Fig 4B ) . Interestingly , all of them were higher in semi-immune volunteers with fever . In semi-immune individuals segregated by headache , several antigens were significant , although only one ( PVX_002550; conserved hypothetical ) was considered seropositive ( Fig 4C ) . To test the hypothesis that several antigens recognized in semi-immune individuals at the peak of the response after challenge were “new” antibodies absent from the baseline profile , as opposed to boosted from antibodies present at baseline , statistical comparison between profiles at day 0 ( baseline ) vs . day 45 ( peak ) was performed . Those that were protected or non-protected ( using fever as the symptom ) were analyzed separately . Volunteers without fever developed antibodies to 13 new antigens , including three members of the MSP family ( one , seven and 10 ) and three hypothetical proteins , whereas individuals with fever had reactivity to 16 new antigens . However , antibodies to only five new antigens were shared by both groups , all of them with higher reactivity in volunteers with fever ( Table 3 ) . These data suggests that only one P . vivax infection is enough to induce antibody response against new antigens .
This study revealed that individuals who were semi-immune to P . vivax had pre-existing antibodies that although present at low levels were associated with clinical protection to P . vivax sporozoite experimental challenge [22] . As expected , semi-immune volunteers showed higher reactivity than naïve individuals to several P . vivax antigens before challenge . Moreover , exposure to a presumably low dose of viable sporozoites inoculated by the bites of only 2–4 mosquitoes was enough to induce a robust antibody response in malaria-naïve volunteers as well as to trigger antibody responses to new antigens in semi-immune volunteers ( Table 3 ) . Another valuable observation was that a proportion of the anti-P . vivax antibodies were short-lived as 138 of the 236 antigens ( >40% ) recognized by day 45 had disappeared by day 145 after challenge . The rapid decay of a subset of antibodies indirectly indicated that semi-immune volunteers had not had recent exposure to the parasites , because several of these antigens were not recognized at pre-challenge time . Before challenge , the Colombian malaria-naïve individuals had significantly higher serological reactivity than the US controls , despite being residents of a non-endemic malaria area . They were confirmed as seronegative against P . vivax blood stages and sporozoites using IFAT . Although infections or experience with protozoa were not studied here , the reactivity observed in Colombian naïve individuals might be due to other pathogens such as Cryptosporidium parvum or others highly prevalent in Colombia [28]; C . parvum shows homology with several Plasmodium proteins [29] . Nevertheless , this serological reactivity did not appear to have played a role in protection as all naïve volunteers developed malaria-related symptoms and patent parasitemia at the expected time [20–22] . The higher reactivity of the semi-immune volunteers to several antigens before challenge as compared to naïve volunteers indicates that in endemic regions , even with low transmission intensity , they develop and maintain P . vivax specific antibodies to a broad number of antigens even after a few previous malaria episodes ( 2–5 episodes ) . However , the degree of immunity conferred by these pre-existing antibodies was not enough to modify the pre-patent period or parasitemia at diagnosis day , although it was highly effective in controlling malaria symptoms . Interestingly , in the subgroup of semi-immune volunteers that developed fever or headache , as in the naïve , the antibody response to challenge was more vigorous than that in asymptomatic volunteers who displayed an attenuated antibody response . This is consistent with findings from P . falciparum vaccination studies in humans where protected individuals did not mount a significant antibody response to challenge , whereas unprotected subjects responded to challenge by elevated signals to many blood stage antigens [11 , 30] . Although in those studies PfCSP was recognized by both the protected and unprotected subgroups , protected individuals had a significantly higher magnitude of response [11 , 30] . At day 45 volunteers with fever showed a significantly higher response to P . vivax antigens such as MSP3 , MSP4 , MSP5 and MSP10 . However , reactivity to PvMSP1 and PvCSP , two established vaccine candidates [14 , 31] , was not different between volunteers with and without fever , as previously seen for the same sera using a recombinant PvMSP1 fragment ( r200L ) and synthetic PvCSP construct by ELISA [22] . These results partially contrast with those of epidemiological studies on P . vivax where an association between sera reactivity to MSP1 , MSP3 and MSP9 proteins and clinical protection has been reported [10 , 32–34] . The higher reactivity to the CSP in P . falciparum studies [30 , 35] is most likely due to the multiple immunization doses , while here only a few mosquito bites were allowed , with possibly low sporozoite density sufficient to induce infection once and a detectable antibody levels against a high number of other P . vivax antigens in all volunteers . In summary , the antibody profiles that developed in humans after experimental exposure to P . vivax sporozoites were defined . It was shown that a single infection was enough to induce detectable specific antibodies in malaria naïve volunteers and to boost the antibodies elicited by natural exposure to malaria in semi-immune individuals . Comparison between semi-immune volunteers segregated by fever showed that those protected had an attenuated serological response after challenge , but also had reactivity to new antigens , which may represent promising targets for vaccine development . Taken together , these findings represent a significant step forward in the understanding of the humoral immune response to P . vivax malaria infection , particularly the extent of priming upon a first parasite encounter .
|
Malaria remains an important public health problem worldwide , with 13 . 8 million cases caused by Plasmodium vivax , a parasite species that predominates in South-East Asia and the American continent . Despite the epidemiological importance of this species , studies of the immune response and their potential for vaccine development are limited . Here we use a high-throughput technique ( protein microarray ) to identify antibodies in serum from malaria naïve and semi-immune Colombian volunteers experimentally infected with P . vivax . We show a higher response in semi-immune individuals before the challenge . Meanwhile , at day 45 after infection , both groups had the highest antibody response to several P . vivax proteins . Additional analysis indicated that semi-immune volunteers without fever recognized new antigens , which may represent promising targets for vaccine development . Taken together , these findings represent a significant step forward in the understanding of the humoral immune response to P . vivax malaria infection , particularly the extent of immune priming upon a first parasite encounter .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"plasmodium",
"immunology",
"tropical",
"diseases",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"membrane",
"proteins",
"apicomplexa",
"protozoans",
"signs",
"and",
"symptoms",
"bioassays",
"and",
"physiological",
"analysis",
"antibodies",
"cellular",
"structures",
"and",
"organelles",
"antibody",
"response",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"malarial",
"parasites",
"proteins",
"cell",
"membranes",
"immune",
"response",
"microarrays",
"biochemistry",
"cell",
"biology",
"fevers",
"physiology",
"biology",
"and",
"life",
"sciences",
"malaria",
"organisms"
] |
2016
|
Antibody Profiling in Naïve and Semi-immune Individuals Experimentally Challenged with Plasmodium vivax Sporozoites
|
Decision making and learning in a real-world context require organisms to track not only the choices they make and the outcomes that follow but also other untaken , or counterfactual , choices and their outcomes . Although the neural system responsible for tracking the value of choices actually taken is increasingly well understood , whether a neural system tracks counterfactual information is currently unclear . Using a three-alternative decision-making task , a Bayesian reinforcement-learning algorithm , and fMRI , we investigated the coding of counterfactual choices and prediction errors in the human brain . Rather than representing evidence favoring multiple counterfactual choices , lateral frontal polar cortex ( lFPC ) , dorsomedial frontal cortex ( DMFC ) , and posteromedial cortex ( PMC ) encode the reward-based evidence favoring the best counterfactual option at future decisions . In addition to encoding counterfactual reward expectations , the network carries a signal for learning about counterfactual options when feedback is available—a counterfactual prediction error . Unlike other brain regions that have been associated with the processing of counterfactual outcomes , counterfactual prediction errors within the identified network cannot be related to regret theory . Furthermore , individual variation in counterfactual choice-related activity and prediction error-related activity , respectively , predicts variation in the propensity to switch to profitable choices in the future and the ability to learn from hypothetical feedback . Taken together , these data provide both neural and behavioral evidence to support the existence of a previously unidentified neural system responsible for tracking both counterfactual choice options and their outcomes .
It is widely agreed that a network of brain areas centered on ventromedial prefrontal cortex ( VMPFC ) and including anterior and posterior cingulate cortex encodes the values of choices that are taken [1]–[3] . Such a representation is assumed to be important for two functions . First , a representation of choice value is needed for decision making [1] . Second , a representation of a choice's value is needed for comparison with the subsequently experienced outcome [4] , [5] . The discrepancy between the two , called the prediction error , is thought to be fundamental for learning because it partly determines the degree to which future reward expectations for the choice should be revised [6] , [7] . While it is essential for organisms to represent the value of the choices that they take , there may also be considerable adaptive advantages associated with representing the reward potential of choices that are untaken . We refer to such potential , but untaken , choices as counterfactual choices . Such representations may confer both decision-making and learning advantages . First , if the reward potential of such choices is maintained neurally , then the organism may be better able to choose them in the future when it is beneficial , even in the absence of learning . Second , such a representation would make it possible to learn valuable information about what would have ensued had another choice been taken without having to incur both the energetic and opportunity costs that making the choice would have entailed . These representations would therefore enable us to exploit valuable , otherwise discarded , information and in turn make superior decisions in complex environments ranging from foraging in the wild to investing in financial markets . Unlike regret-related influences on behavior , which can lead to suboptimal biases in decision making , learning from counterfactual prediction errors should lead to more optimal decision making . There is preliminary evidence that the lateral frontal polar cortex ( lFPC ) may contribute to such a decision-making representation during binary choice; lFPC activity increases with the potential future reward associated with the unchosen option [8] . Organisms , however , are frequently confronted with choices between multiple uncertain prospects , and whether and how lFPC activity might guide decision making in such situations is unknown . When a decision-making problem is no longer binary , several potential schemes for coding unchosen options emerge . First , lFPC may represent the potential future reward of both unchosen options ( Figure 1A ) . There may , however , be limits to the number of potential alternative courses of action that can be represented [9] , [10] . A second possible scheme , therefore , is that lFPC codes for the opportunity cost of the chosen option—that is , the value of the best of the unchosen options—and discards the worst option ( Figure 1B ) . However , a third coding scheme is also possible; the FPC may weigh up the best unchosen option relative to the other options—in other words , relative to both the chosen option and the other unchosen option ( Figure 1C ) . Such a system would be indicative of a mechanism for evaluating the merit of choosing the best pending option at forthcoming decisions rather than a system for evaluating whether it is beneficial to choose either pending option in the future . Such a coding scheme might allow for very efficient transitions in behavior in a changing environment , but as we explain below , it predicts more effective switching to some options than others . There has been considerable recent interest in the possibility that the brain encodes fictive information [11] , [12] . Specifically , it has been shown that activity in the dorsal striatum and parietal cortex is sensitive to the difference between the best possible outcome that could have been attained and the experiential outcome over gains but not losses [9] . Furthermore , single neurons in monkey dorsal anterior cingulate cortex are sensitive to the size of untaken outcomes [12] . It remains unclear , however , whether the brain encodes prediction errors for counterfactual choices—the discrepancy between the outcome for an untaken choice and the reward expectation associated with making that choice—in a separate and parallel manner to experiential prediction errors . This is necessarily difficult to establish in any paradigm in which there is a systematic relationship between the outcomes of counterfactual and experiential choices [11] , [12] . To tackle these and related issues , we conducted an FMRI experiment in which human subjects made voluntary decisions between three options with independent reward probabilities , followed on most trials by decisions between the remaining two options that were unchosen during the first decision . Choices were made on the basis of two pieces of information: the probability of reward associated with each stimulus ( which the participant had to estimate from recent outcomes of both chosen and unchosen options ) and the reward magnitude associated with each stimulus ( which was displayed on the screen beneath each stimulus and changed unpredictably from trial-to-trial ) . A Bayesian model was used to infer the reward outcome probabilities [13] . These manipulations enabled us to dissociate the relevant variables guiding immediate decisions ( the three option expected values ) from those guiding future decisions ( the three option reward probabilities ) and to test for the independent representation of counterfactual prediction errors during learning . Here we show that the lFPC , DMFC , and PMC encode key parameters for both selecting and learning about counterfactual options .
Participants performed a decision-making task in which they repeatedly chose between a face , house , and body stimulus that were presented at one of three locations at random ( Figure 2A ) . On each trial random integers between 1 and 100 were displayed beneath the stimuli that indicated the size of potential reward associated with selecting that option . Participants were informed that since these reward magnitudes were generated randomly on each trial , it was not advantageous to track them across trials . However , participants were not directly cued about the probability with which each option would be rewarded if chosen . Instead , participants were told that these reward probabilities depended only on the recent outcome history . To produce a changeable environment , these reward probabilities varied from trial to trial according to a fixed volatility [13] during the course of the experiment . On two-thirds of trials ( conditions 2 and 3 ) , participants encountered a second decision between the two options that were foregone at the first decision , a manipulation that enabled more accurate estimates of participants' own ranking of the two unchosen options . Following the second decision , feedback on chosen and counterfactual options was presented , thereby allowing us to search for neural correlates of counterfactual prediction errors . On the other third of trials ( condition 1 ) , there was no second decision; instead feedback on the two unselected options was presented ( Figure 2A ) . An optimal Bayesian learner [13] was used to model participant estimates of the probabilities of reward associated with the options given the history of recent choice outcomes ( i . e . , rewarded or unrewarded chosen and unchosen options ) ( Figure 2B ) . The Bayesian learner enabled us to select a reward schedule that de-correlated the reward probabilities associated with each option . In the selected reward schedule , there was limited correlation between the reward probability associated with the three options ( mean r across participants between body part and face stimuli = − . 4; body part and house stimuli = 0 . 01; face and house stimuli = − . 4 ) and between the expected value ( reward probability×reward magnitude ) associated with the three options ( mean r between body part and face stimuli = − . 2; body part and house stimuli = 0 . 03; face and house stimuli = −0 . 14 ) . Although we could not know what choices our participants would ultimately make , this increased the likelihood that chosen and unchosen reward probabilities or expected values would also be de-correlated . As anticipated , there was indeed limited correlation between chosen and unchosen option reward probabilities ( mean r for chosen and best unchosen <− . 1; chosen and worst unchosen = − . 2; best unchosen and worst unchosen = . 37 ) . Similarly , there was little correlation between the chosen and unchosen option values ( mean r for chosen and best unchosen < . 1; chosen and worst unchosen < . 1; best unchosen and worst unchosen = . 43 ) ( Figure 3A ) . It is important to note that the random trial-by-trial fluctuations in reward magnitude meant that only option probabilities had to be maintained for making future choices . This feature of the experimental design also meant that it was optimal to learn about option reward probabilities but not reward magnitudes . Before searching for evidence of a neural representation of unchosen options , it is important to assess whether there is behavioral evidence that people not only update their future reward expectations experientially , from the feedback provided for the chosen option , but also counterfactually , from the feedback provided for unchosen options . Similarly , it is important to establish whether or not there is evidence that behavior is influenced by the values of the different possible options that might be taken . To assess the fit to behavior of the optimal Bayesian model , which used both experiential and counterfactual feedback to update estimates of options' reward probabilities , we computed the log likelihood and Bayesian Information Criterion and compared the fit to two alternative models ( Table 1; Figure S4 ) . To test whether people learn from counterfactual feedback , we compared the fit of the optimal Bayesian model to the fit of an alternative model that we term an “experiential Bayesian model” because it is identical to the optimal model except that it does not update unchosen options . As can be seen from Table 1 , the optimal Bayesian model clearly outperformed the experiential Bayesian model , implying that people learned from both experienced and counterfactual feedback . Finally , we also compared the optimal model with a conventional Rescorla Wagner model that updates both chosen and unchosen options . The optimal model was also a far better fit to behavior than the Rescorla Wagner model that updates both chosen and unchosen options ( Table 1; Text S1 ) . It is notable that these Bayesian models have some parallels with learning models used previously to analyze behavior during experimental games [14] . To further assess whether the optimal Bayesian model captured choice behavior , as well as which variables influenced participant choices , we performed logistic regression analyses . This analysis aimed to determine the degree to which choosing the most valuable option was influenced by the outcome probabilities as estimated by the optimal Bayesian model and reward magnitudes associated with the best , mid , and worst options ( see Text S1 for details ) . This analysis revealed a strong positive effect of the best option ( reward probability: t ( 18 ) = 7 . 56 , p<0 . 0001; reward magnitude: t ( 18 ) = 5 . 45 , p<0 . 0001 ) , a strong negative effect of the mid option ( reward probability: t ( 18 ) = −8 . 50 , p<0 . 0001; reward magnitude: t ( 18 ) = −4 . 86 , p<0 . 0001 ) , and a modest but highly consistent effect of the worst option ( reward probability: t ( 18 ) = −2 . 27 , p = 0 . 02; reward magnitude: t ( 18 ) = −5 . 16 , p<0 . 0001 ) on choices of the best option ( i . e . , optimal choices ) ( Figure 3B ) . In other words , the reward probabilities estimated by the optimal Bayesian model and the explicitly presented reward magnitudes were both strong predictors of participants' choices . Consistent with the experimental design , the reward magnitude from the previous trial by contrast did not have any impact on current choices of the best option ( best reward magnitude trial i-1: t ( 18 ) = 0 . 49 , p = 0 . 31; mid reward magnitude trial i-1: t ( 18 ) = 0 . 67 , p = 0 . 25; worst reward magnitude trial i-1: t ( 18 ) = 0 . 59 , p = 0 . 28 ) . Thus , optimal estimates of reward probability and current but not past reward magnitudes strongly influenced participant behavior ( see Text S1 for more details ) . The analyses described above indicate that although the best and mid options principally drive choice , the worst option also consistently explains a small amount of variance in participant choices of the best option . Whether people learn differently from experiential and counterfactual feedback remains an open question . To address this we constructed an additional model in which separate learning rates scaled chosen and unchosen prediction errors ( Text S1 ) . These two participant-specific learning rates were fitted to each participant's choices using standard estimation procedures . In our experimental task , there was no difference between the learning rates for chosen and unchosen feedback ( t ( 18 ) < . 25 , p>0 . 4 ) , suggesting that people may not learn differently from counterfactual feedback and experiential feedback when these sources of information are both available and equally informative . In order to search for evidence of neural activity encoding the reward association of the best unchosen option , we first tested for voxels across the whole brain where activity correlated with the reward probability of the best unchosen option—one of the relevant metrics to track across trials to inform future switches . We also included the reward probability of the chosen and worse unchosen options as separate terms in the general linear model ( Table S1 ) . This analysis revealed three regions with a positive effect of the reward probability of the best unchosen option ( Z>3 . 1 , p<0 . 001 uncorrected; cluster extent >10 voxels ) : left lateral frontopolar cortex ( lFPC; Z = 3 . 50 , MNI x = −36 , y = 58 , z = −4; Z = 3 . 64 , x = −32 , y = 46 , z = −2 ) , posteromedial cortex ( PMC; Z = 3 . 70 , x = 2 , y = −62 , z = 38 ) , and dorsomedial frontal cortex ( DMFC; Z = 3 . 33 , x = 6 , y = 34 , z = 42 ) ( Figure 4A; Table S2 ) . Although no activations exceeded the threshold in right lFPC , an activation emerged at the reduced threshold of p<0 . 003 , uncorrected ( see Figure 4A ) . It is important to note that we have deliberately refrained from using the most sensitive regressor in this analysis because the purpose was to define ROIs that will be unbiased for later tests . When the best unchosen probability relative to the chosen probability is instead used as the regressor , there is a more robust effect in the lFPC ( Figure S1; Z = 3 . 99 , MNI x = −36 , 58 , −6 ) , as expected based on a previous demonstration that the lFPC encodes a relative signal in a binary choice task [8] . To test whether and how the lFPC , PMC , and DMFC might also encode alternative options , we performed orthogonal analyses on the time courses of these regions identified by the whole-brain analysis . In addition to a positive correlation with the best unchosen reward probability , the lFPC signal correlated negatively with the reward probability for both the chosen ( t ( 18 ) = −3 . 46 , p<0 . 005 ) and other unchosen option ( t ( 18 ) = −2 . 21 , p<0 . 03 ) during the decision-making phase ( Figure 4B ) . The lFPC signal correlates positively with the reward potential of the best alternative and negatively with the reward potential of both the chosen option and the worse unchosen option , suggesting that lFPC forecasts the evidence in favor of choosing the better of the two unchosen options at future choices . Such an activity pattern is inconsistent with FPC simply maintaining a representation of the advantage to be gained from switching to any alternative action . It would , however , be predicted if lFPC represented only one alterative action in a pending state . Under such a scheme , the negative encoding of the reward probability of both the chosen option and the worse unchosen option can be interpreted as reflecting the potential opportunity cost of foregoing the chosen action or the other alternative action if there were to be a switch in behavior to the pending state . We also identified evidence for a very similar pattern of activity in PMC and a closely related one in DMFC . In PMC there was a significant negative correlation with the chosen option ( t ( 18 ) = −2 . 23 , p<0 . 02 ) and the other unchosen option during the decision-making phase ( t ( 18 ) = −2 . 0 , p<0 . 03 ) , whereas in DMFC , there was a significant negative correlation with the chosen option ( t ( 18 ) = −1 . 9 , p<0 . 04 ) , but the effect of the worse unchosen option was not significant ( t ( 18 ) = −1 . 24 , p>0 . 11 ) during the decision-making phase . We repeated these analyses on only those trials on which a single decision had to be made in order to exclude the possibility that activity related to a second decision could confound activity related to the feedback phase of a first decision . Because of the short temporal interval between the first feedback phase and the second decision-making phase in our experiment on some trials , activity related to late time points during the first feedback period is difficult to dissociate from activity related to the second decision in these time course analyses . When there is a second decision the effect of the worst unchosen option flips from being encoded negatively to positively when it frequently becomes the best ( and only ) unchosen option at the second decision-making period . To circumvent such issues , we reexamined the time course during condition 1 in which there was not a second decision that could interfere with the lFPC response to the first decision and feedback phases . Although the number of trials in this analysis is substantially reduced , there were still significant negative effects of the chosen probability ( t ( 18 ) = −2 . 93 , p<0 . 005 ) and worst unchosen probability ( t ( 18 ) = −2 . 62 , p<0 . 01 ) ( Figure 5A ) in lFPC . Similarly , the PMC signal was significantly and negatively correlated with the chosen reward probability ( t ( 18 ) = −2 . 17 , p<0 . 03 ) and worst unchosen reward probability ( t ( 18 ) = −2 . 06 , p<0 . 03 ) . However , in DMFC there was still only a significant negative effect of the chosen option ( chosen option: t ( 18 ) = −2 . 36 , p<0 . 02; worst unchosen option: t ( 18 ) = −0 . 98 , p>0 . 17 ) . The interpretation that this signal reflects the future evidence in favor of choosing the best unchosen option at subsequent decisions makes a testable prediction about behavior . Participants in whom this evidence is better represented should go on to choose the best pending option more frequently . It is important to note that during the initial decision , it is the unchosen option with the highest reward probability that is likely to be the best option at second decisions . In accord with the hypothesis , the greater the effect of the best unchosen probability relative to the worst unchosen probability in the lFPC across participants in conditions 2 and 3 ( i . e . , when there was a second decision ) , the more frequently participants chose the pending option that was associated with the highest reward probability at second decisions ( Figure 5C ) . The identified lFPC coding scheme further suggests that people may be better at adapting behavior to the next best alternative than to the worse alternative when confronted with decisions between multiple options . Such a scheme makes an intriguing prediction about behavior . It is possible that people switch choices to the next best alternative more effectively than they do to the worse alternative . This prediction is testable in our task because the previously worst option might become the best option when random reward magnitudes are introduced at the onset of a new trial . Consistent with this proposition , we found that participants adapted choices to the best pending option when it was optimal significantly more frequently than they did to the worse pending option when it was optimal , even when the analysis was restricted to trials matched for value difference ( t ( 18 ) = 2 . 17 , p = 0 . 02 ) . A valuable source of information during learning comes not only from the experienced outcomes of actions that are taken but also from the consequences of alternative potential actions that might be taken in the future . It was hypothesized that brain regions that encode future reward expectations related to unchosen options might also be involved in updating those expectations . This prediction is based upon recent evidence demonstrating that prediction error-like signals can be identified in brain regions thought to be specialized for visual and social processing when participants must their expectations during visual and social learning , respectively [15]–[22] . We reasoned that the same principle might hold true for learning about unchosen options . Analysis of the time course of the lFPC , DMFC , and PMC regions identified by the whole-brain analysis revealed a significant correlation with the unchosen , but not chosen , prediction error following the delivery of feedback for the second decision in each region ( unchosen prediction error: lFPC: t ( 18 ) = 2 . 01 , p<0 . 05; DMFC: t ( 18 ) = 2 . 8 , p<0 . 01; PMC: t ( 18 ) = 4 . 35 , p<0 . 0005; chosen prediction error: lFPC: t ( 18 ) = . 39 , p>0 . 3; DMFC: t ( 18 ) = . 29 , p>0 . 35; PMC: t ( 18 ) = −0 . 11 , p>0 . 45 ) . Moreover , the pattern of activity in these regions elicited by counterfactual rewards ( Figure 6 ) was similar to that displayed by dopaminergic neurons for experienced rewards [23] . Activity correlated positively with the probability of reward for the unchosen option before the outcome was revealed ( lFPC: t ( 18 ) = 1 . 89 , p<0 . 05; DMFC: t ( 18 ) = 3 . 55 , p<0 . 005; PMC: t ( 18 ) = 2 . 06 , p<0 . 05 ) . Following the delivery of feedback , activity correlated negatively with this same probability ( lFPC: t ( 18 ) = −1 . 78 , p<0 . 05; DMFC: t ( 18 ) = −1 . 92 , p<0 . 05; PMC: t ( 18 ) = −1 . 90 , p<0 . 05 ) and positively with the unchosen outcome ( lFPC: t ( 18 ) = 1 . 81 , p<0 . 05; DMFC: t ( 18 ) = 2 . 65 , p<0 . 01; PMC: t ( 18 ) = 4 . 31 , p<0 . 0005 ) . These regions' activity therefore reflected both components of the counterfactual prediction error—the counterfactual outcome minus the expectation ( Figure 6 ) . Replicating previous findings , we identified experiential reward prediction errors in the ventral striatum , among other regions ( Figure S2; Table S2 ) . We further considered the possibility that neural counterfactual prediction error signals might have an impact on behavior . We reasoned that people in whom there is a greater effect of counterfactual prediction errors may learn more effectively from counterfactual feedback . To test this hypothesis , we compared the model fit to behavior of the optimal Bayesian model that updates both chosen and unchosen options and the experiential Bayesian model that only updates the chosen option . The difference in the fit to behavior of these models provided an index of the extent to which people learned from counterfactual feedback . Across the sample of participants , there was a tight relationship between the effect size for the counterfactual prediction error and the difference between the fits of the optimal and experiential models in the PMC ( r = 0 . 64 , p<0 . 005; Figure 6 ) . There was a similar tendency in lFPC , though this did not reach statistical significance ( lFPC: r = 0 . 39 , p = 0 . 10 ) , but not in DMFC ( r = . 24 , p>0 . 3 ) . Regret should theoretically grow as the difference between the reward magnitude of the foregone outcome and the chosen outcome increases , independently of the reward expectation [24] , [25] . Unlike regret-related FMRI signals that have been identified previously [11] , [25] , activity in the lFPC , DMFC , and PMC was not sensitive to the difference between the size of the outcomes for the unselected and the selected options ( lFPC: t ( 18 ) = −0 . 44 , p>0 . 3; DMFC: t ( 18 ) = −0 . 86 , p>0 . 2; PMC: t ( 18 ) = −0 . 59 , p>0 . 25 ) . Taken together , these findings demonstrate that counterfactual prediction errors are dissociable from regret in the lFPC , DMFC , and PMC .
A number of brain regions have been implicated in the representation of value during learning and decision making [17] , [26] , but in general the focus has been on the choices that participants make and the rewards they experience . Within the frontal cortex , the orbitofrontal and anterior cingulate cortical regions have most often been the focus of such research . Here , however , we show that the lPFC maintains a representation of the net profit to be expected from choosing the next best alternative in the future . The lFPC BOLD signal increases as the probability of obtaining reward from the next best alternative increases . The reward that might be sacrificed by switching away from the current action may be represented as a cost of switching to the alternative; lFPC BOLD activity decreases as the probability of reward associated with the current course of action increases . Similarly , the reward that might be lost by foregoing the worst unchosen option can be conceived of as a cost; lFPC BOLD decreases as the probability of reward associated with the worst foregone option increases . When there are multiple alternatives to choose between , the pattern of lFPC activity is therefore consistent with a system that forecasts the reward potential of the best alternative option and the costs of not taking both the current course of action and the other alternative . This coding scheme is consistent with a system that accumulates evidence in favor of choosing the best pending option in the future so that it can be switched to effectively . There were several plausible schemes according to which the lFPC could represent unchosen alternatives . The data presented here provide evidence in favor of the system depicted in Figure 1C—namely that the lFPC encodes the merit of potential future switches to the next best alternative . This interpretation is supported by a between-subject correlation between the effect of the reward probability associated with the best relative to the worst unchosen option in the lFPC and choices of the best pending option . It is also supported by the finding that participants are superior at adapting behavior to the next best alternative than to the worse alternative when these choices are optimal . The lFPC signal , as in a previous study [8] , contained peaks during both the decision-making and feedback phases ( see Figure 4A ) . This time course is consistent with the notion that the lFPC tracks the relevant decision variable across time for forthcoming choices . Several accounts propose that the FPC maintains information across time for future deployment [9] , [27]–[31] . FPC activity has been shown to increase when an intention or a task set has to be maintained during a delay and then implemented [28]–[30] , [32] , [33] , while damage to left anterior prefrontal cortex , including left FPC , disrupts effective switching between task sets in such paradigms [34] . It was recently shown that monkey FPC encodes the decision ( left or right response ) over an extended delay around the time of feedback , particularly when it was advantageous to maintain this information for use on the next trial [35] . Furthermore , FPC is selectively recruited when participants must maintain information in working memory whilst performing a subtask for the purpose of using the pending information upon completion of the subtask [27] , particularly when the time at which the pending information must be used is unpredictable [36] . On the basis of such findings , it has been posited that FPC has a special role in cognitive branching—the maintenance of pending information related to a previous behavioral episode during an ongoing behavioral episode for future use [9] , [37] . Following this framework , in our experiment the current decision could be conceived of as the ongoing behavioral episode , and the best unchosen option as the pending information , which may be selected in the future . While our findings are consonant with such accounts , we have shown that the FPC not only represents pending information or intentions for future use , but that it encodes the evidence in favor of their future deployment . Moreover , we have demonstrated that when people are confronted with more than two alternatives , the FPC specifically encodes the evidence in favor of the unchosen alternative that would be most advantageous to be selected in the future , a finding consistent with the view that there may be limits to FPC coding during decision making [9] . In our experimental setup , participants should have expected to encounter a second decision on approximately two-thirds of trials . Despite this manipulation we found no evidence that this knowledge influenced participants' initial decisions ( see Text S1 , Experimental Procedures ) . It is nevertheless possible that participants anticipated having to make a second decision at which point reward magnitudes would either remain the same as they were or equate to 50 . In a previous investigation in which participants made binary choices with no intervening second decision , the lFPC was shown to encode the unchosen option positively and the chosen option negatively , consistent with the positive coding of the next best alternative and negative coding of the chosen option we have revealed here in a multi-option context . It would be interesting to examine the coding of lFPC when people make decisions between multiple alternatives in the absence of any requirement to make decisions between the remaining unchosen options . lFPC appears to be only one component of a network of areas that are interconnected and whose activity tracks the advantage to be gained from switching to the next best alternative . The activation in PMC may be in area 31 of the posterior cingulate cortex [38] , [39] , while the DMFC activation appears to be situated between the pre-SMA and dorsal anterior cingulate cortex ( dACC ) . Anterograde and retrograde studies have examined the anatomical connections between these regions in monkeys . Area 31 of the monkey has reciprocal connections with both FPC and parts of DMFC [38] , and FPC also projects to parts of DMFC [40] . A recent study in macaque monkeys has identified neurons in a neighboring region of the PMC that are selective for exploration and switching between four different response alternatives [41] . Moreover , the pre-SMA has been implicated in switching between task sets [42] . Taken together , these findings suggest that the lFPC , PMC , and DMFC regions might form part of an interconnected network dedicated to tracking the evidence in favor of future switches to the best pending option and , in collaboration with the mid-IPS , implementing such switches [8] , [43] . It is notable that the three components of the counterfactual choice circuit are some distance from foci in ventral DMFC , ventral PMC , and VMPFC in which the BOLD signal is correlated with the value of the action that is chosen [1] , [3] , [8] , [44]–[47] . Reinforcement-learning models theorize that agents should learn from both chosen and unchosen outcomes [6] . Nevertheless , to our knowledge prediction error signals related to unchosen options have yet to be identified in the mammalian brain . Lohrenz and colleagues [9] have reported activity in the dorsal striatum and posterior parietal cortex that they refer to as a fictive error signal . Although this metric influences behavior in interesting ways [11] , [48] , it is distinct from the one that we report here because it correlates with the best possible outcome that could have been attained minus the experienced outcome received , over gains but not losses . Crucially such fictive signals pertain to the choice of a different level of the taken action . They do not contain information about alternative actions with independent probabilities of success . By contrast , a counterfactual prediction error—the counterfactual outcome minus its expectation—should theoretically be proportional to the degree to which future reward expectations of unchosen options are updated . We found that the lFPC , DMFC , and PMC—regions whose activity is sensitive to the unchosen option with the highest reward probability during initial decisions—encoded counterfactual prediction errors when participants witnessed counterfactual outcomes of subsequent decisions . A prediction error should theoretically signal the prediction of an event before its revelation and , following its revelation , the discrepancy between the event's occurrence ( or non-occurrence ) and the prediction—a prediction error [6] . It has been well documented , in the context of experienced rewards , that both signals are closely approximated by the firing rate of phasically active dopamine neurons [23] . The pattern of activity in lFPC , DMFC , and PMC similarly exhibited both of these components but in relation to counterfactual rewards: before the outcome was revealed there was a positive correlation with the expectation of reward for the unchosen option; once the outcome was witnessed , there was a negative correlation with this same expectation and a positive correlation with the outcome ( reward or no reward for the unchosen option ) . Notably , in our experimental setup unchosen reward probabilities were relevant for future predictions , but unchosen reward magnitudes were of no relevance because they changed randomly from trial to trial . Counterfactual prediction error coding in lFPC , DMFC , and PMC thus reflected the relevant information for learning about unchosen options in our task—reward probabilities . Consistent with the claim that lFPC , DMFC , and PMC encode counterfactual prediction errors , but not regret [25] , activity in these regions was not sensitive to the difference between the reward magnitudes of obtained and unobtained outcomes . These data therefore constitute the first neural dissociation of counterfactual prediction errors from regret . Intriguingly , neurons that encode counterfactual rewards have recently been identified in the monkey dACC [12] , which neighbors and is interconnected with the DMFC region identified here [39] , [49] and is also interconnected with the PMC and FPC [38] , [40] . These observations raise the possibility that unchosen reward signals in dACC might be integrated with unchosen expectations to compute counterfactual prediction errors in lFPC , DMFC , and PMC . We also tested whether there exists a relationship between the neural coding of counterfactual prediction errors and the propensity to learn from counterfactual information . In the PMC there was a strong relationship between the effect of counterfactual prediction errors and how effectively participants learned from counterfactual outcomes . This finding suggests that neural coding of counterfactual information in PMC influences counterfactual learning behavior . In neuroscience , there is an emerging view that predictive coding extends beyond the domain of experienced reward [18] , [50] , [51] . In the perceptual domain , unsigned prediction error ( or surprise ) responses have been identified in inferior temporal gyrus ( ITG ) when participants observe gabor patches whose orientation does not match the orientation of a template during A , not A decisions [18] . When the stimuli are faces or houses , rather than gabor patches , fusiform face area ( FFA ) and parahippocampal place area ( PPA ) are sensitive to unsigned prediction errors related to predictions concerning faces and houses , respectively [19] , [22] , a modulation that at least partly contributes to the phenomenon of repetition suppression in the FFA [19] . During incidental audio-visual learning , the BOLD response in primary visual cortex and putamen was shown to correlate with unsigned prediction errors , when the appearance ( or absence ) of a black and white shape stimulus was unpredicted ( or predicted ) by an auditory tone [20] . In the social domain , two recent investigations [15] , [21] have revealed signed prediction error responses in the superior temporal sulcus ( STS ) and dorsomedial prefrontal cortex ( DMPFC ) —brain regions implicated in theory of mind tasks [52] , [53]—when participants have to learn about the behavior of another individual . Prediction errors in these regions have been discovered when the objective was to learn about the reputation of a social partner [15] , or when it was to learn about the influence of an opponent's choice on the likely future behavior of the opponent [21] . Collectively , these investigations in the perceptual and social domains carry fundamental implications: First , they suggest that prediction error coding is more ubiquitous than previously thought and , second , that brain regions specialized for a given class of information may also encode prediction errors specifically related to that class of information . The present finding that regions which encode information related to unchosen options also encode unchosen prediction errors adds counterfactual information to the classes of information for which prediction error signals have been identified . In summary , we have delineated the functional contribution of a network centered on lFPC , DMFC , and PMC when human subjects decide between multiple alternatives . The results indicate that this network both forecasts reward expectations related to selecting untaken alternatives in the future and also updates those expectations—key computations for deciding and learning when to take the road less traveled .
Twenty-two healthy volunteers participated in the fMRI experiment . Two volunteers failed to use either the reward probabilities or reward magnitudes in the task , as indicated by values of nearly 0 for each of the free parameters in the behavioral model , and one volunteer failed to use reward probability , as indicated by values of 0 for both β and γ in the behavioral model ( see Behavioral Model description below ) . These participants' data were therefore discarded from all analyses . The remaining 19 participants ( 10 women ) were included in all further analyses . All participants gave informed consent in accordance with the National Health Service Oxfordshire Central Office for Research Ethics Committees ( 07/Q1603/11 ) . In our fMRI paradigm , participants decided repeatedly between three stimuli based on their expectation of reward and the number of points associated with each stimulus option ( Figure 2A ) . Although the number of points was generated randomly and displayed on the screen , the expectation of reward had to be estimated from the recent outcome history . The three stimuli were pictures of a real face , whole body , and house . The identities of the face , body , and house were fixed for the duration of the experiment and across participants . During the first decision-making phase , the three options and their associated points were displayed at three locations on the screen: left , upper middle , and right . The location at which each stimulus was displayed was randomized across trials . When the yellow question mark appeared in the centre of the screen , participants indicated their choices with right-hand finger responses on a button box corresponding to the location of each stimulus . Immediately after participants indicated their choice , the first feedback phase was presented: the selected option was highlighted by a red rectangle that framed the chosen stimulus and the chosen outcome ( reward or no reward ) was presented . If the participant's choice was rewarded , a green tick appeared in the centre of the screen , and the red prize bar also updated toward the gold rectangular target in proportion to the amount of points won on that trial . Each time the prize bar reached the gold target , participants were rewarded with £2 . If the participant's choice was not rewarded , a red X appeared in the centre of the screen , and the red prize bar remained stationary . These initial decision-making and chosen feedback phases were presented on every trial in the experiment . After presentation of the chosen feedback , one of three different conditions followed in pseudorandom order . In condition 1 the outcomes for the two remaining unchosen options were presented . A green tick or a red X appeared on the left of the two options that were unchosen during the first decision-making phase , depending on whether they were rewarded or unrewarded . The red prize bar did not move . This event was followed by presentation of the next trial . This condition was critical because it enabled us to isolate activity during the first decision-making and feedback phases uncontaminated by activity related to a second decision . In conditions 2 and 3 , participants had the opportunity to choose between the two remaining options that were unselected by the participant at the first decision . These two remaining stimuli maintained their spatial locations on the screen . In condition 2 , the option reward probabilities and points associated with the two options remained identical to what they were at the first decision ( Figure 2A ) . The purpose of this condition was to use the participants' responses at the time of the second decision to improve our ability to rank the two unchosen options at the time of the first decision on the basis of expected value . However , in condition 3 only the reward probabilities remained the same; the points for both remaining options were changed to 50 ( Figure 2A ) . Therefore , the only information guiding participant decisions in condition 3 should theoretically be the reward probabilities . This condition was introduced to more accurately rank the two unchosen options at the first decision on the basis of reward probability . For both conditions 2 and 3 , participants indicated their choice after a yellow question mark appeared . This was followed by simultaneous feedback for the chosen and unchosen options from the second decision . During this second feedback phase , a red rectangle framed the selected option and a green tick or red X was presented to the left of the chosen and unchosen options , depending on whether these options were rewarded or unrewarded . If the choice at the second decision was rewarded , the red prize bar updated in proportion to the number of points won . This event was followed by presentation of the next trial . There was no inter-trial interval in any condition . Each event was jittered between 2 . 5 and 5 . 5 s ( uniform distribution ) . There were 60 trials in each condition , making 180 trials in total . Conditions were pseudorandomly interleaved and were uncued . Participants earned between £20 and £28 on the task , depending on their performance . The true reward probabilities associated with each stimulus type varied independently from one trial to the next over the course of the experiment at a rate determined by the volatility , which was fixed in the current experiment . More specifically , the true reward probability of each stimulus was drawn independently from a beta distribution with a fixed variance and a mean that was determined by the true reward probability of that stimulus on the preceding trial . The true reward probabilities that participants tracked are shown in Figure S3 . FMRI data were acquired on a 3T Siemens TRIO scanner with a voxel resolution of 3×3×3 mm3 , TR = 3 s , TE = 30 ms , Flip angle = 87° . The slice angle was set to 15° and a local z-shim was applied around the orbitofrontal cortex to minimize signal dropout in this region [54] , which has previously been implicated in other aspects of decision making . The mean number of volumes acquired was 999 , giving a mean total experiment time of approximately 50 min ( see Text S1 , Experimental Procedures for further details ) . A general linear model ( GLM ) was fit in pre-whitened data space [55] . Twenty-four regressors were included in the GLM ( see Table S1 for a summary ) : the main effect of the first decision-making phase; the main effect of the first feedback phase; the main effect of the foregone outcome phase ( condition 1 ) ; the main effect of the second decision-making phase ( conditions 2 and 3 ) ; the main effect of the second feedback phase ( conditions 2 and 3 ) ; the interaction between chosen probability and the first decision-making phase; the interaction between chosen probability and the first feedback phase; the interaction between the best unchosen probability as determined by the model in conditions 1 and 2 and the first decision-making phase; the interaction between the best unchosen probability as determined by participant choices in condition 3 and the first decision-making phase; the interaction between the best unchosen probability as determined by the model in conditions 1 and 2 and the first feedback phase; the interaction between the best unchosen probability as determined by participant choices in condition 3 and the first feedback phase; the interaction between the worst unchosen probability as determined by the model in conditions 1 and 2 and the first decision-making phase; the interaction between the worst unchosen probability as determined by participant choices in condition 3 and the first decision-making phase; the interaction between the worst unchosen probability as determined by the model in conditions 1 and 2 and the first feedback phase; the interaction between the worst unchosen probability as determined by participant choices in condition 3 and the first feedback phase; the outcome at the first feedback phase; the outcome at the second feedback phase; and six motion regressors produced during realignment . Because there were not any notable differences between z-statistic maps based on the model or participant choices , we defined contrasts of parameter estimates ( COPEs ) for the best and worst unchosen probability as the combination of the regressors based on the model and participant choices . Based on the evidence from our previous investigation [8] that the lFPC encodes reward probability during both the decision and feedback phases , the reward probability regressors were modeled across both phases . To do so , additional COPEs defined the chosen , best unchosen , and worst unchosen probabilities as the sum of regressors over the first decision-making and feedback phases ( Table S1 ) . For group analyses , EPI images were first registered to the high resolution structural image using 7 degrees of freedom and then to the standard [Montreal Neurological Institute ( MNI ) ] space MNI152 template using affine registration with 12 degrees of freedom [56] . We then fit a GLM to estimate the group mean effects for the regressors described above . FMRIB's Local Analysis of Mixed Effects ( FLAME ) was used to perform a mixed effects group analysis [57] , [58] . All reported fMRI z-statistics and p-values arose from these mixed effects analyses on all 19 participants . We report clusters of greater than 10 voxels that survived a threshold of z>3 . 1 , p<0 . 001 , uncorrected . It should be noted that our analyses carefully avoid selection bias in identifying regions related to probability . Based on the findings of our previous study and other investigations [8] , [45] , [46] , we were confident that lFPC would encode the relative probability rather than either chosen or unchosen probability in isolation . One of the central aims of this experiment , however , was to test the hypothesis that the lFPC encoded the best unchosen probability and either the chosen probability , worst unchosen probability , or both ( see Figure 1 ) . For the probability-based analysis , rather than search for regions encoding the relative unchosen probability ( e . g . , the best unchosen probability relative to the chosen probability or the best unchosen probability relative to the average of the other probabilities ) , for which there are large effects in the lFPC ( see Figure S1 ) , we have searched only for regions that encode the best unchosen probability . We have used this analysis because it is orthogonal to the worst unchosen and chosen probability regressors and thus enables us to perform orthogonal tests on the regions of interest ( ROIs ) identified to test competing hypotheses . ROI analyses are presented in detail in Text S1 , Experimental Procedures . We used an optimal Bayesian reinforcement-learning algorithm [13] to model participant estimates of the reward probabilities and their eventual choices . This model has been described in detail in previous investigations [8] , [13] , [15] . Briefly , the model is composed of a “predictor” that estimates the reward probability associated with each option and other environmental statistics given only the observed data ( i . e . , the reward outcomes of chosen and unchosen options ) and a “selector” that chooses actions on the basis of these estimates . Because feedback is given on each option on each trial in our experimental task , the model updates the reward probability associated with each option upon receipt of feedback , as is optimal . These estimates of the reward probabilities were then combined with reward magnitude according to participant-specific free parameters that can differentially weigh probability , magnitude , and their product , to derive estimates of the subjective expected values . We found no evidence that participants' choices at the first decision were influenced by the prospect of a second decision at which reward magnitudes could either remain the same or both change to 50 ( Text S1 ) . We therefore assumed that subjective value at both decisions was computed on the basis of the current decision alone: ( 1 ) where , , and are the subjective value , reward probability , and reward magnitude associated with the stimulus ( face , house , or body ) on trial i . We fitted β , λ , and γ to each individual participant's behavioral data using standard non-linear minimization procedures implemented in Matlab 7 ( Mathworks ) . Finally , the selector assumed that participants chose stimulus s according to the following softmax probability distribution: ( 2 ) where is the subjective expected value of the stimulus , and Ns is the total number of stimuli to choose between ( Ns = 3 at the first decision , Ns = 2 at the second decision ) .
|
Reinforcement learning ( RL ) models , which formally describe how we learn from direct experience , can explain a diverse array of animal behavior . Considering alternative outcomes that could have been obtained but were not falls outside the purview of traditional RL models . However , such counterfactual thinking can considerably accelerate learning in real-world contexts , ranging from foraging in the wild to investing in financial markets . In this study , we show that three brain regions in humans ( frontopolar , dorsomedial frontal , and posteromedial cortex ) play a special role in tracking “what might have been” , and whether it is worth choosing such foregone options in the future . These regions encode the net benefit of choosing the next-best alternative in the future , suggesting that the next-best alternative may be privileged over inferior alternatives in the human brain . When people subsequently witness feedback indicating what would have happened had they made a different choice , these same regions encode a key learning signal—a prediction error that signals the discrepancy between what would have happened and what people believed could have happened . Further analysis indicates these brain regions exploit counterfactual information to guide future changes in behavior . Such functions may be compromised in addiction and psychiatric conditions characterized by an inability to alter maladaptive behavior .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cognitive",
"neuroscience",
"decision",
"making",
"biology",
"neuroscience",
"learning",
"and",
"memory"
] |
2011
|
Counterfactual Choice and Learning in a Neural Network Centered on Human Lateral Frontopolar Cortex
|
Molybdenum ( Mo ) is an essential micronutrient for plants , serving as a cofactor for enzymes involved in nitrate assimilation , sulfite detoxification , abscisic acid biosynthesis , and purine degradation . Here we show that natural variation in shoot Mo content across 92 Arabidopsis thaliana accessions is controlled by variation in a mitochondrially localized transporter ( Molybdenum Transporter 1 - MOT1 ) that belongs to the sulfate transporter superfamily . A deletion in the MOT1 promoter is strongly associated with low shoot Mo , occurring in seven of the accessions with the lowest shoot content of Mo . Consistent with the low Mo phenotype , MOT1 expression in low Mo accessions is reduced . Reciprocal grafting experiments demonstrate that the roots of Ler-0 are responsible for the low Mo accumulation in shoot , and GUS localization demonstrates that MOT1 is expressed strongly in the roots . MOT1 contains an N-terminal mitochondrial targeting sequence and expression of MOT1 tagged with GFP in protoplasts and transgenic plants , establishing the mitochondrial localization of this protein . Furthermore , expression of MOT1 specifically enhances Mo accumulation in yeast by 5-fold , consistent with MOT1 functioning as a molybdate transporter . This work provides the first molecular insight into the processes that regulate Mo accumulation in plants and shows that novel loci can be detected by association mapping .
Plants have developed complex biochemical and regulatory pathways to acquire mineral nutrients from the soil environment and distribute them to appropriate tissues . Natural populations of Arabidopsis thaliana ( Arabidopsis ) provide an excellent system to study how plants have adapted their mineral nutrient and trace element uptake pathways to thrive under different environmental conditions . Molybdenum ( Mo ) is an important micronutrient for plants , being incorporated into molybdopterin , an essential cofactor for enzymes involved in nitrate assimilation , sulfite detoxification , abscisic acid biosynthesis and purine degradation [1] . Molybdenum in either deficiency or excess has been demonstrated to inhibit plant growth and agricultural productivity [2] . The genes comprising the biochemical pathway that synthesizes the molybdopterin cofactor have been identified in plants , animals and microbes , but to date , a Mo transporter in plants has not been found [2] . The first committed step in molybdopterin biosynthesis occurs in the mitochondria [3] , confirming the predicted sub-cellular localization of the enzymes [1] . The remaining 3 steps are thought to occur in the cytoplasm [3] . While a substantial amount is known about the biochemistry , enzymology and underlying genetics of molybdopterin biosynthesis , very little is known about the mechanisms for Mo uptake , distribution and accumulation in plants . In this study , natural variation in whole plant Mo accumulation has been coupled with genomics techniques and genetics to identify a mitochondrial Mo transporter ( MOT1 ) that regulates whole plant Mo content in Arabidopsis . Alleles of this gene are demonstrated to be responsible for low Mo accumulation across a diversity collection of 92 Arabidopsis accessions . All soil grown plant ionomic data from this study is freely available at www . purdue . edu/dp/ionomics [4] .
In a Ler-0×Col-0 cross 51 of the 200 F2 progeny analyzed were found to have low shoot Mo contents similar to the Ler-0 parent . This ratio is consistent with the hypothesis that the low Mo phenotype in Ler-0 is controlled by a single locus , or several closely linked loci ( p<0 . 00001 by Shapiro test for normality ) . Similar segregation patterns were observed in an F2 population derived from a Ler-0 cross to an ionomics mutant ( 14501 ) in the Col-0 background . To obtain a rough map position , a bulk segregant analysis ( BSA ) experiment [9] was performed with microarray detection of genetic markers [10] , [11] , using ( 14501[Col-0]×Ler-0 ) F2 plants . Plants with the lowest shoot Mo contents ( n = 40 ) and plants with Mo shoot contents similar to Col-0 ( n = 40 ) were pooled separately , and genomic DNA from each pool hybridized to the Affymetrix Arabidopsis ATH1 DNA microarray . Using the oligonucleotide probes on the DNA microarray which show differential hybridization between Ler-0 and Col-0 as genetic markers ( Single Feature Polymorphisms or SFP ) , the locus responsible for the low shoot Mo content in Ler-0 was mapped to an area centered at 11 Mb on chromosome 2 ( Figure 4A ) . To further narrow down the map position of the locus controlling low shoot Mo content in Ler-0 , the Col-4×Ler-0 RIL population [12] was used to identify seven recombinants in the mapping region we had previously determined from the BSA . DNA microarray-based genotyping of this Col-4×Ler-0 RIL set was used to further refine the break points between Col-4 and Ler-0 genotypes [7] , [8] . The shoot Mo contents of these seven RILs showed a clear segregation with the genetic markers for Col-4 and Ler-0 within our mapping region , allowing us to classify these lines as having either the Col-0 or the Ler-0 allele for the low Mo locus ( Figure 4B ) . The precise breakpoints identified by Singer et al . [8] allowed the mapping interval to be narrowed to 346 kb on chromosome 2 ( 10 . 771 to 11 . 056 Mb ) , an interval containing 81 genes . Candidate genes from this interval were selected based on annotation and expression differences between Col-0 and Ler-0 ( T . Singer & S . Briggs , personal communication ) . T-DNA insertional alleles for these candidate genes were obtained , and the plants scored for low shoot Mo content compared to wild type Col-0 . A null mutant ( mot1-1 ) with an insertion disrupting the coding region of the gene At2g25680 ( Salk_118311 , Figure 4D ) , was observed to phenocopy the Ler-0 low shoot Mo content when grown in either soil ( Figure 1B ) , or shoots and roots when grown hydroponically ( Figure 2 ) . Differences in the absolute concentrations of Mo between Col-0 in Figure 2A and 2B are related to differences in growth conditions between these two experiments . The At2g25680 gene was originally predicted to be a putative sulfate transporter , and named AtSULT5 . 2 [13] . We have renamed this gene MOT1 ( molybdenum transporter 1 ) based on its phenotype of low shoot Mo content . Furthermore , we detected no change in the shoot content of S in mot1-1 ( Figure 5 ) or in the content of Li , B , Na , Mg , P , K , Ca , Mn , Fe , Co , Ni , Cu , Zn , As , Se , and Cd , providing additional support for the reannotation of AtSULT5 . 2 as MOT1 . Quantitative real-time RT-PCR of the MOT1 transcript in mot1-1 and Ler-0 revealed that both these low shoot Mo lines had significantly reduced expression levels of MOT1 in both root and shoot tissue , compared to Col-0 ( Figure 6 ) . Van-0 , a second low shoot Mo accession ( Figure 1B ) was also confirmed by qRT-PCR to have low expression of MOT1 ( Figure 6 ) . Significantly , we observed no differences in shoot S content between Col-0 and either Ler-0 or Van-0 ( Figure 5 ) . To establish if MOT1 is the locus responsible for the low shoot Mo content of Ler-0 and Van-0 , we crossed mot1-1 and Col-0 to both Ler-0 and Van-0 to test for complementation ( Figure 7 ) . F1 plants from the Ler-0×Col-0 and Van-0×Col-0 crosses contained significantly different levels of shoot Mo from either parent ( p<0 . 005 for all pairwise t-tests ) , suggesting that the Col-0 allele of the low Mo locus could only partially complement the Ler-0 allele . F1 plants from the Ler-0×mot1-1 and Van-0×mot1-1 crosses were found to have significantly lower shoot Mo contents than similar plants from the Ler-0×Col-0 and Van-0×Col-0 crosses ( p<0 . 0005 ) . Such data establishes that mot1-1 is deficient in complementing the low shoot Mo content of both Ler-0 and Van-0 . This is strong evidence confirming that the Ler-0 and Van-0 alleles are allelic with the recessive loss of function allele of MOT1 . To determine the polymorphism in the MOT1Ler-0 allele causal for low shoot Mo content , we sequenced MOT1 , including 1 kb upstream and 200 bp downstream of the coding sequence in both Ler-0 and Van-0 . We observed 18 polymorphisms in common between Ler-0 and Van-0 , as well as several polymorphisms unique to each accession , when compared to the Col-0 reference sequence . These include 15 single nucleotide polymorphisms ( SNPs ) in the 1 kb upstream region , two SNPs that change two amino acids ( I68T and V30L ) in the coding region , and a 53 base pair ( bp ) deletion 27 nucleotides upstream from the translation start site ( Text S1 ) . The deletion includes the TATA box ( Figure 4C ) . The altered expression of MOT1 in both of the low shoot Mo accessions , Ler-0 and Van-0 , suggested that this 5′ deletion may be the causal polymorphism driving low shoot Mo content in these accessions . To obtain further evidence that this deletion is the casual polymorphism driving low shoot Mo content , we performed an association analysis by sequencing the MOT1 5′ promoter region containing the 53 bp deletion in Ler-0 and Van-0 across 92 of the accessions originally screened for shoot Mo content ( Text S2 ) . Combining this information with the Nordborg [5] genotypes , we were able to scan for significant genetic associations with low shoot Mo content . The 53 bp deletion identified in Ler-0 and Van-0 was found in seven of the 92 accessions tested , and all accessions with the deletion had low shoot Mo content compared to the overall distribution of shoot Mo contents ( Figures 4C and 1A ) . The distribution of p-values for genome-wide associations with shoot Mo contents were skewed towards significance , suggesting a relationship between Mo content and the underlying population structure ( Figure S1 ) , making the evaluation of individual loci difficult . However , when kinship and population structure were taken into account [14] , the presence of the 53 bp deletion was found to be highly significantly associated with low shoot Mo content ( p<0 . 0001 ) and accounted for ∼14% of the total variation in Mo accumulation . Once the MOT1Ler loci is accounted for , there were several other markers which showed significant associations with Mo accumulation , and these may represent additional loci of interest ( Text S3 ) . To establish that MOT1 has the capacity to transport Mo , the Arabidopsis MOT1 cDNA was expressed in yeast . Yeast expressing MOT1 were observed to specifically accumulate five time more Mo than vector only controls ( p<1E-14 , Figure 8 ) . Furthermore , the accumulation of no other element , including Na , Mg , P , S , K , Ca , Mn , Fe , Co , Ni , Cu , Zn , and Cd , was observed to be altered more than 20% in two independent experiments . This evidence is consistent with MOT1 being a specific molybdate transporter . To determine the tissue localization of MOT1 expression , Col-0 was transformed with a 1 . 8 kb MOT1 promoter-GUS construct . In all of the promoter-GUS lines examined , GUS staining was observed in the roots , hypocotyls and leaves ( Figure 9 ) . In roots , GUS staining was most pronounced just behind the growing root tip in the primary root ( Figure 9A , B ) , and the lateral roots ( data not shown ) . Cross sections show that the strong GUS staining behind the root tip was restricted mainly to the protodermal cells ( Figure 9B2 ) . At the beginning of the elongation zone , GUS staining was mainly restricted to the epidermis and cortex ( Figure 9C2 ) . Thereafter , GUS staining in the root also occurred in the vascular tissue ( Figure 9D ) . In the hypocotyls the GUS staining was also mainly restricted to the vascular tissue ( Figure 9E ) . In fully expanded leaves GUS staining was found in the vascular tissue . However , the main vein was less intensely stained than the lateral veins ( Figure 9F ) . During flowering GUS activity was also visible in the vasculature of the stem leaves but not in the vasculature of the stem , flowers or developing siliques including seeds ( data not shown ) . The sub-cellular localization of the MOT1 protein was determined by transiently expressing a MOT1 C-terminal GFP translational fusion construct in shoot derived Col-0 protoplasts ( Figure 10A–10D ) , and by stable ectopic expression of a similar construct in Col-0 ( Figure 10E–10G ) . The GFP signal was observed to co-localize with the mitochondrial marker F-ATPase–RFP in shoot derived protoplasts transiently expressing the construct , and with the Mitotracker Red dye in roots of the stably transformed lines . The experimentally determined mitochondrial localization for MOT1 is in good agreement with several sub-cellular prediction programs ( summarized at the SUBA database , [15] ) , which predict a mitochondrial targeting sequence at the N-terminus of MOT1 . Phylogenetic analysis of the sulfate transporter family in Arabidopsis and rice [13] , [16] reveal two Arabidopsis genes , MOT1 and ST5 . 1 , to have diverged significantly from the other sulfate transporter family members , and each Arabidopsis gene has a rice ortholog . Given the sequence similarity between ST5 . 1 and MOT1 , the shoot Mo content of a st5 . 1-1 T-DNA insertion line ( Salk_015044 ) was analyzed to determine if it also shows the low Mo phenotype observed in mot1-1 . We were unable to detect any significant changes in the Mo content of shoot of st5 . 1-1 when grown in soil and the roots and shoot Mo accumulation of a mot1-1/st5 . 1-1 double mutant was not significantly different from the mot1-1 mutant alone . ( Figures 1B and 2B ) . The Mo content of yeast heterologously expressing the ST5 . 1 cDNA was found not to be reproducibly significantly different from vector only controls in two independent experiments ( Figure 8 ) . Thus , we can find no evidence that ST5 . 1 is also a Mo transporter . A search of the PiiMS database [4]; www . purdue . edu/ionomics ) found four T-DNA lines with insertions in sulfate transporters , representing the ST1 , ST2 and ST3 subfamilies ( At3g15990 , At4g02700 , At1g77990 , At4g08620 ) , none of which accumulated significantly different levels of Mo than Col-0 .
Molybdenum is one of the 14 essential minerals required by plants . Despite its importance as a cofactor in processes ranging from nitrogen metabolism to hormone biosynthesis , we still know relatively little about the regulation , uptake and transport of this transition metal . Using an ionomics approach to identify genes that affect the elemental composition of plants , we identified MOT1 as the causal gene driving reduced shoot Mo in various accessions of Arabidopsis . We have combined several lines of evidence to support this conclusion . First , a 54 bp deletion in the promoter of MOT1 was found to be strongly associated ( p<0 . 0001 ) with low shoot Mo content across 92 Arabidopsis accessions . Given the strong association between the presence of this deletion and low shoot Mo content , and the rate of linkage disequilibrium decay in Arabidopsis [17] , if this deletion is not the causal polymorphism , it is within 20 kb of the causal polymorphism . Second , deficiency complementation with a T-DNA allele ( SALK_118311 ) indicates that the MOT1 allele found in Ler-0 and Van-0 , accessions with a low content of shoot Mo , is responsible for this reduced Mo content . Third , MOT1 shows reduced expression in both Ler-0 and Van-0 compared to Col-0 , and the mot1-1 null allele in the Col-0 background phenocopies the low shoot Mo observed in both Ler-0 and Van-0 . Given the fact that MOT1 belongs to the sulfate transporter superfamily , and can transport Mo when expressed in yeast , it is easy to imagine how reduced expression of a Mo transporter could lead to low shoot Mo content , either via reduction in uptake from the soil and/or translocation to the shoot . In Col-0 , the functional MOT1 allele is strongly expressed in the root differentiation zone and in mature vasculature tissue of both the roots and the shoots , suggesting a defect in either or both roles could be the possible explanation for the observed phenotype . Our grafting experiments clearly show that the Mo defect is associated with the roots . Surprisingly , the MOT1 protein is not localized to the plasma membrane of roots cells where it could function in uptake into cells , but rather MOT1 is localized to the mitochondria . Based on these observations , and the grafting data , it is hypothesized that MOT1 regulates whole plant Mo accumulation at the level of the mitochondria in the root . Interestingly , the first committed step in molybdopterin biosynthesis has recently been shown to occur in the mitochondria , and this is consistent with the mitochondria acting as a control point in regulating whole plant Mo content [3] . Given that characterized members of the sulfate transporter superfamily are SO4−2/H+ co-transporters , we speculate that MOT1 is transporting MoO4−2 from the acidic mitochondrial intermembrane space to either the cytoplasm or the matrix . While we have localized MOT1 to the mitochondria , the question of how Mo enters the root cells remains . Based on its homology , ST5 . 1-1 is a candidate for this role , however a ST5 . 1 C-terminal GFP fusion was found to localize to the vacuole ( Buchner and Hawkesford , unpublished ) . Additionally , the st5 . 1-1 T-DNA insertional mutant does not have altered shoot Mo content , and heterologous expression of ST5 . 1 cDNA in yeast has no effect on Mo accumulation . ST5 . 1 also appears not to interact with MOT1 since the double mutant st5 . 1mot1-1 showed the same reduction in Mo content as mot1-1 . Alternatively , the putative plasma membrane molybdate transporter could be from a family unrelated to the SUL gene family , as multiple gene families have been shown to transport Zn2+ and Ca2+ , for example [18] , [19] . Finally , it remains possible that MoO42− is transported across the plasma membrane through a promiscuous transporter ( s ) with broad ion specificity as shown for E . coli [20] . Both sulfate and phosphate starvation have been shown to increase Mo accumulation , which suggests that Mo may be transported across the plasma membrane by sulfate and phosphate transporters [21] , [22] . Sequencing results show that the MOT1Ler-0 allele has a frequency of approximately 7% ( 7/92 ) in the natural diversity collection of Arabidopsis obtained from populations collected from a broad geographical region . This frequency is higher than we would expect if the lowered Mo content strongly reduced fitness . For example , the overall reduction in Mo content might cause a reduction in molybdopterin , negatively impacting the activity of MoCo containing enzymes like nitrate reductase . However , we found no significant reduction in growth , N accumulation or nitrate reductase activity in mot1-1 lines when grown with nitrate as the sole N source ( data not shown ) . This suggests that MoCo levels are not limiting in mot1-1 . Alternatively , loss of MOT1 function might increase the fitness of Arabidopsis by some as yet unknown process , and the association of Mo with population structure is consistent with Mo accumulation having a selective effect . However , an analysis of haplotype sharing [23] around this gene gave no indication of recent selection on the MOT1Ler-0 allele ( C . Toomajian , unpublished ) . Finally , it has been noted that large effect QTLs' might generally be too rare to be detected by association mapping [5] , [17] , [24] . The MOT1Ler-0 allele provides a clear counter-example , demonstrating that , given the dense marker maps that are now obtainable , association mapping will at least sometimes be a useful tool for finding such loci . While this paper was in review , Tomatsu et al . [25] also reported that MOT1 is the locus responsible for low Mo in Ler . However , they did not confirm this directly by complementation . Here we conclude that MOT1 is the locus responsible for low Mo in Ler-0 and Van-0 based on genetic complementation , and further show that a deletion in the 5′ UTR of MOT1 is strongly associated with low Mo in 92 different Arabidopsis accessions . The sequence for the MOT1Ler locus reported by Tomatsu et al . [25] , derived from the Cereon resequencing project [7] , differs from the results presented here by two additional amino acid altering SNPs . The MOT1Ler sequence presented here is based on resequencing after PCR amplification , and is consistent with that published by Clark et al . [26] . Furthermore , here we conclude that MOT1 is localized to the mitochondria . This is in contrast to Tomatsu et al . [25] , who suggest that MOT1 is localized to the plasma membrane and the secretory and/or endocytic pathways . We attribute these inconsistencies to differences in constructs and localization systems used in the two studies . To determine the localization of MOT1 Tomatsu et al . [25] reported that they prepared a construct in which GFP was fused to the N-terminus of MOT1 , blocking the predicted mitochondrial localization signal , and likely mislocalizing the GFP::MOT1 fusion protein in the tobacco BY-2 cells used for transient expression . Results reported here were obtained using a C-terminally fused MOT1::GFP construct that was expressed both transiently and stably in Arabidopsis and that clearly localized to the mitochondria . In summary , natural variation has been used here to identify a mitochondrially localized Mo transporter that controls both root and shoot Mo content in Arabidopsis . This discovery demonstrates that natural accessions of Arabidopsis are a rich source of interesting alleles , useful for the functional characterization of genes of unknown function . Furthermore , the identification of MOT1 as a regulator of total plant Mo accumulation provides molecular insight into plant Mo homoeostasis . Note added in proof: While this paper was in review , Tejada-Jimenez et al . [27] published the identification and characterization of the MOT1 ortholog in Chlamydomonas reinhardtii .
All accessions were obtained from the ABRC or Lehle seeds . The insertion of the T-DNA into the MOT1 coding region in mot1-1 and into the ST5 . 1 coding region in st5 . 1-1 were verified by sequencing , and the mutant was confirmed to be null for MOT1 expression by RT-PCR ( Figure S2 ) . All T-DNA lines analyzed were homozygous for the T-DNA insertion . Plants used for elemental profiling by ICP-MS analysis were grown in a controlled environment , 8 h light:16 h dark ( 90 µmol·m−2·s−1 light intensity ) and 19 to 22°C [6] . Briefly , seeds were sown onto moist soil ( Sunshine Mix LB2; Carl Brehob & Son , Indianapolis , Indiana , United States ) with various elements added at subtoxic concentrations [As , Cd , Co , Li , Ni , Pb , and Se [6]] and stratified at 4°C for 3 d . Plants were bottom-watered twice per week with 0 . 25× Hoagland solution in which iron was replaced with 10 µM Fe-HBED [N , N′-di ( 2-hydroxybenzyl ) ethylenediamine-N , N′-diacetic acid monohydrochloride hydrate; Strem Chemicals , Inc . , http://www . strem . com ) . For elemental analysis after 5-weeks , plants were nondestructively sampled by removing one or two leaves . The plant material was rinsed with 18 MΩ water and placed into Pyrex digestion tubes . To alter the concentration of Mo in the watering solution , 0 . 25× Hoagland solution was made without any MoO42− which was then added back in varying concentrations from a solution of dissolved MoO3 . At Purdue University seeds of Col-0 and Ler-0 were germinated in the dark at 4°C for 2 days on 0 . 5× Murashige and Skoog media with 0 . 5× MS Vitamins ( Caisson Laboratories , Inc . ) , 3 mg/L Benomyl [methyl 1- ( butylcarbamoyl ) -2-benzimidazolecarbamate; Sigma , http://www . sigmaaldrich . com ) , and 10 µM Fe-HBED solidified with 1 . 5% agar in 1 . 5 ml eppendorf microcentrifuge tubes before being transferred into growth conditions described above . For the first 5 days after germinationthe tubes containing the seedlings were kept covered to maintain high humidity . The bottom of each tube was then removed and the tube inserted into foam floats placed in tubs containing ∼4 L of 0 . 25× Hoaglands solution with 10 µM Fe-HBED . The solution was changed weekly until harvest at 4 weeks from planting . Leaves were harvested as described above . The roots were rinsed twice in distilled water and a third time in double distilled water before being put in Pyrex digestion tubes . At Rothamsted seeds of Col-0 , mot1-1 and st5 . 1mot1-1 were germinated on 0 . 5% agarose in 0 . 5 ml tubes ( with excised lower portion ) in racks placed on boxes ( with transparent lids ) filled with 700 ml nutrient solution: 1 . 0 mM KNO3 , 0 . 5 mM Ca ( NO3 ) 2 , 1 . 0 mM KH2PO4 , 1 . 0 mM MgSO4 , 100 µM FeEDTA , 30 µM H3BO3 , 5 µM MnCl2 , 1 µM ZnCl , 1 µM CuCl and 0 . 1 µM Na2MoO4 . To synchronize germination , the boxes were incubated at 4°C over night and then transferred to a controlled growth room under light/dark cycle 16/8h at 20°C . The nutrition solution was exchanged 2 times per week . After 3 weeks , shoot and root materials were harvested . Roots were washed twice by dipping in deionised water and dried on paper towels before freezing in liquid nitrogen . At Purdue University tissue samples were dried at 92°C for 20 h in Pyrex tubes ( 16×100 mm ) to yield approximately 2–4 mg of tissue for elemental analysis . After cooling , seven of approximately 100 samples from each sample set were weighed . All samples were digested with 0 . 7 ml of concentrated nitric acid ( OmniTrace; VWR Scientific Products; http://www . vwr . com ) , and diluted to 6 . 0 ml with 18 MΩ water . Elemental analysis was performed with an ICP-MS ( Elan DRCe; PerkinElmer , http://www . perkinelmer . com ) for Li , B , Na , Mg , P , S , K , Ca , Mn , Fe , Co , Ni , Cu , Zn , As , Se , Mo , and Cd . All samples were normalized to calculated weights , as determined with an iterative algorithm using the best-measured elements , the weights of the seven weighed samples , and the solution concentrations , implemented in the PiiMS database [4] . Alternatively , for the data shown in Figure 2B , samples were analyzed at Rothamsted . Frozen plant material was homogenized with a mortar and pestle in liquid nitrogen . After transfer into 2 ml tubes , the plant material was freeze-dried . Samples were acid digested ( 83% HNO3 & 13% ( 70% ) HClO4 ) and analyzed by ICP-MS ( Agilent ICP-MS 7500ce , Agilent Technologies , Santa Clara , CA , US ) [28] . Broad sense heritability was calculated using a general ANOVA to account for line and growth tray variation . A linear mixed model adjusting population structure confounding effects as in [14] , [29] was used to test the marker-trait associations . A brief summary of the model is given below , where y is the vector of phenotype , α is the vector of fixed allele effects , β is the vector of subpopulation effects , u is the vector of random effects reflecting the genome-wide relatedness , and X , Q , Z are known incidence matrices relating the observations to fixed and random effects , respectively . The variance of phenotype was modeled asThus , the phenotypic variance can be partitioned into two parts: σ2g , the genetic variance attributable to genome-wide effects , and σ2e , the residual variance . The Q and K* matrix was the same as in [14] , with Q being the population assignments by Structure [30] and K* being the kinship coefficient matrix , estimated as the proportion of shared haplotypes between individual pairs . DNA microarray-based BSA was realized as previously described [11] , [31] . Briefly , SFPs were identified between Col-0 and Ler-0 by hybridizing labeled genomics DNA from each one of the accessions to Affymetrix ATH1 microarrays and comparing them to Col-0 hybridizations downloaded from http://www . naturalvariation . org/xam . Two genomic DNA pools from an F2 population of a cross between Ler-0 and the 14501 mutant in the Col-0 background were created and hybridized to separate DNA microarrays . Each one of the pools contained plants with either shoot Mo contents similar to Col-0 ( “control” pool ) or low shoot Mo contents similar to Ler ( “Low Mo” pool ) . At loci unlinked to the low Mo phenotype , the pools should have equivalent amounts of each genotype , and the hybridization signal at each SFP should be intermediate between the two parent accessions , for an average difference between the two DNA microarrays of zero . At linked loci , the difference between the two DNA pools should be approximately two-thirds the difference between the parent accessions . By smoothing the signal across multiple SFPs , noise is reduced and the peak of the differences in hybridization signal will correspond to the chromosomal region of the loci controlling the low Mo trait . Raw hybridization data ( . CEL files ) for each probe on the ATH1 DNA microarrays used in these experiments have been submitted to the Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo ) for public distribution ( Reference #GSE10039 ) . Seedlings to be grafted as previously described by [10] . Plants were harvested for analysis of shoot Mo content 4 weeks after transfer to soil . Postharvest analysis of graft unions was performed under the stereoscope to identify any adventitious root formation from grafted individuals . Individuals with adventitious roots emerging at or above the graft union or without a clear graft union were eliminated from subsequent analyses . Plants were first analyzed by ICP-MS and further used to determine MOT1 transcript levels as described previously by Rus et al . , [10] . For MOT1 ( At2g25680 ) transcript quantification the following primers were used: forward primer 5′-GGT GGG TGT GTG GCA CTG T-3′ and reverse primer 5′-AGC ACA CCA ACC GGA AAC TT-3′ . Four reactions were done per biological sample and three independent replicate samples per genotype were used . to evaluate the transcript abundance of MOT1 . Data was analyzed using the SDS software ( Applied Biosystems version 1 . 0 ) , following the method of [32] . Ct values were determined based on efficiency of amplification . The mean Ct values were normalized against the corresponding ACTIN 1 gene ( At2g37620 ) and Ct values calculated as ( Ct AtMOT1- Ct Actin1 ) . The expression of MOT1 was calculated as the 2−ΔCt method . The final standard error was estimated by evaluating the 2−ΔCt term using 2−ΔCt plus standard deviation and 2−ΔCt minus the standard deviation [32] . Genomic DNA was isolated from 10-day-old Arabidopsis Col-0 seedlings using a DNeasy plant mini kit ( Qiagen , Valencia , CA ) . MOT1 clones containing approximately 1 kb upstream to the ORF , and approximately 0 . 250 kb downstream to the ORF were sequenced . The following primers were used to amplify different PCR products spanning the selected genomic region . Forward primers were: FP1 5′-CGA GCA AAC TAG AAA AGA GAT CG-3′ , FP2 5′-CAG GTG TTA GCT GTT TAA CTG-3′ , FP3-5′-GCG ATT TCG TCT ACC GC-3′ and FP4 5′-GGT TTG GAG CAA TGC C-3′ . The corresponding reverse primers were: RP1 5′-CAA AAA CCA AAG CGT TGA CA-3′ , RP2 5′-GGA CAC CGT AAA CTG C-3′ and RP3 5′-GGG AAG ATG TAG GTG G-3′ . Conditions used in all PCR reactions were: initial denaturation 94°C followed by 30 cycles of 94° C 30 sec , 50° C 40 sec , 72° C 1 min 30 sec and final extension at 72° C for 10 min . All PCR products were cloned using TOPO XL PCR cloning kit ( Invitrogen Corporation , Carlsbad , CA ) , and sequenced using Big dye terminator v 3 . 0 method ( Applied Biosystems Foster city , CA ) with T7 and universal M13 primers . The MOT1 cDNA was isolated by RT-PCR . Oligonucleotide primers MOT1-5′ ( 5′-ATG GAG TCT CAG TCT CAG AGA GGT CAA-3′ ) and MOT1-3′ ( 5′-TCA AGC ATG TTC ACC GGA TTG CGG GGG-3′ ) were designed to amplify the coding sequence . Total RNA was extracted from roots of 4-week-old Arabidopsis plants grown on solidified one-half Murashige and Skoog + B5 vitamin medium using RNeasy Plant Mini Kit ( Qiagen , Hilden , Germany ) . Reverse transcription was carried out at 50°C in 20 ul solution containing 1 ug total RNA , 0 . 5 ug oligo ( dT ) primer , 10 nmol dNTPs and 200 units Superscript II reverse transcriptase ( Gibco BRL , Rockville , MD , USA ) . PCR was carried out on the first-strand cDNA using Ex Taq DNA polymerase ( Takara ) following the manufacturer's protocols . After a standard PCR of 30 cycles , aliquots were run on an agarose gel . The fragment of putative size was cloned into a pGEM-T Easy vector ( Promega ) and identified by DNA sequencing . For yeast expression studies , the MOT1 cDNA insert from pGEM-T Easy vector ( Promega ) was amplified using PCR with the primers ( 5′-CGG ACT AGT CCA TGG AGT CTC AGT CTC AGA GAG GT-3′ ) and ( 5′-TCC GGA TCC TCA AGC GTA ATC AGG AAC ATC GTA AGG GTA AGC ATG TTC ACC GGA TTG CGG GGG-3′ ) , with an HA tag encoded by the reverse promoter . The fragments were cloned into p416 GPD ( -uracil ( ura ) ) ( ATCC nos . 87360 ) . Two separate MOT1::GFP fusion protein constructs were cloned , one at Purdue ( v1 ) and one at Rothamsted ( v2 ) . To construct the MOT1::GFP ( v1 ) fusion protein , MOT1 was amplified from cDNA using PCR with 5′-TGC TCT AGA GCA TGG AGT CTC AGT CTC AGA GAG GT-3′ ( MOT1 , 5′ primer ) and 5′-TCC GGA TCC TTC CTC CTC CAG CAT GTT CAC CGG ATT GCG GGG G-3′ ( MOT1 , 5′ primer ) primers . The PCR product was confirmed by sequencing and cloned into the XbaI and BamHI sites of the 326-SGFP plasmid ( kindly provide by Inhwan Hwang , POSTECH , Korea ) to created chimeric GFP fusion constructs under the control of the 35S promoter . The use of the two specific oligonucleotide primers 5′-GGA TCC GCA GTC GAG CTT ACC AAT TCT C-3′ and 5′-GGA TCC GAA ACA GAG CAA TAA GCG TAT CTC-3′ allowed isolation of a 1778 bp MOT1-promoter DNA fragment ( −1801 to −21 from the translation start site ) from Arabidopsis Col-0 genomic DNA by PCR . After subcloning in pGEM-Teasy ( Promega ) and sequencing , the BamHI restriction sites in the primer sequences were used for further cloning of the promoter DNA fragment into the BamHI site of the pBI101 vector ( Clontech , Mountain View , USA ) harboring the promoterless β-glucuronidase gene . For preparation of the MOT1::GFP ( v2 ) construct the full ORF without the stop codon ( TGA ) was amplified from total cDNA by using the two specific oligonucleotide primers 5′-GGA TCC AAT GGA GTC TCA GTC TCA GAG AG-3′ and 5′-GGA TCC AGC ATG TTC ACC GGA TTG CGG GG-3′ . The Bam HI restriction site replacing the stop codon allowed the cloning of the coding DNA in frame to the GFP gene in the binary vector pBIN19 [33] , which contained the cauliflower mosaic virus 35S-promoter-MCS-GFP cassette . The binary plasmids were transformed into Agrobacterium tumefaciens GV3101 ( pM P90 ) [34] by the freeze/thaw method . Arabidopsis Col-0 plants were transformed using the floral dip method ( Clough and Bent , 1998 ) . Transgenic plants were selected on solidified ½ MS media [35] containing 50 mg/l kanamycin sulphate . Kanamycin-resistant homozygous T3 progenies derived from 5 independent transgenic lines were used for analysis . Total RNA and first strand cDNA was isolated from Col-0 seedlings as described for qRT-PCR . The first strand cDNA was used as a template to synthesize ST 5 . 1 cDNA using the following primers: SpeI-FP 5′-CGA CTA GTA TGG CGG TCG CAA TAT CTG GGA GT-3′ BamHI-HA-3′RP 5′-CGG GAT CCT TAA GCG TAA TCC GGA ACA TCA TAC GGG TAG TTT GCG AGT ATC GGG TT3′ . The reverse primer contained the HA tag . PCR conditions are as follows: initial denaturation 94°C followed by 30 cycles of 94° C 30 sec , 50° C 40 sec , 72° C 1 min 20 sec and final extension at 72° C for 10 min . PCR products were cloned and sequenced as before . Plasmid DNA containing ST5 . 1 HA were digested with SpeI and BamHI and the cDNA fragment cloned into p416 GPD ( Glycerol 3-phosphate dehydrogenase promoter ) using the same restriction enzymes . The ligated products were chemically transformed into E . coli DH5α and the positive clones were identified using a vector primer ( FP-5′-AAT GGA GTG ATG CAA CCT-3′ ) and the reverse primer BamHI-HA-3′RP HA . Transgenic Arabidopsis Col-0 plants were germinated and grown vertically on 0 . 5× MS agar plates containing 1% sucrose . Plants were transferred to 5 cm Petri dishes containing 2–4 ml staining solution , containing 0 . 05% 5-bromo 4 chloro 3 indolyl β-D glucuronic acid in 0 . 1 M sodium phosphate , 10 mM EDTA , 0 . 5 mM potassium ferricyanide , 0 . 5 mM potassium ferrocyanide , 0 . 3% triton X-100 and 10% methanol ( pH 7 . 5 ) . Vacuum infiltration for 1 min was performed twice and the staining reaction allowed to proceed in the dark at 37°C until the blue indigo color appeared . After straining plant samples were rinsed twice in 70% ethanol for 30 min then in 100% ethanol until chlorophyll was removed . After staining and destaining samples were analysed by optical microscopy . The MOT1::GFP ( v1 ) construct and a mitochondria marker , F1-ATPase-γ-RFP [36] , were co-transformed into purified Arabidopsis leaf protoplasts using polyethyleneglycol method [37] . The protoplasts were incubated for 20 hours at 22°C , and examined under an epi-fluorescence microscope , Nikon Eclipse 80i ( Nikon USA , Melville , NY ) . Images of the mitochondria marker and the green fluorescence of MOT1::GFP were sequentially taken within 1 second in the same cell , because plant mitochondria actively moves along F-actins [38] . The filter sets used were 31303 for RFP and 31001 for GFP from Chroma Technol Corp ( Rockingham , VT ) . The plants containing stably transformed 35S-Promoter-MOT1::GFP ( v2 ) fusion constructs described above were grown vertically on B5 plates under continuous light for two weeks . The seedlings were harvested , and incubated with 0 . 2 µM MitoTracker Red CM-H2X ROS ( Invitrogen , Carlsbad , CA ) for 10 min at room temperature , and examined using a confocal microscope , Nikon Eclipse 80i ( Nikon USA , Melville , NY ) using green HeNe laser line ( 543 nm ) for the GFP and red HeNe laser line ( 633 nm ) for MitoTracker . The MOT1-HA and ST5 . 1-HA containing plasmids were transferred into the Saccharomyces cerevisiae yeast wild-type strain BY4742 ( MATα his3 leu2 lys2 ura3 ) by the lithium acetate method [39] . The transformants were selected on SD minimal media containing 20 g l−1 glucose or glactose and required amino acids . Yeast transformants were first pre-grown at 30°C and 320 rpm in four different culture tubes per construct for one day in SD URA- minimal media . Each independent culture was used to inoculate 8 wells of a 2 mL square deep well 96-well plate ( 600 µL per well defined minimal growth medium inoculated with 20 µL of yeast culture ) . The plate was covered with a breathable seal and incubated at 30°C with shaking at 400 rpm for 24 hr . An AcroPrep® 96 PVDF ( Polyvinylidene fluoride ) filter membrane ( 0 . 45 µm , 350 µL ) micro well plate ( Pall Life Sciences ) was wetted by partially filling with methanol and filtered using a vacuum block . Wells were then filled with deionized water and similarly filtered . Yeast culture samples , grown in the 2 ml deep well 96-well plates , were transferred to the 96-well filter plates ( 200 µL/well ) and the yeast growth media removed by filtration . The same amount of culture samples was concurrently transferred into Clear View® micro plate ( Whatman ) for optical density measurement with a Dynex Opsys MR plate reader . Yeast cells on the membrane were washed four times , respectively , with 1 µM EDTA , pH 8 , and deionized water . 96-well filter plates were dried at 88°C for 2 h . Concentrated nitric acid ( 100 µL/well ) was added , the plate covered with polypropylene lid , and placed in a preheated heating block set at 88°C for 35 min to digest the yeast cells . The digested samples were diluted with a gallium internal standard ( 5 ppb final concentration ) and filtered into a 2-mL square deep well 96-well plate containing Triton X-100 ( 0 . 025% , 300 µL ) . The final dilution volume was 1 . 6 mL , giving an acid and Triton X-100 concentration of about 6 . 5% and 0 . 005% , respectively . The filtered culture medium , as well as the original medium , was diluted 50 times but otherwise the matrix prepared in the same manner as the samples . The samples and the media were analyzed on a Perkin Elmer Elan DRC-e ICP-MS coupled with Elemental Scientific SC-2 autosampler and Apex Q nebulizer sample introduction system , and the following analytes quantified: Na , Mg , P , S , K , Ca , Mn , Fe , Co , Ni , Cu , Zn , Mo and Cd . Solution concentrations for all yeast sampleswere normalized to the measured optical densities of the corresponding yeast culture . The amount of yeast was converted from OD to number of cells using the conversion factor of 1ml of 1 OD culture containing 3×107 cells . The Arabidopsis Biological resource Center ( ABRC ) accessions used in this paper were CS22660 , CS20 , CS1965 , CS1968 , CS1971 , CS1986 , CS1925 , CS1939 , CS1975 .
|
Plants must acquire all the mineral nutrients they require for survival from the complex chemical and biological environment of the soil . A better understanding of the way plants do this would not only allow improvements in sustainable agricultural productivity , but could also improve human health through enhancement of the nutritional quality of foods . One such essential mineral nutrient required by plants is molybdenum ( Mo ) , which is needed as a cofactor in several critical biochemical reactions , including the utilization of nitrogen from the soil . By searching through numerous natural populations of Arabidopsis thaliana ( Arabidopsis ) , we were able to identify a DNA deletion that drives the natural variation in Mo accumulation observed in these populations . This deletion reduces expression of a gene ( MOT1 ) that the authors establish to encode a mitochondrially localized molybdenum transporter . Loss of expression of MOT1 in the roots of Arabidopsis causes a significant reduction in whole plant Mo accumulation , though the mechanism by which this Mo transporter regulates whole plant Mo from the mitochondria remains to be established .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"plant",
"biology/plant",
"genetics",
"and",
"gene",
"expression",
"plant",
"biology/plant",
"biochemistry",
"and",
"physiology"
] |
2008
|
Variation in Molybdenum Content Across Broadly Distributed Populations of Arabidopsis thaliana Is Controlled by a Mitochondrial Molybdenum Transporter (MOT1)
|
Heavy-ion therapy has an advantage over conventional radiotherapy due to its superb biological effectiveness and dose conformity in cancer therapy . It could be a potential alternate approach for hydatid cyst treatment . However , there is no information currently available on the cellular and molecular basis for heavy-ion irradiation induced cell death in cystic echinococcosis . LD50 was scored by protoscolex death . Cellular and ultrastructural changes within the parasite were studied by light and electron microscopy , mitochondrial DNA ( mtDNA ) damage and copy number were measured by QPCR , and apoptosis was determined by caspase 3 expression and caspase 3 activity . Ionizing radiation induced sparse cytoplasm , disorganized and clumped organelles , large vacuoles and devoid of villi . The initial mtDNA damage caused by ionizing radiation increased in a dose-dependent manner . The kinetic of DNA repair was slower after carbon-ion radiation than that after X-rays radiation . High dose carbon-ion radiation caused irreversible mtDNA degradation . Cysts apoptosis was pronounced after radiation . Carbon-ion radiation was more effective to suppress hydatid cysts than X-rays . These studies provide a framework to the evaluation of attenuation effect of heavy-ion radiation on cystic echinococcosis in vitro . Carbon-ion radiation is more effective to suppress E . multilocularis than X-rays .
Alveolar echinococcosis , which is caused by larvae of E . multilocularis , is common in Europe , China , and Siberia . It's a serious disease to human that has a significantly high fatality rate . The most common form of treatment for cystic echinococcosis is surgical removal of the cysts combined with chemotherapy using albendazole and/or mebendazole . However , it would not be practical for surgery if cysts are in risky locations or in multiple organs or tissues . In this case , PAIR ( puncture-aspiration-injection-reaspiration ) combined with chemotherapy become an alternative options of treatment [1] . There are currently new studies looking for new treatment by mean of high-energy electromagnetic waves ( such as X-rays ) . In vivo and in vitro studies shows that X-rays radiation could significantly increase protoscolex mortality and inhibit cysts growth [2] , [3] . Recent case study represented radiation therapy could be an alternative treatment modality for hydatid disease of the chest wall after medical and surgical therapy failure [4] . However , conventional megavolt photon therapy such as X-rays may not be a good choice for hydatid cyst treatment , because E . multilocularis is mostly localized in the liver and lungs and the therapeutic effect of X-rays is limited and unsatisfactory on liver and lungs on the basis of their poor tolerance to irradiation [5]–[7] . Heavy-ion radiotherapy has been proved to be an ideal treatment for lung cancer and hepatocellular carcinoma due to its excellent local control and survival in cancer patients [8] , [9] . Higher-energy beams of charged nuclear particles could offer significant advantages for the deep-seated hydatid cysts treatment in comparison to conventional megavolt photon therapy . Heavy-ion radiation could be an effective way for treatment of hydatid cysts , taking its full advantage of well-defined range , small lateral beam spread and an enhanced biological effectiveness . Until now there are no in vitro studies reported the cellular and molecular effect of heavy-ion radiation on hydatid cysts . This is the first report on the cellular damage induced by carbon-ion irradiation in hydatid cysts .
All animals experiments described have been conducted according to the Guideline on the Humane Treatment of Laboratory Animals stipulated by the Ministry of Science and Technology of the People's Republic of China ( MOST ) and were approved by the Institutional Animal Care and Use Committee ( IACUC ) of Institute of Modern Physics , CAS for the use of laboratory animals . The Regulations for the Administration of Affairs Concerning Experimental Animals ( 1988 . 11 . 1 ) is affiliated with Institute of Modern Physics , CAS . Hydatid cysts of E . multilocularis were freshly isolated from livers of female mice after secondary infection with hydatid cyst tissue homogenates , which were originally obtained from a sheep naturally infected with E . multilocularis in Qinghai , China . Collected cysts were rinsed with sterile saline for several times and then cultivated with RMPI 1640 medium at 37°C in a 5% CO2 incubator . X-rays were generated with an X-rays machine ( FAXITRON RX650 , USA ) operated at 130 keV . An exposure-rate meter ( AE-1321 M , Applied Engineering Inc , Japan ) was used for the dosimetry . The dose rates were 1 . 3 Gy/min . Carbon-ion irradiations were performed at room temperature at the Heavy Ion Accelerator Center ( HIRFL ) of the Institute of Modern Physics ( Lanzhou China ) with 300 MeV/n carbon ions; the LET value for carbon ions was 40 KeV/µm . The dose rates were 1 Gy/min . LD50 for hydatid cysts was monitored by protoscolex death after irradiation . The viability of protoscoleces 24 h after irradiation was assessed by observation under microscopy . The corresponding numbers of viable/non-viable protoscoleces were determined in 10 randomly chosen fields by phase contrast microscopy at 10×magnification . Cysts were fixed in 4% glutaraldehyde ( 24 h ) and 1% osmium tetroxide sequentially , dehydrated with acetone gradient , and embedded in Epon 812 epoxy resin . The 60-nmsections were cut with an ultrathin section machine , stained with uranium and lead electron stains , and observed under the TEM . Cysts were also observed under light microscopy according to standard protocol . Western blotting protocol followed standard protocol . Briefly , protoscoleces were lysed in buffer containing 50 mmol/l Tris at pH 7 . 4 , 50 mmol/l NaCl , 0 . 1% Triton X-100 , 0 . 1% SDS , 0 . 3 mmol/l sodium orthovanadate , 1 mmol/l dithiotheritol , 10 mg/l leupeptin , and 5 mg/l aprotinin . Protein concentrations of lysates were determined using a BCA protein assay kit ( Pierce , Rockford , IL , USA ) . An aliquot of each extract ( 40 µg protein ) was fractionated by electrophoresis in an SDS–polyacrylamide gel and transferred to a PVDF membrane . Membranes were blocked with 10 ml TBST containing 0 . 5 g FBS at room temperature for 2 h , followed by incubation with antibodies against Caspase-3 ( Biosynthesis Biotechnology , Beijing , China ) at 4°C overnight . After washing with TBST for 30 min , appropriate HRP-conjugated secondary antibody was added to the membranes , which were incubated at room temperature for 1 h . Membranes were washed three times for 15 min each with TBST . Reactive proteins were visualized using a chemiluminescence kit ( Santa Cruz Biotechnology , Santa Cruz , CA ) according to the manufacturer's instructions . Detections were performed according to a previous description using the caspase-3 activity colorimetric assay kit instruction ( Beyotime Institute of Biotechnology , Jiangsu , China ) [10] . Briefly , 3-mg samples were added with 100-µl lysis buffer , ground , kept on ice for 15–20 min and centrifuged at 4°C , 17 , 000×g for 15 min . The supernatants were harvested and added into the reaction system on an assay plate with a control group according to the kit's instruction; plates were incubated at 37°C for 15 h and detected with a microplate reader for the absorbance at 405 nm ( A405 ) . The activated caspase-3 in samples catalyzed colorless substrate Ac-DEVD-pNA into yellow pNA: which concentrations could then be calculated according to pNA standard curve and sample A405; the activity of caspase-3 in samples was finally deduced based on the pNA concentration . Long PCR for mtDNA damage evaluation was performed using the GeneAmp XL PCR kit ( PerKin–Elmer , Boston , MA ) . Quantitative long PCR were performed in an Eppendorf Mastercycler PCR system ( Eppendorf , Hamburg , Germany ) . The PCR cycle test was performed before to ensure the PCR in the exponential phase . The sequence information of the primers was listed in Table 1 . Briefly , a highly conserved nuclear single copy nuclear gene , DNA-directed RNA polymerase II ( rpb2 ) , was used as a reference for nuclear DNA copy number; Cytochrome c oxidase subunit II ( Cox2 ) , which encode an essential part of mitochondrial Electron Transport Chain ( ETC ) , was used for mtDNA copy number reference . The PCR was initiated with a 75°C hot-start addition of the polymerase and allowed to undergo the following profile: an initial denaturation for 1 min at 94°C followed by 25 cycles for large fragments or 20 cycles for small fragments of 94°C denaturation for 15 sec and 68°C extension for 15 min . A final extension at 68°C was performed for 10 min at the completion of the profile . An aliquot of each PCR product was resolved on a 1% vertical agarose gel and electrophoresed in TBE for 4 hr . The gels were then digitally photographed and quantified with FluorChem FC2 ( Alpha Innotech corporation ) . The DNA damage was quantified by comparing the relative level of amplification of the large fragments of mtDNA ( 8761 bp ) normalizing this to the amplification of smaller ( 126 bp ) fragments . Statistical analysis was performed on the means of the data obtained from at least three independent experiments . Data are presented as means±SD . Student's t-test program in Microsoft Excel was used to detect statistical significance . p<0 . 01 was considered to be statistically significant .
The decrease in the vitality of the protoscoleces became evident after 3 h following the radiation exposure , since their movements has deceased ( data not shown ) . Loss of protoscolex viability in IR-treated cultures became significant after 24 h , with a 100% protoscolex death at 60 Gy X-rays radiation and at 30 Gy carbon-ion radiation , respectively . The dose lethality after ionizing radiation is presented in Figure 1 . The LD50 was 28 . 5 Gy for X-rays and 15 . 5 Gy for carbon-ion , respectively . The LD50 of X-rays irradiated cysts was significantly higher than that for carbon-ion irradiated cysts . Both X-rays and carbon-ion radiation induced dose-dependent mtDNA damage . Carbon-ion radiation caused relative higher mtDNA damage than X-rays at the same dose ( Figure 2A ) . MtDNA damage was repaired equal efficiently in both groups ( Figure 2B ) . Although mtDNA damage was completely repaired within 8 hours after radiation , we noticed significant decrease in the reference signal of cox2 4 hours after radiation , which was indicative of mtDNA copy number reduction . MtDNA copy number was then quantified by real-time PCR . Results showed that mtDNA copy number was dramatically reduced following radiation ( up to 30% ) and persisted for 24–48 hours before their gradual recovery ( Figure 2C ) . To further investigate the cellular effects of radiation on hydatid cysts , irradiated parasites and control were observed by thin section electron microscopy 24 hours after radiation . In control cysts , the germinal cells were intact , villi beneath the cuticle was visible . Organelles such as mitochondria and golgi with distinct outlines were distributed around the cell with ordered structure . In irradiated cysts , the germinal layer showed apparent damage . Abnormal appearance including sparse cytoplasms , absence of organelles such as mitochondria and golgi , and devoid of villi . These attributes are indicative of distressed or dying cells ( Figure 3 ) . The morphology was clear and intact under light microscope in control group . The protoscolex , germinal layer and cuticles were intact and clear . Abnormal protoscolex and detachment of germinal layer from cuticles were observed in X-rays irradiated group . Protoscolex contraction , loss of suction cups and scolex hooks were extensive in carbon-ion irradiated group ( Figure 4 ) . Since the metabolic pathway of programmed cell is currently unknown in E . multilocularis , caspase 3 , an effector molecule common to all know metabolic routs of apoptosis procedure was used as indicator of apoptosis . Caspase 3 expression was detected in hydatid cysts after 30 Gy carbon-ion or X-ray radiation , indicative of apoptotic index ( Figure 5A ) . The specific activity of active caspase 3 in irradiated cysts was higher than that in control . In addition , caspase 3 activity was more pronounced in carbon-ion irradiated cysts than X-rays irradiated cysts exposed to the same dose . Caspase 3 activity reached a plateau after 30 Gy radiation exposure ( Figure 5B ) .
In this study , we provide the first comprehensive study on cellular and molecular alternations in E . multilocularis hydatid cysts treated with high doses of X-rays or carbon-ion radiation . Exposure of high doses radiation causes the attenuation of hydatid cysts . The morphological changes of hydatid cysts after radiation were similar to other reports using drugs [10] , [11] , confirming the inhibitory effect of radiation on hydatid cysts . The significant decrease in mtDNA copy number indicates that the mtDNA degradation process becomes profound with increasing oxidative damage , presumably due to saturation of the repair capacity of mitochondria in E . multilocularis hydatid cysts . The persistent depletion in mtDNA content appears to be a direct consequence of active mtDNA degradation and may be the basis to the “persistent mtDNA damage” reported in several studies [12] , [13] . The profound mtDNA damage and degradation then could lead to mitochondrial dysfunction and persistent oxidative stress [14] . Carbohydrates , as the main energy source of E . multilocularis , can be catabolized by arobotic respiration or two complementary anaerobic pathways . Mitochondrial dysfunction induced by radiation , thus could lead to severe hydatid cysts growth inhibition and cell death due to its extensive parasitic life style . High level of apoptosis has been reported to be involved in hydatid cyst infertility in Echinococcus granulosus hydatid cysts [15] , [16] . Lymphocytes apoptosis by modulator of hydatid fluid is reported to be one of the survival mechanisms for hydatid cysts [17] . These studies indicate that apoptosis play an important bifunctional role in hydatid cysts survival . Oxidative stress play a pivotal role in apoptosis induction [18] . Hanhua et al . reported that H2O2 and dexamethasone could induce the cellular apoptosis of protoscoleces [10] . However , oxidative stress induced apoptosis in E . multilocularis has not been reported . Here we found that ionizing radiation such as X-rays and carbon-ion irradiation could efficiently induce apoptosis in E . multilocularis hydatid cysts , which may caused by exacerbated oxidative stress . Due to its physical and biologic advantages over conventional radiation therapy , heavy-ion therapy has high local control rates with relatively low toxicity compared with photo and proton radiation therapy [19] , [20] . Our results also showed that carbon-ion radiation caused more severe damage on hydatid cysts than X-rays . However , the side-effect of heavy-ion therapy should not be ignored . Severe late complications has been reported in patients who received high dose heavy-ion radiation for esophageal cancer [20] . Further in vivo experimental data for the effects of heavy-ion therapy on hydatid cysts should be provided . Our results provide a rationale future for exploring the application of radiotherapy as nonsurgical treatment method in treating this parasitic disease . Additionally , we find that carbon-ion radiation is more effective to damage hydatid cysts than X-rays , which may be a more suitable candidate for hydatid disease treatment .
|
Surgical removal of cysts may be impractical in cases that cysts are in multiple organs or tissues , or in risky locations . In that case , alternative treatment should be employed . Heavy-ion radiation could be an effective way for treatment of hydatid cysts , taking its full advantage of well-defined range , small lateral beam spread and an enhanced biological effectiveness . In this study , we found that carbon-ion radiation could result in extensive mitochondrial DNA damage and apoptosis in hydatid cysts . Cellular and ultrastructural changes were observed after ionizing radiation , which were indicative of cysts growth inhibition . To our knowledge , this is the first study reporting the biological effect of carbon-ion radiation on E . multilocularis hydatid cysts .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2013
|
Suppression of E. multilocularis Hydatid Cysts after Ionizing Radiation Exposure
|
The innate immune response to viruses is initiated when specialized cellular sensors recognize viral danger signals . Here we show that truncated forms of viral genomes that accumulate in infected cells potently trigger the sustained activation of the transcription factors IRF3 and NF-κB and the production type I IFNs through a mechanism independent of IFN signaling . We demonstrate that these defective viral genomes ( DVGs ) are generated naturally during respiratory infections in vivo even in mice lacking the type I IFN receptor , and their appearance coincides with the production of cytokines during infections with Sendai virus ( SeV ) or influenza virus . Remarkably , the hallmark antiviral cytokine IFNβ is only expressed in lung epithelial cells containing DVGs , while cells within the lung that contain standard viral genomes alone do not express this cytokine . Together , our data indicate that DVGs generated during viral replication are a primary source of danger signals for the initiation of the host immune response to infection .
The recognition of virus-specific pattern associated molecular patterns ( PAMPs ) is a pivotal event in the initiation of the host innate response to infection . In recent years , it has been established that most viral danger signals are derived from oligonucleotide structures exposed during the replication of the viral genomes [1] , [2] , [3] , [4] , [5] , [6] . However , most viruses produce proteins that antagonize and effectively delay signaling by the primary viral oligonucleotide sensor molecules retinoic acid inducible gene I ( RIG-I ) and melanoma differentiation–associated gene 5 ( MDA5 ) , allowing the virus to replicate to high titers and produce large amounts of danger signals prior to host intervention [7] , [8] . It is currently unclear how the host immune response overcomes viral evasion to initiate a protective antiviral response . Defective viral genomes ( DVGs ) arise when the viral polymerase loses processivity during virus replication at high titers , thereby generating truncated versions of the viral genome that contain deletions and/or complementary ends ( the later known as copy-back or snap-back genomes ) [9] , [10] . DVGs with the ability to interfere with standard virus replication were first described by Von Magnus in the early 1950s as the genomes of incomplete forms of influenza virus called defective interfering ( DI ) viral particles [11] . DVGs have been identified in multiple distinct viral families when the viruses are grown in the laboratory at high multiplicity of infection and span a broad range of hosts , from plants to mammals [12] . Importantly , DVGs are found in patients infected with hepatitis A [13] , hepatitis B [14] , [15] , hepatitis C [16] , HIV [17] , dengue virus [18] , and influenza virus [19] . However , the biological role of DVGs in the context of natural infections is not well understood . We and others have shown that stocks of Sendai virus ( SeV ) with a high content of copy-back DVGs with interfering activity trigger enhanced production of cytokines in vitro and more potently induce antigen presentation by mouse and human dendritic cells than do virus stocks lacking this kind of DVGs [20] , [21] , [22] , [23] , [24] , [25] . Our group has also demonstrated that in contrast to standard viral genomes , SeV copy-back DVGs induce the expression of MDA5 and of a number of other interferon-stimulated genes in the absence of type I IFN positive feedback [23] , [26] , [27] . Remarkably , SeV copy-back DVGs show this potent in vitro stimulatory activity even in the presence of functional viral encoded antagonists of the host response [23] , [24] . Here , we demonstrate that DVGs that trigger a robust activation of the transcription factors IRF3 and NF-κB accumulate at a high rate in infected cells becoming the main source of viral PAMPs . These DVGs arise naturally during acute respiratory viral infections in mice and provide essential stimuli for the initiation of the antiviral innate immune response in the lung . These data demonstrate the generation of DVGs in vivo during acute respiratory viral infections and suggest a critical role of these kinds of viral genomes in determining the quality of the host response to infection .
To further investigate the cellular mechanisms responsible for the efficient activation of the antiviral response by SeV DVGs , we evaluated the phosphorylation of transcription factors that are critical for the expression of type I IFNs in cells infected with equivalent amounts of infectious particles of a SeV strain Cantell stock containing high levels of copy-back DVGs ( SeV Cantell HD ) or with SeV Cantell depleted of DVGs ( SeV Cantell LD ) . Virus stocks were prepared from the same parental virus and their content of DVGs was determined by calculating the ratio of infectious particles to total particles ( ratios are specified in the material and methods section ) . In addition , copy-back DVGs of these stocks were identified by PCR . One predominant copy-back genome was present in cells infected with SeV Cantell HD ( amplicon of 278 bp ) , while no copy-back defective genome was detected in cells infected with SeV Cantell LD up to six hours after infection ( Figs . 1A and S1 ) . Cloning and sequencing of the 278 nt long amplicon confirmed that it corresponded to a previously described SeV Cantell copy-back DVG of 546 nt in length ( DVG-546 ) [28] . Phosphorylation of IRF3 and of the NF-κB repressor IκBα in response to SeV Cantell HD occurred rapidly and was sustained even in type I IFN receptor KO cells ( Ifnar1−/− ) ( Fig . 1B and C ) , while no phosphorylation of IRF3 or IκBα was observed for up to ten hours post-infection with SeV Cantell LD despite equivalent or higher expression of the viral protein Np ( Fig . 1D ) . Corresponding with the strong activation of transcription factors , Ifnb mRNA was expressed in Ifnar1−/− cells infected with SeV Cantell HD ( Fig . 1E ) . In contrast , type I IFN signaling was required for the cellular response to Newcastle disease virus ( NDV ) , an avian virus that only partially inhibits the type I IFN pathway , triggering the expression of type I IFN and other cytokines in the absence of DVGs . To further validate the role of SeV copy-back DVGs as triggers of type I IFN-independent antiviral responses , we cloned DVG-546 under the control of the T7 polymerase promoter and used this construct to prepare a SeV stock containing a single recombinant DVG ( rDVG ) . For this purpose we used SeV strain 52 that normally does not produce highly immunostimulatory copy-back DVGs [24] . Equivalent infectious units of SeV 52 and SeV 52 plus rDVGs had similar levels of total RNA ( Fig . 1F ) but infection with virus containing rDVGs strongly induced the antiviral response while virus that lacked DVGs did not ( Fig . 1G ) confirming the DVG immunostimulatory activity . In addition , presence of rDVGs significantly reduced the expression of SeV Np mRNA , demonstrating their strong interfering capacity ( Fig . 1G ) . Notably , mouse embryo fibroblasts lacking the type I IFN receptor expressed Ifnb mRNA in response to SeV 52 containing rDVG ( Fig . 1H ) and virus containing rDVGs triggered IRF3 phosphorylation independently of type I IFN feedback ( Fig . 1I ) , mirroring the response to SeV Cantell HD . Altogether , this evidence conclusively shows that SeV copy-back DVGs confer potent immunostimulatory ability to SeV stocks , independent of type I IFN feedback . Notably , potent Ifnb mRNA expression in response to SeV DVGs was independent of IRF1 , IRF5 , and IRF8 while only partially dependent on IRF7 ( Fig . S2 ) . This response was maintained in a variety of cell types ( Fig . S3 ) . To determine whether standard viral genomes and DVG RNAs have distinct intrinsic properties that explain their differential immunostimulatory activities , we compared naked RNA purified from a stock of SeV Cantell LD with in vitro transcribed DVG-546 . RNAs were transfected into cells before or after treatment with phosphatase or with RNase A that cleaves 3′ of single stranded C and U residues , and/or RNase V1 that cleaves base paired nucleotides . Both genomic RNA and DVG RNA were susceptible to treatment with phosphatase , as well as to treatment with RNases ( Fig . 2A ) , corresponding with the literature that demonstrates a crucial role for 5′-triphosphate-RNA in the induction of type I IFNs . While transfected DVGs induced stronger expression of Ifnb than an equivalent concentration of gSeV ( Fig . 2A ) , transfection of equivalent molar amounts of genomic and DVG RNA resulted in higher immunostimulatory activity of genomic RNA compared to DVG RNA ( Fig . 2B ) , demonstrating that SeV LD RNA can strongly trigger the host response to infection when delivered naked into the cells . Paradoxically , cells infected with SeV Cantell LD alone failed to induce strong type I IFN production even when used at a 10 times higher infectious dose than SeV Cantell HD ( Fig . 2C ) . Although the amount of gSeV RNA was significantly higher in cells infected with an moi of 15 of SeV Cantell LD compared with ten times less SeV Cantell HD at 6 h post-infection ( Fig . 2C ) , DVGs were only detected in cells infected with SeV Cantell HD , confirming a strong correlation between the presence of DVGs in the infected cells and the induction of the host response to infection . To determine whether the amount of total input viral RNA affected the immunostimulatory activity of SeV Cantell LD and HD , we measured the RNA content in equivalent infectious doses of these stocks . SeV Cantell HD had less than two fold higher the amount of total RNA than SeV Cantell LD and total RNA levels were equivalent between SeV Cantell LD and HD when LD was at twice the infectious dose ( Fig . 2D ) . Thus , differences in the net input amount of viral RNA cannot explain the more than >1000 fold difference in the expression of Ifnb mRNA between cells infected with equivalent infectious doses of SeV Cantell LD and HD . DVGs have an increased rate of replication compared to standard viral genomes due to their shorter size and promoter properties [29] . To determine whether DVGs replicate faster than gSeV , we calculated the rate of replication of gSeV and DVGs in cells infected with SeV Cantell HD . Although at an early time point more copies of gSeV than DVGs were detected in the cells , DVGs dominated by 12 h post-infection ( Fig . 2E ) accumulating at a 4 times faster rate than gSeV ( Fig . 2F ) . These data demonstrate that DVGs rapidly surpass the number of gSeV in infected cells , providing large quantities of pathogen associated molecular patterns . Supporting previous observations that DVGs from SeV stimulate the cellular antiviral response through signaling by RIG-I like receptors ( RLRs ) [1] , [23] , [24] , the essential RLR adaptor protein mitochondrial antiviral signaling protein ( MAVS ) was required for the activation of the transcription factors IRF3 and NF-κB and for expression of numerous antiviral and pro-inflammatory molecules upon infection with SeV Cantell HD . In contrast , MAVS was not required for the response to herpes simplex virus , which can trigger the host response independently of RLRs ( Fig . S4 ) . In addition , only DVG RNA , but not standard viral genomes , could be amplified from endogenous RIG-I and MDA5 complexes immunoprecipitated from infected cells ( Fig . S4C ) , supporting published evidence that DVGs bind to RIG-I preferentially over the standard viral genomes in infected cells [1] . As predicted , association of DVGs with RLRs correlated with type I IFN induction , but not with the level of virus replication ( Fig . S4D ) . Importantly , in addition to the primary role of RIG-I in the response to SeV DVGs , MDA5 participates in the induction of type I IFN in primary mouse lung fibroblasts infected with SeV HD ( Fig . S4E ) , similar to what we have observed in DCs [23] , [27] . Overall , these data demonstrate that DVGs are produced in the infected cells at a higher rate than genomic RNA and that DVGs are the predominant ligands for both RIG-I and MDA5 during SeV infection . Based on the potent ability of SeV stocks containing a high content of copy-back DVGs to induce the host response to infection in vitro [23] , [24] , [25] , [28] ( Fig . 1 ) and on our prior reports of strong host responses to DVGs regardless of the presence of functional virus-encoded antagonists [23] , [24] , we hypothesized that DVGs that arise in situ during viral infections provide essential stimuli to initiate an antiviral immune response . To test this hypothesis , we first determined if SeV strains that accumulate copy-back DVGs early in infection induced faster Ifnb mRNA expression in vitro than viruses with delayed DVG accumulation . For these experiments we used SeV preparations that did not show immunostimulatory activity or evidence of copy-back DVG accumulation by 2 h post-infection and all the viruses were used at a multiplicity of infection of 1 . 5 TCID50/cell . While standard viral genomes of all the different SeV strains used were detected at all tested time points , copy-back DVGs of different sizes were detected starting at 6 h post-infection in cells infected with SeV Z and at later time points in cells infected with SeV 52 , Enders , or Cantell LD in both murine lung epithelial cells ( TC-1 ) and bone marrow-derived dendritic cells ( BMDCs ) ( Fig . 3A and B and data not shown ) . Sequences of the starred PCR products confirming the amplification of copy-back DVGs are shown in Fig . S5 . Remarkably , accumulation of DVGs was directly associated with phosphorylation of IRF3 ( Fig . 3C ) and with the expression of Ifnb mRNA ( Fig . 3D ) , demonstrating that standard viral genomes alone are not sufficient to initiate this response during infection in vitro and strongly supporting a unique ability of naturally arising DVGs to initiate the cellular antiviral response . To evaluate the impact of DVGs during SeV infection in vivo , we infected mice with SeV Cantell HD or LD . Mice infected with SeV Cantell HD showed diminished morbidity than mice infected with the same infectious dose of SeV Cantell LD ( Fig . 4A ) despite equivalent levels of virus in the lungs at early times post-infection ( Fig . 4B ) , agreeing with reports of reduced virulence in virus stocks with a high content of DVGs [30] , [31] , [32] , [33] , [34] , [35] . Reduced virulence of SeV Cantell HD was associated with a stronger stimulation of the host antiviral response as shown by the expression of Ifnb mRNA ( Fig . 4C ) . To conclusively demonstrate the role of DVGs in diminishing virulence in vivo , we co-infected mice with SeV Cantell LD and purified viral particles containing DVGs ( defective particles; DPs ) . Confirming their critical role , DVGs reduced the pathogenicity of SeV Cantell LD in mice , while UV-inactivated DP particles did not provide significant protection ( Figs . 4D–F ) . Interestingly , infection in the presence of DPs resulted in reduced expression of SeV NP protein in the lung at day 7 post-infection , suggesting that in this system , DPs reduce virulence by interfering with virus replication . To determine whether immunostimulatory DVGs were generated in situ in the lung during infection , we infected mice with SeV Cantell LD , and we followed the appearance of copy-back DVGs in the lung by PCR . SeV copy-back DVGs were detected in whole lung homogenates at the time of high viral replication ( Fig . 5A ) . Notably , upon infection with SeV Cantell LD , a copy-back DVG of high molecular weight was detected at day 3 post-infection in the lung , while a DVG of low molecular weight ( amplicon of 278 bp ) that predominates in the parent stock of SeV Cantell HD ( Fig . 1A ) was only detectable at day 5 post-infection . Copy-back DVGs also appeared in the lung of mice infected with SeV 52 ( Fig . S6 ) , showing that DVGs naturally arise during infection in vivo independent on the virus strain . Interestingly , accumulation of copy-back DVGs during infection with SeV Cantell LD was associated with the expression of Ifnb and Il-6 mRNA in the lung ( Fig . 5B ) . To determine whether DVGs were necessary for the expression of antiviral cytokines in vivo , we took advantage of IFNβ-YFP reporter mice . To demonstrate that YFP expression serves as readout for DVG activity , we first infected BMDCs prepared from IFNβ-YFP reporter mice with SeV Cantell LD alone , or together with increasing doses of purified DPs . As shown in Fig . 5C , at 6 h post-infection , YFP was expressed only in the presence of DPs and in a dose-dependent manner , and the YFP expression was lost when UV-treated DPs were used . DPs alone were also able to induce YFP in a dose-dependent manner , albeit at much lower levels than during co-infection with SeV . These data agree with our previous reports that demonstrate that the immunostimulatory activity of DPs is greatly amplified during DVG replication by the cognate polymerase provided by co-infecting SeV [23] and validate the IFNβ-YFP reporter system as a readout for DVG activity . We then infected IFNβ-YFP reporter mice with SeV Cantell LD and analyzed viral genomes in YFP+ cells . We focused our analysis on the CD45− ( non-hematopoietic ) cellular fraction of the lung as SeV replicates predominantly in the lung epithelium [36] . Although full-length viral genomes were detected in both YFP+ and YFP− CD45− populations sorted three days after infection , DVGs were only found in YFP+ cells ( Fig . 5D ) , suggesting that the presence of DVGs promotes IFNβ production in response to virus infection in vivo . Together , these findings show that DVGs are normally generated in situ in the lung during respiratory infection with SeV , and that their accumulation is associated with the expression of IFNβ in the lung . To determine whether type I IFNs produced early upon infection promoted the generation of DVGs in the lung , we infected wild type or type I IFN receptor deficient mice ( Ifnar1−/− ) with SeV Cantell LD and analyzed the lungs at different times post-infection . As shown in Fig . 5E , DVGs accumulated in the lung at a higher rate in mice unable to respond to type I IFNs compared with wild type mice , corresponding with the predicted enhanced rate of virus replication in the lack of type I IFN signaling and demonstrating that type I IFNs are not required for the generation of SeV copy-back DVGs in vivo . To investigate whether the content of DVGs in IAV stocks affects virulence similar to SeV , we obtained IAV strain PR8 stocks with a high content of DVGs ( HD ) or lacking DVGs ( LD ) . The stock of IAV PR8 HD produced two predominant DVGs derived from the PA and PB1 genomic segments in infected cells , while no DVGs were detected in cells infected with IAV PR8 LD ( Fig . 6A ) ( Strategy for IAV detection and sequences for the IAV DVGs present in infected cells can be found in Fig . S7 ) . Mice infected with IAV PR8 HD showed reduced morbidity compared to mice infected with IAV PR8 LD ( Fig . 6B ) despite similar levels of virus replication ( Fig . 6C ) . Similar to SeV Cantell HD , reduced morbidity was associated with enhanced Ifnb mRNA expression in the lung ( Fig . 6D ) . To determine whether IAV DVGs were generated in situ in the infected lung , we tracked their appearance in mice infected with IAV PR8 LD . Accumulation of DVGs was clearly observed at day 3 post-infection ( Fig . 6E ) . Representative sequences of starred IAV DVGs products are shown in Fig . S8 . Similar to SeV infection , accumulation of DVGs corresponded with enhanced expression of mRNA for Ifnb and Il-6 ( Fig . 6F ) despite evidence of reduced genomic viruses at that time point ( Figs . 6E and F ) . These data demonstrate that DVGs are generated de novo in the lung during infections with IAV , and suggest an important role of these types of genomes in promoting the host response to IAV in vivo .
We have shown that DVGs are naturally generated in the lung during infection with SeV and IAV and provide primary danger signals for the triggering of the host response to infection . The generation of DVGs during virus growth in tissue culture is a highly conserved phenomenon among viruses of different species and is tempting to speculate that DVGs provide an evolutionary advantage to the virus by contributing to the preservation of both the virus and the host . Interestingly , immunostimulatory DVGs result from drastic truncations in the genome of the virus that render it a dead end product unable to persist in the absence of helper virus . It will be relevant to determine how DVGs relate to viral “quasispecies” that result from mutations as a consequence of having a viral polymerase with a lower fidelity and processivity [37] . Viral quasispecies have been shown to be essential for viral fitness and virulence [37] . Whether DVGs a tradeoff of this viral polymerase characteristic that enables more rapid virus evolution but makes the virus more vulnerable to innate immune detection remains to be established . Our data demonstrate that DVG recognition is not necessary for the response to NDV , while is required for the response to SeV . We speculate that the differential DVG requirement may be explained by the poor adaptation of the avian NDV to grow in mice while the murine SeV is fully adapted to grow in this species . A critical factor of this adaptation is the activity of the virally encoded V and C proteins that effectively block the induction of type I IFNs , as well as the type I IFN-mediated amplification of the type I IFN pathway [38] , [39] , [40] , [41] . As NDV is adapted to grow in birds , its antagonistic proteins are not fully functional in mammalian cells [42] allowing unrestricted production and amplification of type I IFNs . In contrast , the SeV C and V proteins very effectively block the cellular response to SeV [40] , [42] , [43] , [44] and no cellular response is observed unless DVGs are present . Interestingly , in the absence of type I IFN feedback ( or of the IFN-inducible transcription factor IRF7 ) SeV DVGs induce a more potent cellular response compared to NDV , suggesting that DVGs have a unique ability to bypass both virus antagonism and the requirement for IRF7 for strong type I IFN production . We have reported that SeV Cantell HD , but not NDV , has the ability to induce the expression of the viral sensor MDA5 independently of type I IFN feedback [27] , and that MDA5 is involved in the recognition of SeV DVGs ( [23] and data in Fig . S4 ) . Although it is unclear why this newly synthetized MDA5 is less susceptible to inhibition by the SeV V protein , preferential binding of DVGs to both RIG-I and MDA5 compared to standard SeV genomes , together with the availability of high levels of MDA5 , may explain the strong activation of transcription factors and type I IFN expression in response to DVGs , regardless of type I IFN feedback . Remarkably , DVGs arise in vivo independently of type I IFN feedback , demonstrating that DVGs do not appear in response to host pressure via type I IFNs . Notably , viruses containing DVGs have significantly reduced virulence . In a previous study , we reported that a SeV strain with a lower propensity to produce DVGs ( SeV 52 ) persisted longer in the lung than a SeV strain able to produce high levels of DVGs ( Cantell ) [26] . Consistently , we observed higher levels of viral NP protein at day 7 post-infection in the lung of mice infected with SeV Cantell LD , compared to mice infected with SeV Cantell LD plus DPs ( Fig . 4G ) . In additional studies , we have not observed significant differences in the rate of SeV-specific T cells in the lung of mice infected with SeV LD alone or in the presence of DPs ( data not shown and [26] ) . Although interpretation of this observation is complicated by the reduced amount of SeV antigen ( NP protein ) present in infection with SeV LD plus DPs , we favor the hypothesis that DPs diminish virulence by competing for the viral polymerase , thus interfering with the replication of the standard virus , a well-defined characteristic of DVGs . Additionally , enhanced production of type I IFNs in response to DVGs likely contributes to dampened viral replication . Highly immunostimulatory SeV DVGs are of the copy-back type . Intriguingly , during SeV infections in vivo , accumulation of DVGs of high molecular weight preceded the appearance of low molecular weight DVGs , suggesting that the smaller ones may be secondary to longer defective genomic products . Copy-back DVGs are not transcribed due to their promoter properties [10] , thus their stimulatory activity likely derives solely from their genomic composition . IAV DVGs are truncated versions of one of the genomic segments that have natural complementarity among their 3′ and 5′ ends providing the theoretical capacity to form structures similar to copy-back DVGs . Notably , it is apparent that both SeV genomic and DVG RNAs have the potential to induce a host response when delivered naked into the cells . Based on our data , we predict that in the context of infection , DVG RNA is more available for detection due to their enhanced rate of replication compared to standard viral genomes ( Fig . 2 ) . Interestingly , we have shown that DVGs have the ability to bypass viral-encoded antagonists of the immune response even upon overexpression of viral antagonistic proteins [24] . It remains to be investigated what is the molecular mechanism behind this DVG property . Notably , DVGs of different forms and compositions have been described in the sera of patients chronically infected with a number of different viruses [13] , [14] , [15] , [16] , [17] and DVGs of various viruses have been shown to promote persistent infections in tissue cultures [45] , [46] , [47] , [48] , [49] , [50] , [51] , [52] , [53] supporting a role for DVGs in the maintenance of chronic viruses . The role of naturally arising DVGs in promoting virus persistence in vivo remains to be investigated . In summary , we have demonstrated that DVGs arise naturally during an acute respiratory virus infection and that they play a critical role in regulating the virus-host cycle in vivo . Importantly , the recognition of DVGs as stimuli for the onset of immunity has multiple practical implications , most directly: ( i ) DVGs represent novel determinants of virus pathogenesis that could be targeted for therapy , and ( ii ) DVGs are novel candidate biomarkers to predict the outcome of infections and the rate of virus spread in the population .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol ( 803176 ) was approved by the Institutional Animal Care and Use Committee , University of Pennsylvania Animal Welfare Assurance Number A3079-01 . TC-1 cells ( mouse lung epithelial cells , ATCC , #DR-L2785 ) , LLC-MK2 cells ( monkey kidney cells , ATCC , #CCL7 ) , MDCK ( Madin-Darby canine kidney cells , kindly provided by Dr . T . Moran , Icahn School of Medicine at Mount Sinai ) , Baby hamster kidney-21 ( BHK-21 ) cells expressing the T7 RNA polymerase ( BSR-T7 ) ( kindly provided by Dr . C . Basler , Icahn School of Medicine at Mount Sinai ) , and WT and Ifnar1−/− mouse embryo fibroblasts ( kindly provided by Dr . B . tenOever , Icahn School of Medicine at Mount Sinai ) were cultured in DMEM supplemented with 10% fetal bovine serum , 1 mM sodium pyruvate , 2 mL L-Glutamine , and 50 mg/ml gentamicin . C57BL/6 mice were obtained from Taconic Farms , Inc . IFNβ-YFP reporter mice ( B6 . 129-Ifnb1tm1Lky/J ) were obtained from The Jackson Laboratories . Ifnar1−/− mice were a kind donation of Dr . Thomas Moran ( Icahn School of Medicine at Mount Sinai ) and were bred in our animal facility . SeV strains Cantell , 52 , Enders , and Z , and influenza A/PR8/34 virus were grown in 10 days hen embryonated eggs ( SPAFAS; Charles River Laboratories ) . SeV Cantell was passaged to retain its original high DI particle content ( HD ) or to deplete it of DI particles ( LD ) as we previously described [24] . In brief , SeV 52 , Enders , Z , and Cantell HD were grown in embryonated hen eggs inoculated with 30 , 000 medium tissue culture infectious dose ( TCID50 ) for 40 h . SeV Cantell HD TCID50 was calculated by end point dilution in LLCMK2 cells in the presence of trypsin , as described below [24] . SeV Cantell HD total particles were calculated by end point dilution of hemagglutination of chicken red blood cells . SeV Cantell HD stocks had consistently an infectious:total particle ratio of 5 , 000–15 , 000 . SeV 52 had an infectious:total particle ratio of 24 , 472 . SeV Enders had an infectious:total particle ratio of 38 , 746 . SeV Z had an infectious:total particle ratio of 391 , 553 . SeV Cantell LD was prepared by inoculating embryonated hen eggs with 3 TCID50 for 40 h . Under these conditions only 33% of the eggs grew virus . Allantoic fluid from those eggs was pooled and diluted 1/106 for subsequent inoculation into embryonated hen eggs in a total volume of 100 µl . Allantoic fluid containing virus ( 80% of the inoculated eggs ) was pooled and tittered as described below . SeV Cantell LD stocks had consistently an infectious:total particle ratio of 100 , 000–200 , 000 . IAV strain PR8 ( LD ) was grown by inoculating hen embryonated eggs with 30 , 000 TCID50 obtained directly from infected lung homogenates . Allantoic fluid containing the virus was collected 40 h later . Egg's allantoic fluid was snap frozen in an ethanol/dry-ice bath and stored at −80°C . IAV PR8 containing high dose of DI particles ( HD ) was kindly provided by Dr . Laurence C . Eisenlohr , V . M . D . , Ph . D ( Thomas Jefferson University ) . IAV HD was grown by inoculating hen embryonated eggs with 10 , 000 pfu of egg-passed virus . Eggs were incubated at 35°C and allantoic fluid containing the virus was collected 48 h later . IAV PR8 HD and LD stocks were originated from the same parent stock but were extensively passaged in the different conditions described . Permissive cells were infected with serial 1∶10 dilutions of lung homogenates or virus stocks in the presence of 2 mg/ml of trypsin to determine the medium tissue culture infectious dose ( TCID50 ) . LLCMK2 cells were used for SeV titration , while MDCK cells were used for IAV titration . After 72 h of incubation at 37°C , 50 µl of supernatant from each well was tested by hemagglutination of chicken red blood cells ( RBCs ) for the presence of virus particles at the end point dilution . To do this , 1∶4 dilutions of the cell supernatant were incubated in 0 . 5% chicken RBCs at 4°C for 30 min . Hemagglutination of RBCs indicated the presence of SeV or influenza virus particles . BMDCs were generated as previously described [24] . Detailed procedure can be found in the Supplemental Information Material and Methods . BMDCs were infected after 4 days in culture with viruses at an multiplicity of 1 . 5 as we have previously described [24] . For SeV infections , mice were anesthetized with tribromoethanol ( Avertin®; Acros Organics ) and inoculated in the nostrils with 30 µl of PBS containing 104 or 105 TCID50 of SeV . For IAV infections animals were infected intranasally with 100 TCID50/mouse in a 30 µl volume . Lungs were extracted at different times post-infection , homogenized in 0 . 1% w/v Gelatin-PBS and snap frozen in dry-ice/ethanol for preservation . Total RNA was extracted from cell lines or lungs with TRIzol ( Invitrogen ) according to the manufacturer's specifications and total RNA was reversed transcribed using the high capacity RNA to cDNA kit from Applied Biosystem . For sorted cells , 500 ng of RNA were reversed transcribed , for all other experiments 1–2 µg of RNA were reversed transcribed . cDNA was diluted to a concentration of 10 µg/µl and amplified with specific primers in the presence of SYBR green ( Applied Biosystem ) . For the detection of DVGs , isolated total RNA was reverse transcribed using Superscript III without RNase H activity , to avoid self-priming by the DVGs complementary ends and recombinant RNase H ( Invitrogen ) was added later to the samples . For the detection of the standard virus genome , the negative strand of the full-length genome was reverse transcribed with Transcriptor First Strand cDNA synthesis kit ( Roche ) . PCR detection for IAV was performed using as it has been previously described [54] . Primers and detailed PCR conditions can be found in the Supplemental Information Material and Methods . Detailed primers and PCR conditions can be seen in the Supplemental Information Material and Methods . Whole cellular extracts were prepared by lysing 3×106 of cells in a NP-40-based lysis buffer containing phosphatase inhibitors , proteinase inhibitors ( Roche and Thermo Scientific ) , and 0 . 5 M EDTA . The concentration of protein was measured by Bradford assay ( Themo Scientific ) . Samples ( 25 µg ) were boiled for 5 min and resolved on 10% Bis-Tris pre-cast gels ( Bio-rad ) . Resolved proteins were transferred to a polyvinylidene fluoride ( PVDF ) membrane ( Millipore ) . The membrane was blocked with 5% non-fat milk and immunoblotted with the indicated antibodies . Anti-rabbit IRF3 , anti-rabbit phospho-IRF3 ( Ser396 ) , anti-mouse IκBα , anti-mouse phospho-IκBα ( Ser32/36 ) , and anti-rabbit IgG ( HRP-conjugated ) were purchased from Cell Signaling . Anti-mouse GAPDH was purchased from Sigma . Anti-mouse IgG and anti-mouse IgG1 ( HRP-conjugated ) were purchased from Jackson Immunologicals . Lumi-Light western blotting substrate was used for HRP detection ( Roche ) . DP purification was performed as previously described [24] . In short , allantoic fluid from 100 infected hen eggs was pooled and concentrated by high-speed centrifugation . Pellets were suspended in 0 . 5 ml of PBS/2 mM EDTA and incubated overnight at 4°C in a 5–45% sucrose ( Fisher ) gradient that was prepared using a gradient maker ( BioComp ) . Gradients were centrifuged at 4°C for 1 . 5 h at 28 , 000 rpm and fractions containing low-density viral particles were collected , pelleted , suspended and re-purified using the same procedure . Collected low-density fractions were concentrated by centrifugation at 4°C for 2 h at 21 , 000 rpm . Pellets were suspended in PBS , snap frozen , and stored at −80°C . The content of DI particles was determined by calculating the ratio of infectious over non-infectious particles as described above . A 591 nt long product containing the sequence of the T7 promoter followed by the 546-nucleotide long copy back DVG from SeV Cantell , and flanked by the restriction enzymes SpeI and SapI at the 3′ an 5′ ends was synthetically synthetized ( DNA 2 . 0 ) and clone into the pSL1180 vector ( Amersham Pharmacia Biotech ) containing the sequences for the hepatitis delta virus ribozyme and the T7 polymerase terminator . In order to optimize the transcription of the DVG , 3 G residues were introduced downstream of the T7 promoter by site-directed mutagenesis ( Stratagene , CA ) using the oligonucleotides 5′CCACTAGTTAATACGACTCACTATAGGGACCAGACAAGAGTTTAAGAG-3′ and 5′CTCTTAAACTCTTGTCTGGTCCCTATAGTGAGTCGTATTAACTAGTGG-3′ . BSR-T7 cells were infected with a moi of approximately 66 of partially inactivated SeV strain 52 . Virus inactivation was performed by exposing diluted virus to UV light ( 254 nm model MRL-58 , UVP Upland , CA ) for 53 sec at a distance of 9 inches from the light source . Virus inactivation diminished the virus replication rate , while allowing the expression of viral proteins necessary for the replication of DVGs . Cells were incubated at 37°C for 1 h before transfection of 3 µg of vector encoding DVG . Transfection was performed with XtremeGENE transfection reagent ( Roche ) according to manufacturer instructions . Cells were cultured in Dulbecco's modified Eagle medium ( DMEM ) supplemented with 1% bovine serum albumin , 2% NaCO3 , 0 . 5 µg/ml trypsin ( Worthington ) and 0 . 1% penicillin-streptomycin ( Invitrogen ) and incubated in 7% CO2 at 37°C . Cells and supernatant containing SeV 52 and rDPs were harvested after 48 h and 200 µl of the suspension were inoculated in the allantoic cavity of 10-day embryonated hen eggs ( B & E Eggs , Silver Springs , PA ) . After 40 h allantoic fluid was harvest and 200 µl of undiluted fluid were inoculate in 10-day embryonated eggs for virus growth and egg inoculation was repeated for three consecutive passages . Allantoic fluid from the third passage was quick-frozen in dried ice/ethanol and used for infections . Presence of recombinant DVG was confirmed by PCR . No other DVGs were detected . DVG RNA was in vitro transcribed ( Ambion ) from the T7-DVG 546 plasmid . Standard genomic RNA was extracted from SeV Cantell LD stocks . To remove 5′-triphosphates , 1 µg of RNA was incubated with 10 U of Calf Intestinal Phosphatase ( New England BioLabs ) for 60 min at 37°C . To cleave single stranded RNA , 1 µg of RNA was incubated with 1 ng of RNase A ( Ambion ) for 15 min at room temperature . To cleave double stranded RNA , RNA was incubated with 0 . 1 U of RNase V1 ( Ambion ) for 15 min at room temperature . After treatments , RNA was purified using TRIzol or precipitation/inactivation buffer according to the manufacturer's specifications . LLC-MK2 cells were transfected with 250 ng or indicated doses of DVG and genomic RNA using lipofectamin 2000 ( invitrogen ) . At 4 hours post transfection , the cells were harvested and total RNA was isolated using TRIzol according to the manufacturer's specifications . Infected IFNβ-YFP cells were collected at 6 h post-infection . Reporter mice were sacrificed 3 days post-infection . Lungs were collected and dissociated with collagenase ( Roche ) , followed by suspension on 0 . 5 M EDTA and RBC lysis buffer . Single cell suspensions were then incubated with CD16/CD32 FcBlock for 20 min at 4C , followed by incubation with biotinylated mouse anti-CD45 . 2 . Washed cells were incubated with anti-biotin microbeads ( Miltenyi ) for 20 min and passed through a magnetic column for negative selection . CD45− cells were sorted based on YFP expression using a FACS Vantage SE sorter . Statistical analyses were performed as indicated in each figure . GraphPad Prism version 5 . 00 for Windows , GraphPad Software , San Diego California USA , www . graphpad . com , was used for analysis . Genes NCBI ID numbers . tuba1b: 22143; rps11:27207; ifnb: 15977; ifnl2: 330496; illb: 16176; il12b: 16160; tnf: 2926; Il-6 , 16193 .
|
In infections with viruses well adapted to the host virus-encoded proteins that delay the cellular response allow the virus to replicate to high titers prior to host intervention . The mechanisms overcoming viral evasion of the immune system and leading to the production of the primary antiviral cytokine IFNβ are not well established . Here , we demonstrate that truncated forms of viral genomes that are generated in situ during virus replication are a primary source of danger signals for the initiation of the host immune response to respiratory viral infections in vivo . Defective viral genomes ( DVGs ) are able to function as triggers of the immune response even in the absence of type I IFN signaling and are strong triggers of the host response to infection while overcoming viral antagonism .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Defective Viral Genomes Arising In Vivo Provide Critical Danger Signals for the Triggering of Lung Antiviral Immunity
|
Human β-defensin 3 ( hBD3 ) is a cationic host defence peptide and is part of the innate immune response . HBD3 is present on a highly copy number variable block of six β-defensin genes , and increased copy number is associated with the autoimmune disease psoriasis . It is not known how this increase influences disease development , but psoriasis is a T cell-mediated disease and activation of the innate immune system is required for the initial trigger that leads to the amplification stage . We investigated the effect of hBD3 on the response of primary macrophages to various TLR agonists . HBD3 exacerbated the production of type I Interferon-β in response to the viral ligand mimic polyinosinic:polycytidylic acid ( polyI:C ) in both human and mouse primary cells , although production of the chemokine CXCL10 was suppressed . Compared to polyI:C alone , mice injected with both hBD3 peptide and polyI:C also showed an enhanced increase in Interferon-β . Mice expressing a transgene encoding hBD3 had elevated basal levels of Interferon-β , and challenge with polyI:C further increased this response . HBD3 peptide increased uptake of polyI:C by macrophages , however the cellular response and localisation of polyI:C in cells treated contemporaneously with hBD3 or cationic liposome differed . Immunohistochemistry showed that hBD3 and polyI:C do not co-localise , but in the presence of hBD3 less polyI:C localises to the early endosome . Using bone marrow derived macrophages from knockout mice we demonstrate that hBD3 suppresses the polyI:C-induced TLR3 response mediated by TICAM1 ( TRIF ) , while exacerbating the cytoplasmic response through MDA5 ( IFIH1 ) and MAVS ( IPS1/CARDIF ) . Thus , hBD3 , a highly copy number variable gene in human , influences cellular responses to the viral mimic polyI:C implying that copy number may have a significant phenotypic effect on the response to viral infection and development of autoimmunity in humans .
HBD3 is a member of the β-defensin multigene family . The amphipathic , antiparallel β-sheet structure , stabilised by disulfide bonds , via six canonical cysteines is conserved throughout evolution [1] and between family members despite significant sequence diversity [2] . These powerful cationic antimicrobials directly kill fungi , bacteria and viruses , and recently it has become clear that this gene family has roles in other processes including male fertility , immunomodulation and inflammatory disease [3] . Defensins are primarily expressed from mucosal surfaces , some exclusively in the reproductive tract and others in skin , intestine and gingival surfaces [4–6] . Of β-defensin genes , hBD3 is probably the most versatile and studies both in vitro and in vivo demonstrate its ability to chemoattractant immune cells [7]; encourage wound healing [8] and modulate innate signalling [9–11] . HBD3 ( gene name DEFB103 ) and its mouse orthologue ( Defb14 ) are promiscuous ligands with ability to bind the receptors CCR6 , CCR2 , CXCR4 . In addition , a dominant mutation in the DEFB103 gene in dogs and wolves causes an increase in canine β-defensin 3 ( CBD103 ) peptide level allowing off-target binding to melanocortin receptor 1 ( MC1R ) , which results in black coat colour [7 , 12–15] . DEFB103 is present on hypervariable clusters of six β-defensin genes and alteration in copy number may influence innate immune responses . Increased copy number of the cluster is associated with psoriasis [16 , 17] . Increased defensin peptide level has been reported in serum of psoriasis patients , although the influence of defensins on the pathogenesis of the disease is not understood . Psoriasis is a T cell-mediated disease predominantly orchestrated by Th-17 cells . Amplification of the disease process is triggered by an initial phase modulated by an increase in innate immune signalling through pattern recognition receptors ( PRR ) such as toll like receptors ( TLR ) [18] . Psoriatic patients have an increase in dendritic cells and cationic antimicrobial peptides in the skin [19] . It has been shown that self and viral nucleic acids trigger an increase in the type I Interferon-α response of plasmacytoid dendritic cells ( pDC ) that are specialised cells for Interferon-α production through TLR9[20] . Blocking production of Interferon-α by these cells prevents T cell–dependent development of psoriasis in a xenograft model . The antimicrobial peptide LL-37 has been identified as a molecule that encourages recognition of self DNA and RNA through TLRs on pDC to induce release of Type I Interferon [21] . Recently antimicrobial peptides hBD2 , hBD3 and lysozyme have also been shown to bind self-DNA and activate pDC through TLR9 to release Interferon-α . The presence of these peptides in psoriatic plaques suggests a concerted role for them in the pathogenesis of psoriasis [9 , 19] . Here we determine the effect of hBD3 on the response of primary macrophages to known pathogen-associated molecular patterns ( PAMPs ) to investigate the influence of hBD3 on signalling from innate immune receptors . We have previously shown that hBD3 suppresses the TLR4-mediated response to bacterial lipopolysaccharide ( LPS ) which is mediated by both MyD88 ( Myeloid Differentiation Primary Response 88 ) and TICAM1 ( Toll-Like Receptor Adaptor Molecule 1-also known as TRIF ) [10] . We show that polyI:C in the presence of hBD3 has an exacerbated Interferon-β response and decreases CXCL10 production , in vitro and in vivo in both mice and human primary cells . PolyI:C is a synthetic double stranded RNA ( dsRNA ) and consequently a viral ligand mimic , which is recognised by endosomally located TLR3 and also by cytoplasmic RIG-I-like receptors ( RLRs ) [22] . Recently high molecular weight ( HMW ) or long poly:IC has been shown to preferentially access RLRs in conventional dendritic cells [23] . The RLRs include RIG-I ( also known as DDX58 ( DEAD ( Asp-Glu-Ala-Asp ) box polypeptide 58 ) and MDA5 ( Melanoma Differentiation-Associated protein 5 , also known as IFIH1 -interferon induced with helicase C domain 1 ) [24] . HMW polyI:C is recognised primarily by MDA5 ( without the need for transfection ) and secondarily ( with transfection ) by RIGI [23] . Activation of MDA5 and consequent interferon-β production has been shown to be associated with autoimmune disorders [25–28] . Both TLR3 and RLR types of receptor are MYD88 independent , with MDA5 and RIGI requiring the recruitment of the adaptor protein MAVS ( mitochondrial antiviral signalling protein -also known as VISA , IPS-1 and CARDIF ) . Signalling through the MAVS pathway in macrophages results in IRF3/7 driven expression of type I Interferon and NF-ᴋB induction of inflammatory cytokines and chemokines [29 , 30] . TLR3 specifically associates with the adaptor TICAM1 and mediates signal transduction which activates IRF3 and NF-ᴋB [30] . Our studies have been carried out using HMW polyI:C throughout . We dissect the effect on the pathways responsible for the altered response of macrophages to HMW polyI:C in the presence of hBD3 and reveal the mechanism that enables increased Interferon-β and decreased CXCL10 production .
We previously reported that hBD3 suppresses Toll Like Receptor 4 ( TLR4 ) -induced signalling in response to LPS both through MYD88 and TICAM1 pathways [10] . To investigate the effect of hDB3 on other TLR pathways we exposed primary bone marrow derived macrophages from mice ( BMDM ) to a variety of TLR ligands and found that the response to TLR 2 , 2/6 , 1/2 , 7 or 9 ligands were not significantly altered in the presence of hBD3 ( Fig 1A ) . In agreement with our previous findings , hBD3 inhibited LPS-induced TNFα . In contrast , the response to HMW polyI:C was significantly exacerbated in the presence of hBD3 , increasing TNF-α; Interferon-β and IL-6 production , however hBD3 significantly suppressed the CXCL10 ( also known as IP10 ) response to polyI:C ( Fig 1B ) . The enhancing effect of hBD3 on polyI:C cytokine induction , was also seen in the human monocytic cell line THP1 , where hBD3 significantly enhanced polyI:C-induced TNFα , IL-6 and IL-8 . However , in contrast to what we observe in mouse cells , there was no significant effect of hBD3 on polyI:C-induced CXCL10 in human cells ( Fig 1C ) . In addition this enhancing effect was also seen on Interferon-β gene expression in primary human peripheral blood monocyte derived macrophages ( PBMDM ) , measured by qRT-PCR ( Fig 1D ) . The amount of hBD3 required to induce an enhanced Interferon-β response to polyI:C differed from that required to decreased CXCL10 response ( Fig 2A ) . PolyI:C-induced CXCL10 was inhibited by 4μg/ml hBD3 , but no longer inhibited at 2μg/ml hBD3 , whereas concentrations as low as 0 . 05 μg/ml hBD3 enhanced the Interferon-β response to polyI:C , although at this concentration the effect is beginning to diminish . We tested the importance of the hBD3 cysteine-stabilised structure using hBD3 with the canonical defensin motif of six cysteines ( which form three intramolecular disulfide bonds ) replaced with serines . This modified peptide ( hBD3 Cys:Ser ) did not augment cytokine production in response to polyI:C , suggesting that the enhancing effect of hBD3 on polyI:C signalling is dependent on the 3-dimensional structure of the hBD3 peptide ( Fig 2B ) . This lack of effect was not due to the inability of the linear peptide to rapidly enter the cell as TAMRA ( tetramethylrhodamine azide ) -labelled hBD3 Cys:Ser peptide entered BMDM in 10 min ( Fig 2B ) similar to canonically folded hBD3 , as shown in our previous studies[10] . To test whether these in vitro effects were relevant in vivo , we injected wild type mice with polyI:C in the presence of synthetic hBD3 peptide ( Fig 3A ) . We found that TNF-α and IL12p40 responses to polyI:C were significantly increased in the presence of synthetic hBD3 . Interestingly the cytokine CXCL10 was not decreased in the presence of hBD3 ( Fig 3A ) although there was a trend for CXCL10 to be lower in the presence of hBD3- in keeping with the significant reduction of this cytokine seen in the BMDM stimulated with polyI:C and hBD3 . There is a concern that unless synthetic hBD3 is correctly oxidised and the correct disulphide bonding achieved , the properties of the peptide may be altered [7] . The importance of structure is confirmed in our experiment with hBD3 Cys:Ser shown in Fig 2B . The oxidised synthetic peptide we used here ( obtained from Peptide Institute , Japan ) gives details of preparation and oxidation [31] and implies correct cysteine bonding ( C1-C5; C2-4 and C3-C6 ) . However , in order to investigate our effect with hBD3 oxidised in vivo , we expressed the gene ( DEFB103 ) that encodes hBD3 , as a transgene in mice . Transgenic mice were made using the pCAAG promoter ( as described previously by Candille et al . [14] ) by introducing a transgene containing the genomic copy of DEFB103 into ES cells . The vector ( shown in S1A Fig ) expresses hBD3 only after CRE recombinase-mediated deletion of the floxed DsRed , puromycin STOP spacer . After deletion , hBD3 and EGFP are expressed as a polycistronic mRNA and translated as independent proteins using an IRES site . We made several ES cell lines with the DEFB103 expression construct and isolated a clone with strong expression by virtue of DsRed expression and a single site of insertion of the transgene ( located to the sub-telomeric region of chromosome 12 by FISH and chromosome painting ( S1B Fig ) . The control DsRed transgenic mice were made with this ES clone . HBD3 transgenic mice were made from the same ES clone after treatment with CRE recombinase . These cells had strong expression of EGFP and hBD3 and only the ES cells containing the CRE-excised vector showed expression of DEFB103 mRNA ( S1C Fig ) . HBD3 mRNA expression was detected in all tissues tested from mice made using the hBD3-Tg ES cells and hBD3 protein was detected in various tissues including BMDM by immunohistochemistry ( S1D Fig ) using an hBD3 monoclonal antibody [32] . No hBD3 was detected in DsRed-Tg control mice . The effect of polyI:C exposure in mice expressing physiologically secreted hBD3 was tested by exposing heterozygote hBD3-Tg and DsRed-Tg mice to polyI:C ( 100μg , i . p . ) . The transgenic animals expressing hBD3 demonstrated an increased level of both Interferon-β and TNF-α ( Fig 3B ) . In addition , homozygous transgenic hBD3 mice had a significantly raised basal level of Interferon-β compared to controls ( Fig 3C ) . A number of additional cytokines were investigated , including IL-6 and CXCL10 . hBD3-Tg mice treated with polyI:C showed a trend towards enhanced IL-6 induction and reduced CXCL10 induction compared to controls , however these data did not reach significance ( see S2 Fig ) . HMW polyI:C enters macrophages without a transfection reagent , however complexing with the cationic lipid , Lipofectamine 2000 , enhances the amount of polyI:C entering cells by endocytosis , allowing more ligand to be available to both endosomal and cytoplasmic receptors [33] . We hypothesised that hBD3 , being positively charged , may form complexes with polyI:C , enhancing uptake in a similar way to lipofection . To investigate this , we compared cellular uptake of FITC-labelled polyI:C in the presence of Lipofectamine 2000 and/or hBD3 , by flow cytometry . Compared to polyI:C alone , the addition of hBD3 significantly increased the amount of labelled ligand per cell ( Fig 4A ) . Lipofectamine alone also significantly augmented the amount of polyI:C entering cells , with the amount of FITC-polyI:C uptake in the presence of Lipofectamine not differing significantly from the level of FITC-polyI:C in hBD3 treated cells . Uptake in the presence of both lipofectamine and hBD3 was similar to either hBD3 or lipofectamine alone , so no additive effect was apparent . We further compared the effects of lipofectamine and hBD3 on the production of Interferon-β and CXCL10 by polyI:C . Transfecting polyI:C with lipofectamine into wildtype BMDM resulted in enhanced CXCL10 production ( Fig 4B ) . In contrast hBD3 inhibited CXCL10 induction by polyI:C . In the presence of a combination of lipofectamine and hBD3 , CXCL10 production was still inhibited compared to polyI:C and lipofectamine , but to a lesser extent than hBD3 alone . In contrast , hBD3 and hBD3/lipofectamine significantly increased Interferon-β production in response to polyI:C in BMDM ( Fig 4C ) . Lipofectamine and polyI:C did not demonstrate an enhanced Interferon-β response and lipofectamine decreased the amount of Interferon-β produced in response to polyI:C . To visualise the uptake of polyI:C by BMDM we used FITC-labelled polyI:C ( green fluorescence ) . Stronger fluorescence , was observed in cells when either hBD3 or lipofectamine was also present , supporting the flow cytometry data in Fig 4A showing both hBD3 and lipofectamine increase the entry of polyI:C into the cells . HBD3 increased the intensity of cyloplasmic polyI:C staining compared to either lipofectamine or polyI:C alone . In the presence of lipofectamine , polyI:C had a more punctate pattern within the cell , consistent with endosomal location . In the presence of hBD3 , polyI:C appeared to have increased cytoplasmic distribution in addition to foci of staining . These results show that lipofectamine and hBD3 , both enhance the amount of polyI:C that enters the cell but the different signalling responses triggered by each , suggests that hBD3 is directing polyI:C towards cellular compartments , that are not those targeted by the polyI:C-lipofectamine complexes . To further investigate the mechanism of the hBD3 effect on polyI:C we carried out fluorescence co-localisation studies using TAMRA-labelled hBD3 , FITC-labelled polyI:C and immunohistochemistry for the early endosomal marker ( EEA1 ) . TAMRA-labelled hBD3 entered the cells by 10mins ( Fig 5Aii ) , whereas FITC-polyI:C was not visible in the cells until 30 minutes ( Fig 5Biii ) . FITC-polyI:C added to the cells without hBD3 ( Fig 5B and 5C ) , demonstrated FITC fluorescence in punctate regions which stained to some extent with the early endosome marker EEA1-1 antibody ( Pearson coefficient , r = 0 . 615; Fig 5Civ ) . HBD3 alone ( Fig 5A and 5D ) also localised to discrete regions however these did not co-localise with EEA1 staining ( Pearson coefficient , r = 0 . 205; Fig 5Div ) . Adding TAMRA-hBD3 and FITC-polyI:C together onto BMDM , again showed more polyI:C entering the cell in the presence of hBD3 ( Fig 5Biii , 5Cii vs 5Eiii ) . However we could see no evidence for co-localisation of hBD3 and polyI:C ( Pearson coefficient , r = 0 . 073; Fig 5Evii ) . PolyI:C appeared more cytoplasmic in the presence of hBD3 than when it entered the cell alone and the co-localisation coefficient with EEA1-1 staining was lower than in the absence of hBD3 . ( Pearson coefficient , r = 0 . 562 compared to r = 0 . 615 Fig 5Civ versus Fig 5Evi ) . In the presence of polyI:C , hBD3 remained localised in the discrete foci similar to those seen with hBD3 treatment alone , and again hBD3 did not co-localise with EEA1 ( Pearson coefficient r = 0 . 05; Fig 5Eviii ) . This implies that hBD3 , alters the localisation of polyI:C allowing less polyI:C to access the early endosome . Although nucleic acids can induce type I Interferon by activation of TLR signalling [34] in the endosome , Interferon-β and IL-6 can also be produced by activation of cytoplasmic receptors . To examine the consequences of the altered localisation of polyI:C by hBD3 and to determine the signalling pathways responsible for production of Interferon-β and CXCL10 , we used cells from knockout mice specific for the two main pathways known to be involved . Firstly , we determined the Interferon-β response of BMDM exposed to polyI:C in the absence of the TLR3 adaptor TICAM1 and found that this was reduced in Ticam1-/- cells compared to wild type BMDM , indicating that the majority of the polyI:C effect on Interferon-β was not TICAM1 dependent . In the presence of hBD3 , the polyI:C-induced Interferon-β response in Ticam1-/- BMDM was still significantly enhanced compared to polyI:C alone ( Fig 6A ) , which suggests that hBD3 is not enhancing signalling through the TLR3/TICAM1 pathway . Conversely , in the absence of MAVS , where Interferon production is only through TLR3/TICAM1 signalling , a smaller amount of Interferon-β was induced by polyI:C . In the presence of hBD3 this Interferon-β induction was significantly inhibited ( Fig 6A ) , suggesting that hBD3 inhibits TLR3/TICAM1 signalling . Treatment of BMDM from Ticam1 ( -/- ) Mavs ( -/- ) double knockouts with polyI:C did not induce Interferon-β indicating that all the Interferon-β response to polyI:C in BMDM is dependent only on these two pathways ( S4A Fig ) . In contrast to Interferon-β , the induction of CXCL10 by polyI:C in wildtype BMDM was significantly reduced in the presence of hBD3 . In Ticam1 ( -/- ) BMDM the response to polyI:C was eliminated ( Fig 6B ) demonstrating that production of this cytokine is controlled primarily by TLR3/TICAM1 activation . Supporting this finding , polyI:C-induced CXCL10 in Mavs ( -/- ) BMDM ( a TLR3-TICAM1 dependent response ) was not significantly different to wildtype cells indicating that MAVS does not influence CXCL10 production and the production of CXCL10 in response to polyI:C in Mavs ( -/- ) cells was still significantly inhibited by hBD3 ( Fig 6B ) . Both RIGI and MDA5 are upstream of MAVS , so to dissect the effects of hBD3 on MAVS signalling we used the specific RIGI ligand , 5’ triphosphate double stranded RNA ( 5’ppp ) and Mda5 ( -/- ) mice [35] . Treatment of wildtype BMDM with 5’ppp resulted in a significant increase in Interferon-β , TNFα and CXCL10 , and as expected with this ligand , these responses were dependent on the presence of MAVS ( Fig 7A ) . The addition of hBD3 to the macrophages shortly after transfection of 5’ppp , resulted in a significant decrease in cytokine and Interferon-β production ( Fig 7A ) , demonstrating that RIGI responses to 5’ppp are inhibited by the presence of hBD3 . In contrast however , macrophages from Mda5 ( -/- ) mice , revealed that Interferon-β induced by polyI:C was reduced compared to wildtype cells , indicating that MDA5 signalling was responsible for the majority of the polyI:C induced Interferon-β production ( Fig 7B ) . This residual response which is likely to be TLR3-TICAM1 signalling , was not amplified in the presence of hBD3 , indicating that MDA5 is required for the hBD3 enhancing effects on polyI:C-induced Interferon-β .
We show here that hBD3 enhances the production of various cytokines in response to polyI:C ( TNF-α , IL6 in mouse cells and TNF-α , IL8 in human cells ) and Interferon-β in both human and mouse cells . We demonstrate that this effect is dependent on the correct , disulfide stabilised structure of hBD3 . Importantly , we show that this response is not specific to our synthetic hBD3 peptide , as transgenic animals expressing hBD3 from a genomic transgene , also demonstrate an increased type I Interferon response when injected with polyI:C . It was important to reproduce the results shown with our synthetic hBD3 peptide in this transgenic system to validate the augmenting effects of hBD3 as it has been demonstrated that synthetic peptides may be incorrectly folded giving misleading results [7] . The synthetic hBD3 peptide we use here , gives equivalent functional results to hBD3 produced in vivo . When we dissect the pathways known to be activated by polyI:C , it is evident that although signalling through the RLR co-adaptor MAVS mediated pathway is up-regulated in the presence of hBD3 , signalling through the endosomally located TLR3/TICAM1 pathway is suppressed . The inhibition of the TICAM1-mediated signalling pathway supports our findings with LPS , where we previously reported that hBD3-mediated inhibition of LPS signalling through TLR3/TICAM1 was lost in Ticam1 KO mice and could be inhibited by hBD3 cDNA in HEK293 cells [10] . In our experiments here , we use HMW ( long ) polyI:C which activates MDA5 with or without transfection and show that in BMDM , MDA5 is predominantly responsible for the Interferon-β response . Interestingly , it has been demonstrated previously that LMW and HMW polyI:C are recognised by different receptors with HMW polyI:C being recognised by MDA5 and LMW polyI:C by RIGI [24] . In addition large RNA structures generated by viral replication are believed to be important in effectively triggering MDA5 [36] . Recently Zou et al reported that forced delivery of HMW polyI:C to the cytoplasm with transfection was not necessary for RLR stimulation in GM-DCs and CD11bhiCD24lo DCs , although LMW polyI:C required transfection to interact with MDA5 [23] . We show here that similarly to the DCs , BMDM also take up HMW polyI:C effectively without transfection and activate MDA5 . HBD3 exacerbates the signalling through MDA5 . These researchers also report release of endosomal Cathepsin D and induction of necrosis by the activation of MDA5 . We see no evidence of cell death ( using an LDH assay , see S5 Fig ) using polyI:C at 10μg/ml , but this is 5-fold less than that used by Zou et al [23] . Cationic lipids such as Lipofectamine are known to cause endosomal localisation [33] . Although hBD3 is a cationic peptide we do not observe similar outcomes when we compare the effects of hBD3 with the actions of lipofectamine on polyI:C stimulation of macrophages . In wild type cells stimulated with polyI:C , hBD3 increased Interferon-β and decreased CXCL10 production . In contrast , polyI:C and lipofectamine increased CXCL10 production and decreased Interferon-β . This effect is likely to be due to the change in localisation of polyI:C as a result of being in the presence of lipofectamine or hBD3 ( see Fig 8 ) . Our immunostaining demonstrates that hBD3 encouraged polyI:C to be more cytoplasmic compared to lipofectamine which causes increased endosomal localisation . Despite the likely electrostatic interaction of polyI:C and the highly charged hBD3 ( +11 ) in the cell , our cellular uptake experiments using fluorescently labelled derivatives revealed that at 30min after addition to the cells hBD3 and polyI:C do not co-localise appreciably . It is possible that initially they may have interacted , allowing polyI:C to access the cell as hBD3 has been described as having cell penetrating properties [37] . In the presence of hBD3 , polyI:C does not localise to the early endosome so presumably the cytoplasmic location of the ligand allows an increase in interaction with MDA5 . It is possible that cationic hBD3 complexed with polyI:C , enables the ligand to rapidly escape the acidic endolysosome , perhaps in a similar way to pH-dependent fusogenic peptides that assist macromolecules to access the cytoplasm [33] . However we see polyI:C localised to the early endosome ( by EEA1 positive immunostaining ) in the presence of hBD3 implying that the structure of the early endosome is not disrupted by the presence of hBD3 . The main consequence of increased MDA5 signalling in response to polyI:C is increased IFN-β . This increase is additive when lipofectamine is also present ( S4B Fig ) . However no increase is observed in the absence of MAVS which implies that the exacerbated response requires the RLR . It may be that lipofectamine complexes the polyI:C to create higher order structures that activate MDA5 more optimally when [36] hBD3 increases its cytoplasmic localisation . MDA5 is important in relation to autoimmunity and mutations that inactivate or reduce expression of MDA5 have been shown to protect individuals from type I diabetes mellitus risk [38 , 39] . In addition , a mutant form of MDA5 in mice that is active without viral infection induces a type I Interferon-dependent autoimmunity with similarities to lupus [25] . However Interferon-β has also been described as a protector against some types of inflammation such as dextran sodium sulphate induced colitis [40–42] and this protection can be observed in mice that express increased Interferon-β in response to dsRNA-producing intestinal bacteria [40–42] . Increased copy number of the cluster of β-defensins on human chromosome 8 is linked to increased incidence of the autoimmune disease psoriasis and effective treatment of psoriasis with UV irradiation is linked to suppression of type I Interferon and Th17 cells [43] . In addition the most effective treatments currently for psoriasis are monoclonal antibodies directed against IL-17 cytokine production or IL-12p40 ( the cytokine subunit common to both IL-12 and IL-23 ) [18 , 44] . It is thus potentially highly significant that we see strong elevation of IL12-p40 subunit in mice injected with both hBD3 and polyI:C . It is also possible that the other defensins on the CNV cluster may also demonstrate this effect and we have shown that hBD2 also heightens the response of mouse BMDM to polyI:C ( S6 Fig ) . It has recently been shown that human pDC produce Interferon-α in response to self or other DNA through TLR9 [9 , 19] . We show here that macrophages increase the Interferon-β response to polyI:C in the presence of hBD3 through MDA5 . Production of type I Interferons is normally the consequence of pattern recognition receptors binding virally produced nucleic acid pathogen associated molecular patterns ( PAMP ) , such as double stranded RNA ( dsRNA ) produced during viral replication . Psoriasis has been reported to be exacerbated by the use of Interferon-α as therapy for Hepatitis C [45] and by Interferon-β therapy for multiple sclerosis [46] . Investigation of the psoriasis transcriptome has identified an increase in RIG-I like receptors ( RLR ) , which also recognise viral PAMP leading to type 1 Interferon production [47] . During a pathogen infection , hBD3 expression increases [32 , 48] , and hBD3 has been shown to demonstrate potent anti-viral action in vitro [49] . Expression in pDC , monocytes and epithelial cells of the non-copy number variable defensin hBD1 has been shown to increase in response to virus exposure , while expression of the murine orthologue of DEFB103 ( Defb14 ) increases in response to polyI:C [50 , 51] . MDA5 is specialised for protecting mice against infection with various RNA viruses including picornaviruses ( including Theiler’s and Mengo viruses and Encephalomyocarditis virus ( EMCV ) ) as well as paramyxovirus and Norovirus [52 , 53] . MDA5 knockout mice are highly susceptible to EMCV [35 , 54] . During infection , rapid killing , detection and innate response are essential; therefore in this regard , high hBD3 copy number and potentiation of PRR may be beneficial . However an undesirable effect of increased copy number of the defensin cluster ( and concomitant increase in expression of defensin peptides ) may be over stimulation of PRRs leading to exuberant production of type I interferons . This double edged sword may provide protection against pathogens in the short term , but in the longer term contribute to the development of psoriasis in individuals with an increased copy number of the β-defensin cluster .
Animal studies were covered by Project License ( PPL 60/4475 ) , granted by the UK Home Office under the Animal Scientific Procedures Act 1986 , and locally approved by the University of Edinburgh Ethical Review Committee . Human venous blood was collected with written patient consent from healthy volunteers according to Lothian Research Ethics Committee approvals ♯08/S1103/38 . Ultra pure Lipopolysaccharide ( LPS ) from E . coli 0111:B4 , Lipoteichoic acid ( LTA ) , Pam3CSK4 , FSL-1 , HKLM , polyI:C ( HMW ) , FlTC-labelled polyI:C ( HMW ) , R848 , CpG and 5’ triphosphate double stranded RNA ( 5’ ppp-dsRNA ) were purchased from InvivoGen ( San Diego , USA ) , M-CSF , ELISA DuoSets and IFNβ antibodies were obtained from R&D Systems , ( Abington , UK ) . Fluorescently labelled secondary antibodies were purchased from Jackson ImmunoResearch Laboratories ( PA , USA ) . hBD3 ( GIINTLQKYYCRVRGGRCAVLSCLPKEEQIGKCSTRGRKCCRRKK ) was from Peptides International , and cys-ser hBD3 and cys-ser-TAMRA hBD3 were from Almac ( Almac Group Ltd , Craigavon , UK ) . The peptide was produced on a CEM Liberty1 microwave peptide synthesizer using standard Fmoc ( fluorenylmethyloxcarbonyl chloride ) chemistry . Amino acids were purchased from AAPPTec and were assembled on H-Rink amide ChemMatrix resin . Fmoc protecting groups were removed using 20% piperidine and 0 . 1 M hydroxybenzotriazole ( HOBt ) in dimethylformamide ( DMF ) . Amino acids were coupled using 5 molar equivalents of diisopropylcarbodiimide ( DIC ) and 10 molar equivalents of HOBt in DMF . The N-terminal labelling of peptides with fluorescent dye was performed on resin-bound peptide using 4 equivalents of 5- ( and-6 ) -carboxytetramethylrhodamine , succinimidyl ester ( 5 ( 6 ) -TAMRA SE purchased from Biotium ) and 6 equivalents of diisopropylethylamine ( DIEA ) in DMF , incubating for 2 hours . The peptide resin was then rinsed with DMF to remove excess fluorescent dye , washed with dichloromethane ( DCM ) , and dried . Cleavage of the peptide from resin was performed in a trifluoroacetic acid ( TFA ) / triisopropylsilane ( TIS ) / 1 , 2-ethanedithiol ( EDT ) / phenol ( 90:4:4:2 ) mixture for 90 min . The resin was filtered and the filtrate was added to 90 mL of cold dry diethyl ether . The precipitate was collected by centrifugation and the diethyl ether was discarded . The peptide was purified on a C18 reverse phase HPLC column and the correct molecular weight was confirmed by ESI-MS . Oxidative folding was achieved in folding buffer ( 0 . 5–1 . 0 M guanidine hydrochloride ( GuHCl ) , 0 . 1 M Tris , 1 mM glutathione ( GSH ) , 0 . 1 mM oxidized glutathione ( GSSG ) , pH 8 . 5 ) at a peptide concentration of 0 . 1 mg/mL and stirred for 48 hours . Folding was monitored by reverse phase HPLC , which revealed one major species that was used in subsequent experiments . Folding procedures were developed to give the correct HBD3 structure , as verified previously by nuclear magnetic resonance structure determination ( Nix et al . , 2013 ) . The folded products were purified on a C18 reverse phase HPLC column and identified as fully oxidized peptides by ESI-MS . Quantitative concentrations were determined with amino acid analysis at the molecular structure facility at UC Davis . The Mavs−/− ( Cardif , Ips-1 , or Visa ) mutant line was generated by the Tschopp group in Lausanne . It is homozygous-viable null mutant in the C57B6J background . Ticam1 ( Trif ) -/- mice and MDA5-/- mice were used with the generous permission of Professor Shizuo Akira ( Osaka University , Japan ) [35 , 55] . hBD3-Tg and DsRed-Tg were constructed by electroporation of ScaI linearised parental vector , which has a 1 . 5 Kb genomic fragment of the entire hBD3 gene DEFB103 including exons 1 and 2 and the intervening intron cloned into pTLC plasmid ( a kind gift from Josh Brinkman , Danish Stem Cell Centre , DanStem , University of Copenhagen ) using DEFB103 primers with 5' NheI and PacI sites for cloning , into ES cells . Cells that strongly expressed DsRed were selected by FACS and used to make DsRed-Tg mice . Transient cre treatment of these cells produced DsRed negative cells due to lox -mediated excision of the DsRed gene and allowed expression of hBD3 and EGFP ( see S1 Fig ) . The ES cells were made into macrophages using the method of ( Yeung et al 2015 ) which were strong expressors of EGFP . Clones before and after CRE treatment were injected into blastocysts at the University of Edinburgh MRC Evans Building , Transgenic Unit . THP-1 cells were grown in RPMI with 10% fetal bovine serum ( FBS ) and differentiated into macrophages by the addition of 150nM PMA , 2 days before treatment . Mouse primary macrophages ( BMDM ) were generated from femur bone marrow grown for 8 days in DMEM with 10% fetal bovine serum and 20ng/ml M-CSF . Cells , seeded at 2 x 105 into 48 well plates , were grown without M-CSF for 24 hours prior to treatment . Replicate experiments were done with separate primary cell preparations from at least 3 mice for each experiment . Mouse BMDM , seeded at 2 x 105 cells into 48 well plates were treated in serum free media with TLR or RIGI agonists at the concentrations indicated in the figures , in the presence or absence of hBD3 ( 5μg/ml ) . After an 18 hour incubation at 37°C , 5% CO2 , TNF-α , IL-6 , CXCL10 and Interferon-β were measured using mouse DuoSet ELISA ( R&D Systems ) . Statistical significance was determined by an unpaired t-test using GraphPad software , with values expressed as mean +/- SEM and p < 0 . 05 considered significant . Human venous blood was collected from healthy volunteers according to Lothian Research Ethics Committee approval , using sodium citrate anticoagulant ( Phoenix Pharma , Gloucester , UK ) , and cells were separated by Dextran sedimentation , followed by discontinuous , isotonic Percoll gradient centrifugation as previously described [56] . PBMC were incubated at 4×106/mL in IMDM ( PAA Laboratories , Somerset , UK ) at 37°C , 5% CO2 , for 1 h . Non-adherent cells were removed and adherent monocytes cultured for 6 days in IMDM with 10% autologous serum to generate monocyte-derived Mφ . PolyI:C was applied in concentrations indicated and cells harvested at 18 hours . RNA was isolated and human Interferon-β gene expression was measured using the Applied Biosystems Taqman Gene Expression Assay following the manufacturer’s instructions . Cytospins of mouse BMDM were fixed in 4% PFA , washed , then blocked for 2 hours at room temperature ( RT ) in 10% donkey serum ( Sigma , Poole , UK ) in PBST ( 0 . 1% Tween in PBS ) . Slides were then incubated overnight at 4°C with hBD3 antibody ( 1:200 ) ( DSHB , Iowa University , USA ) . After washing , slides were incubated with TxRd labelled anti-mouse antibody ( 1:400 ) for 2 hours at RT , further washed then mounted with Vectashield containing 1μg/ml DAPI . hBD3 immunostaining was visualised using a Zeiss Axioplan 2 microscope ( Carl Zeiss UK Ltd . , Welwyn Garden City , UK ) equipped with Ludl filter wheel ( Ludl Electronic Products Ltd , Hawthorne , NY , USA ) and Chroma 83000 triple bandpass filter set ( Chroma Technology Corp , Rockingham , VT , USA ) . In-house scripts written for IPLab ( Scanalytics Corp , Fairfax VA , USA ) were employed for image capture and image processing . BMDM ( 2 x 104 cells/well ) were cultured overnight on 8 well glass chamber slides ( Nunc Inc , IL , USA ) in DMEM with 10% FCS . Cells were treated with FITC-polyI:C ( 2 . 5μg/ml ) in Optimem media ( Life Technologies ) with or without lipofectamine 2000 ( Invitrogen , at 1:100 dilution , 10μl/ml ) in the presence or absence of hBD3 or TAMRA-labelled HBD3 ( 0 . 5μg/ml ) . After 2 , 10 , 15 or 30 mins cells were washed in PBS and fixed in 4% PFA . For early endosome staining , cells were blocked with 10% donkey serum and incubated with anti-EEA11 antibody ( Abcam , UK ) for 1 hr at RT , then 30 min with donkey anti-rabbit Cy5 . Cells were imaged using a 40x 1 . 3NA oil immersion objective on a Nikon A1R confocal microscope using Nikon Nis-Elements AR software for image acquisition ( Nikon Instruments Europe , Netherlands ) . Image analysis was carried out in ImageJ ( http://imagej . nih . gov/ij/ ) . Pearsons coefficients were calculated using the JaCoP ImageJ plugin [57] . Male C57 Black/6 mice ( 6–8 weeks old ) and hBD3 transgenic male mice ( 8 weeks ) were injected intraperinoneally ( i . p . ) with polyI:C ( 100μg/mouse ) in 200μl of physiological saline . Half of the C57 Black/6 mice also received an i . p injection of synthetic hBD3 ( 20μg/mouse ) . After 4 hr , mice were killed by cervical dislocation , exsanguinated and serum levels of TNFα and Interferon-β measured by ELISA . Mouse BMDM plated at 1 x 106 cells on a 6-well plate were treated with FITC-pI:C ( 10μg/ml ) in the presence of lipofectamine 2000 ( at 1:100 dilution , 10μl/ml ) or hBD3 ( 5μg/ml ) . For treatment with polyI:C in the presence of lipofectamine 2000 , media was replaced with optimem before the addition of polyI:C complexed with lipofectamine 2000 ( L-pI:C ) and addition of hBD3 was delayed for 5 min to avoid direct interaction with L-pI:C complexes . After 18 hours cells were gently washed and gently removed from the dish into PBS containing 1% BSA . Fluorescence was measured with a BD FACSARIAII SORP ( BD Biosciences , Oxford , UK ) , using a 640nm laser ( 670/14nm bandpass filter ) . Data analysis was done using FlowJo Version 7 . 5 . 5 ( Treestar Inc , Olten , Switzerland ) . This experiment was carried out on 3 different preparations of BMDMs
|
Defensins are classically known as antimicrobial peptides due to their ability to rapidly kill pathogens including bacteria , viruses and fungi . They are produced in the presence of infectious agents at body surfaces exposed to the environment . Increasingly , their functional repertoire is expanding , and they have been shown to modulate the immune system . In humans , there is a block of six β-defensin genes that varies in copy number in the population . Individuals with an increased number of β-defensin genes have an increased likelihood of developing the skin autoimmune disease psoriasis . It is not known how this increase in gene copy number influences development of the disease , and psoriasis is a complex interplay of genomic and environmental factors that trigger disease progression and include exposure to viruses . We examined whether a molecular pattern characteristic of viruses produces an altered immune response in the presence of the defensin human β-defensin 3 ( hBD3 ) . We find that hBD3 triggers a larger interferon defence response to this viral mimic by increasing accessibility to a cellular receptor that recognises viral patterns . Interferon is known to be important in autoimmunity and our work may explain why individuals with increased β-defensin number are predisposed to develop psoriasis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Human β-D-3 Exacerbates MDA5 but Suppresses TLR3 Responses to the Viral Molecular Pattern Mimic Polyinosinic:Polycytidylic Acid
|
Vector-borne diseases remain a threat to public health , especially in tropical countries . The incompatible insect technique has been explored as a potential control strategy for several important insect vectors . However , this strategy has not been tested in Culex pipiens pallens , the most prevalent mosquito species in China . Previous works used introgression to generate new strains that matched the genetic backgrounds of target populations while harboring a new Wolbachia endosymbiont , resulting in mating competitiveness and cytoplasmic incompatibility . The generation of these incompatible insects is often time-consuming , and the long-term stability of the newly created insect-Wolbachia symbiosis is uncertain . Considering the wide distribution of Cx . pipiens pallens and hence possible isolation of different populations , we sought to test for incompatibilities between natural populations and the possibility of exploiting these incompatibilities as a control strategy . Three field populations were collected from three geographic locations in eastern China . Reciprocal cross results showed that bi-directional patterns of incompatibility existed between some populations . Mating competition experiments indicated that incompatible males could compete with cognate males in mating with females , leading to reduced overall fecundity . F1 offspring from incompatible crosses maintained their maternal crossing types . All three populations tested positive for Wolbachia . Removal of Wolbachia by tetracycline rendered matings between these populations fully compatible . Our findings indicate that naturally occurring patterns of cytoplasmic incompatibility between Cx . pipiens pallens populations can be the basis of a control strategy for this important vector species . The observed incompatibilities are caused by Wolbachia . More tests including field trials are warranted to evaluate the feasibility of this strategy as a supplement to other control measures .
Vector-borne diseases especially those transmitted by mosquitoes such as malaria , dengue fever , Japanese encephalitis , West Nile fever , Chikungunya and lymphatic filariasis are still important scourges responsible for millions of deaths each year , especially in tropical and developing countries . The lack of effective vaccines for major mosquito-borne diseases and the development of resistance to available chemotherapies both underscore the importance of reducing the numbers of major vector mosquitoes . Insecticides are the major weapons for mosquito control . Earlier efforts using insecticides to reduce malaria and other neglected tropical diseases had been met with success [1] , [2] . But , long-term intensive use of insecticides has led to the development of insecticide resistance in important vector mosquito species , jeopardizing the effectiveness of insecticide-based vector control [3] . In addition , the adverse effects of insecticides on human health and the environment could not be ignored [4] , [5] . With these concerns , biological approaches are called upon as alternatives to chemical control . One approach is to render the arthropods incapable of disease transmission . This strategy uses transgenesis or paratransgenesis to introduce foreign genes into arthropod populations . The expression of the transgenes makes these arthropods resistant to infections or unsuitable for parasite development to their infective stages [6] , [7] . Introduction of endosymbionts such as Wolbachia , maternally inherited obligatory intracellular bacteria , into certain mosquito species can also decrease their vectorial capacity and have yielded novel strategies for population replacement [8]–[11] . Another approach resembles chemical control in that it aims at population reduction . For example , transgenes or biological agents such as Wolbachia can be introduced into mosquito populations to reduce fertility [12]–[15] . Because females of many insect species mate only once in their life-time , unproductive mating effectively precludes their reproduction or reduces their fecundity . Two strategies have been developed to take advantage of their monogamous behaviors , namely the sterile insect technique ( SIT ) and the incompatible insect technique ( IIT ) . SIT uses radiologically or chemically sterilized males to suppress target populations . IIT uses incompatibilities especially cytoplasmic incompatibility ( CI ) between males and females to reduce female fecundity . CI is an incompatibility between a sperm and an egg that interferes with the normal development of a zygote . It is caused by endosymbionts in the cytoplasm such as Wolbachia [15]–[17] . Since their initial identification in the ovaries of Culex mosquitoes [18] , Wolbachia have been found in the reproductive tissues of the majority of tested arthropod species [19] . Wolbachia cause a variety of reproductive abnormalities besides CI , such as feminization of genetic males , thelytokous parthenogenesis ( female offspring being produced from unfertilized eggs ) , and male-killing [15] , [20] . Among these abnormalities , CI has been explored as a potential tool for the development of novel and environmentally friendly strategies for insect controls [9] , [21] . Because CI exists between sperm of Wolbachia-infected mosquitoes and eggs of an uninfected or an incompatible strain of Wolbachia-infected mosquitoes , CI-based IIT is applicable to both Wolbachia-negative and Wolbachia-positive target populations . CI prevents embryonic development , thus making the mating events unproductive . Consequently , those mated females are unable to reproduce or have their fecundity greatly reduced . The application of CI to suppress insect populations predated the identification of Wolbachia as its causative agent . Laven used an introgressed Culex pipiens fatigans strain with maternal cytoplasm derived from a Paris strain to successfully suppress the incompatible local Cx . pipiens fatigans populations in Myanmar [22] . Subsequently , CI-based suppression was trialed for the European cherry fruit fly Rhagoletis cerasi [23] and almond warehouse moth Cadra ( Ephestia ) cautella [24] . To improve the safety of this suppression strategy , an irradiation step was included in several studies on Culex mosquitoes to minimize inadvertent release of fertile females and minority-type ( compatible ) males [25]–[27] . Medfly Ceratitis capitata transinfected with Wolbachia derived from R . cerasi have been generated for IIT approaches to suppress medfly populations [28] . CI has also been utilized to develop strategies to control Aedes mosquitoes . Through interspecific hybridization and introgression , an Aedes polynesiensis strain harboring Wolbachia derived from Aedes riversi was generated . This new strain is bi-directionally incompatible with the natural strain [29] . Wolbachia-infected mosquito strains that display novel patterns of CI for population suppression are usually created by introgression or transinfection [30]–[32] . The preparation of incompatible mosquitoes using such methods is often laborious and time-consuming . Because introgression involves putting Wolbachia into new host genetic backgrounds , the long-term stability of such novel mosquito-Wolbachia symbioses is unknown . In addition , although IIT has been applied to a variety of insects , it has not been tested in Culex pipiens pallens , one of the most prevalent mosquito species in China . In this study , using three Cx . pipiens pallens field populations collected in eastern China , we tested the applicability of exploiting incompatibilities between naturally occurring populations to suppress target populations . We report that incompatible males could compete with cognate males in mating with females , resulting in reduced fecundity . Though the incompatibilities were not absolute , they were preserved with hybrid offspring keeping the maternal crossing types . Our data indicate that this strategy may be a viable and sustainable approach over time to suppress populations of Cx . pipiens pallens .
WX ( Wuxi , Jiangsu Province , 31°33′58 . 47″N , 120°18′9 . 88″E ) , NJ ( Nanjing , Jiangsu Province , 32°3′30 . 11″N , 118°47′47 . 28″E ) , and TK ( Tangkou , Shandong Province , 34°52′34 . 97″N , 117°22′53 . 69″E ) populations of Cx . pipiens pallens were used in this study ( Figure 1 ) [33] . WX and TK populations were collected in 2008 from July to August . NJ population was collected in July , 2006 . For each population , several hundreds of larvae were collected with 350-ml dippers from over 50 larval habitats in public areas of the respective location . Fourth-instar larvae were identified to species by morphology , including the aspect ratio of the air tube and the number of pectens and tufts on it . The larvae were then brought back and reared in the insectary . Mosquitoes were kept at 28°C and 75% relative humidity with 14 h∶10 h light∶dark cycle . Adult mosquitoes were fed 10% ( w/v ) glucose solution prior to blood meals . To separate virgin females and males , pupae from each population were put into 15-ml tubes with water for individual emergence . Afterward , male and female adults were raised in 30 . 5×30 . 5×30 . 5 cm cages . Females 2 days after emergence and males 3 days after emergence were used in mating experiments . Each set of crossings included combination groups of virgin males and females from two different Cx . pipiens pallens populations , with combinations of males and females from the same populations as positive controls . Females and males placed in the same cages were given 2 days to mate . Females were blood fed after mating , then the egg rafts were given 48 hours after oviposition to hatch in separate containers . The numbers of eggs and larvae were counted under stereoscope and the hatching rates ( HR ) for individual rafts were calculated . Unhatched egg rafts were all examined for fertilization through microscopic observation of embryonic development [34] . Each experiment was performed twice , shown here is a representative result . Total DNA of individual Cx . pipiens pallens mosquito was extracted using a method previously described [35] . Polymerase chain reaction ( PCR ) was first carried out using generic primers to amplify wsp gene which encodes the major surface protein of Wolbachia to determine the infection status of the mosquito [36] . Studies have reported that the majority of insect Wolbachia belong to supergroups A and B , specific wsp primers were used accordingly for typing [37] . All primers were adopted from published studies . Primers wF and wR ( see Table 1 for primer sequences ) were used to amplify a 590–632 bp fragment from all Wolbachia strains; primers wAF and wR were used to amplify a 556 bp fragment characteristic of supergroup A; primers wBpipF and wR were used to amplify a 501 bp fragment characteristic of wPip strains of supergroup B; and primers wBcauBF and wR were used to amplify a 466 bp fragment characteristic of the wCauB strains of supergroup B . In addition , PCR was carried out to amplify the ankyrin domain-encoding gene ank2 to differentiate wPip strains into groups I-V based on 313–511 bp amplification fragments [38] . All PCR reactions were composed as follows: 1 unit of Pyrobest Taq DNA polymerase ( Takara , Japan ) , 5 µl 10× Pyrobest Buffer II ( Mg2+ Plus ) , 4 µl of 2 . 5 mM dNTPs , 5 ng total DNA , 2 µl of each 10 µM primer , and ddH2O was added to bring up the total volume to 50 µl . The amplification of wsp gene was performed as follows: 32 cycles of 94°C for 30 s , 55°C for 30 s and 72°C for 60 s , followed by a final extension at 72°C for 7 min . The amplification of ank2 gene was performed as follows: 35 cycles of 94°C for 30 s , 52°C for 30 s and 72°C for 60 s , followed by a final extension at 72°C for 7 min . PCR products were resolved by 1% agarose gel electrophoresis and stained with ethidium bromide . PCR products of ten mosquitoes from each population were used for sequencing of wsp and ank2 alleles by chain-termination method ( Invitrogen ) . Sequence analysis was carried out using DNAMAN software . The unique DNA sequences were deposited into GenBank ( accession numbers JX050182 - JX050187 ) . Tetracycline treatment to eliminate Wolbachia from Culex populations was carried out according to published methods [17] . Tetracycline ( Amresco ) at a concentration of 0 . 05 mg/ml was used for the treatment through both larval and pupal stages . Eggs were placed on tetracycline water solution to hatch . Surviving larvae were transferred to fresh tetracycline solution every 24 hours . A normal infusion was prepared in parallel and fed to larvae in tetracycline solution . The elimination of Wolbachia was checked by Hoechst 33342 ( Sigma ) staining [39] . Egg rafts within 1 hour of oviposition were placed in 5 ml fixation buffer ( 182 mM KCl , 46 mM NaCl , 3 mM CaCl2 , 10 mM Tris pH 7 . 2 , 3 . 7% formaldehyde ) and overlaid with 5 ml n-heptane in a 50-ml tube . After incubation at room temperature for 15 min with constant shaking , fixation buffer was replaced with 10 ml methanol . The eggs were incubated at room temperature for 10 min with constant shaking . The n-heptane layer was replaced with 10 ml methanol . The eggs were washed for 2 times with methanol , and stained with Hoechst 33342 ( 1 µg/ml in PBS ) for 15 min before visualization under microscope . Statistical differences in hatching rate among crossing groups were examined using the Student's t-test . Linear regression analysis was conducted to determine the correlation coefficient between hatching rate and percentage of incompatible males . All statistical analyses were carried out using SPSS Statistics 17 . 0 .
We first tested the compatibilities between available Cx . pipiens pallens populations collected from three geographic locations in China . Reciprocal crosses among these three field populations were performed as outlined in Table S1 . In the mating combinations of NJ and WX populations , no significant incompatibility was detected in either NJ♀×WX♂ or WX♀×NJ♂ cross as compared to NJ♀×NJ♂ and WX♀×WX♂ control groups ( Table S1 and Figure 2A ) . In the mating combinations of TK and WX populations , bi-directional incompatibility was observed , with both TK♀×WX♂ and WX♀×TK♂ crosses having significantly reduced average hatching rate as compared to TK♀×TK♂ and WX♀×WX♂ control groups ( Table S1 and Figure 2B ) . In the WX♀×TK♂ group , 14 larvae hatched out of 21 egg rafts , making the average hatching rate 0 . 005±0 . 026 . The incompatibility was more pronounced in the TK♀×WX♂ group , with 2 larvae hatched out of a total of 18 egg rafts , resulting in an average hatching rate of 0 . 001±0 . 001 . In comparison , the hatching rates in the control groups TK♀×TK♂ and WX♀×WX♂ were 0 . 873±0 . 028 and 0 . 856±0 . 003 , respectively . Similarly , bi-directional incompatibility was detected in the mating combinations of NJ and TK populations ( Table S1 and Figure 2C ) . The average hatching rates for TK♀×NJ♂ and NJ♀×TK♂ crosses were 0 . 004±0 . 003 and 0 . 004±0 . 002 , respectively . In comparison , the average hatching rates were 0 . 883±0 . 028 and 0 . 922±0 . 015 for TK♀×TK♂ and NJ♀×NJ♂ control groups , respectively . The observed bi-directional incompatibility between TK and WX populations and between TK and NJ populations could be a result of mating failure or some post-mating events . To distinguish between these two possibilities , insemination status and embryonic development in incompatible crosses were examined . The female genitalia of Culex mosquitoes include three oval-shaped spermathecal capsules for spermatozoa storage which are connected to vagina through spermathecal ducts . During coitus , some or all of these capsules are filled with seminal fluid from a male and become inflated . The females from incompatible crosses were examined and compared to virgin females . Because three spermathecal capsules from the same individual female had variable sizes before insemination as seen in virgin females , and it was likely that they also varied from mosquito to mosquito , inflation was not always a reliable indicator of successful insemination . Instead , the presence of spermatozoa inside the spermathecal capsules or ovarioles was checked . In all crossing experiments we conducted , the females from both TK♀×NJ♂ and TK♀×WX♂ crosses contained spermatozoa , indicating that they had mated . In addition , the egg rafts were examined 48 hours after oviposition . The eggs from incompatible crosses showed clear embryonic development ( Figure S1 ) . Inside the eggs from TK♀×WX♂ cross , there was segment formation along the axis . At the anterior end of the eggs , two red spots representing primitive eyes ( stemmata ) were visible . Consistent with previous reports , these embryos showed some abnormalities [34] . One evident difference between TK♀×WX♂ cross and compatible crosses was that the incompatible embryos had disorientated bristles . In the eggs from TK♀×NJ♂ cross , similar development was observed . 1–3 ( mostly 2 ) pigmented stemmata were formed . Most stemmata were located in the head region , but some did not seem to have a specific localization . In some eggs , two stemmata were aligned anteroposteriorly instead of being side-by-side . In addition , there was segment formation along the axis in some eggs . Disoriented bristles were also observed in the eggs from TK♀×NJ♂ cross . These observations further confirmed that the females in the incompatible crosses had mated . For those females needed for subsequent mating experiments , only embryonic development inside the eggs was checked to confirm their insemination status . Female monogamy in insects is common but often not absolute . Considering these populations were geographically isolated , it was possible that mating with incompatible males did not preclude subsequent mating of the inseminated females . To test if mating with incompatible males made the females refractory to re-mating , females were retrieved from the incompatible crosses and tested for their ability to mate with cognate males . In the aforementioned crossing experiment , the combination of TK and WX populations displayed higher level of incompatibility than that of TK and NJ populations , so TK and WX populations were chosen in subsequent experiments . TK♀ from TK♀×WX♂ cross and WX♀ from WX♀×TK♂ cross were separated from WX♂ and TK♂ , respectively . Each group was equally divided into two subgroups , with one subgroup mixed with cognate males and the other kept alone ( Table S2 ) . If subsequent mating could happen , the subgroup mixed with cognate males would become inseminated with both compatible and incompatible spermatozoa and result in higher hatching rates than the subgroup kept separate from cognate males . As shown in Figure 3 and Table S2 , the hatching rates in TK♀ subgroup mixed with TK♂ and in TK♀ subgroup kept alone without males were not significantly different ( t = −1 . 013 , df = 18 , P = 0 . 324 ) . Similarly , the hatching rates in WX♀ subgroup mixed with WX♂ and in WX♀ subgroup kept alone without males were not significantly different ( t = −1 . 0 , df = 6 , P = 0 . 356 ) . These results indicate that both TK♀ and WX♀ became refractory to subsequent mating after they mated with incompatible males . Because these incompatible females and males can successfully mate in the absence of compatible males and produce reduced numbers of offspring , we then tested if incompatible mating would still occur when compatible males were available . We also tested if the reduction in fecundity depended on the number of incompatible males introduced . To that end , two sets of experiments were performed using TK and WX females ( Table S3 ) . In each set , an equal number of females were placed in different cages together with no male ( blank ) , equal number of compatible males ( positive control ) , equal number of compatible males plus equal number of incompatible males , or equal number of males plus 3× as many incompatible males . In addition , as in the previous experiment , 48 hours after the first oviposition , egg rafts were checked for embryonic development to ascertain that the females were inseminated . Subsequently , females were collected from each group and divided into two subgroups to be mixed with either no male or cognate males ( Table S4 ) . The hatching rates from the second oviposition were compared to determine if mating with mixed male populations made these females refractory to subsequent mating even when only cognate males were available . The results ( Table S3 , Figure 4A ) show that in the positive control group ( TK♀×TK♂ ) the average hatching rate was 0 . 861±0 . 020 . When an equal number of WX males were included in the cage [TK♀× ( TK♂+WX♂ ) ] , the average hatching rate dropped to 0 . 211±0 . 071 . This value was further decreased to 0 . 063±0 . 039 in the group that included 3× as many WX males [TK♀× ( TK♂+3×WX♂ ) ] . In comparison , the average hatching rate of TK♀ mixed with only WX♂ was 0 . 004±0 . 003 and TK♀ kept alone produced no larva . These data indicate TK♀ mated with TK♂ or WX♂ in the presence of mixed male population , i . e . , even in the presence of TK♂ , mating between TK♀ and WX♂ still occurred , which resulted in reduced overall fecundity of the groups . The extent of fecundity reduction correlated with the ratio of WX♂ to TK♂ . Similarly , in parallel experiment with WX♀ ( Table S3 , Figure 4B ) , the average hatching rate of WX♀ was reduced from 0 . 903±0 . 013 in the positive control group ( WX♀×WX♂ ) to 0 . 607±0 . 066 in the group with an equal number of competing TK♂ included [WX♀× ( WX♂+TK♂ ) ] and further to 0 . 407±0 . 077 in the group with 3× competing TK♂ included [WX♀× ( WX♂+3×TK♂ ) ] . The average hatching rate of WX♀ mixed with only TK♂ ( WX♀×TK♂ ) was 0 . 001±0 . 001 . WX♀ kept alone produced no larva . These data also indicate that the frequency of WX♀×TK♂ mating in the presence of mixed male population can also be increased with an increasing TK♂∶WX♂ ratio . As shown in Figure 4C , after retrieved TK♀ were mixed with TK♂ or no male , these two subgroups produced similar hatching rates in their second oviposition . Similarly , subsequent mixing of retrieved WX♀ with WX♂ did not significantly increase the hatching rates in the second oviposition compared to the subgroups of retrieved WX♀ kept alone ( Figure 4D ) . These results indicate both TK♀ and WX♀ became refractory to subsequent mating after they mated with mixed male populations . This was similar to the scenario in which their first mating was with incompatible males only . Polyandry was not common or completely absent for these female populations . In addition , the average hatching rates in their subsequent ovipositions correlated with the hatching rates in their first oviposition . A higher number of incompatible males during the first mating event resulted in lower fecundity during the first and subsequent gonotrophic cycles . The correlation between hatching rate and the proportion of incompatible males was plotted for both TK♀ and WX♀ . As shown in Figure 5 , based on the curves , the fecundity of TK♀ could be diminished when inundated by excess WX♂ . 9× WX♂ can achieve nearly complete suppression of TK♀ fecundity . On the other hand , the maximal fecundity reduction of WX♀ by TK♂ is around 90% in one generation . The r2 value is 0 . 92 ( P<0 . 05 ) for TK♀ curve ( Figure 5A ) , and 0 . 9363 ( P<0 . 05 ) for WX♀ curve ( Figure 5B ) . Since the incompatibility between TK and WX populations is not absolute , when incompatible males are released to suppress a target population , some hybrids will be generated . If these hybrids are compatible with the released males , then subsequent male release will help these hybrids to reproduce . This would pose a danger of gradually replacing the target population with a compatible hybrid population or creating a balance between the original target population and the hybrid population . In these cases , the populations are not effectively suppressed by male release . To test if the use of naturally incompatible populations as a control strategy is sustainable , the compatibility between the hybrids and TK and WX populations were measured . In the case of releasing WX♂ to suppress TK population , both male and female F1 offspring will be generated from TK♀×WX♂ cross , designated as F1♀ ( TK♀×WX♂ ) and F1♂ ( TK♀×WX♂ ) . These hybrids will encounter TK♀ , TK♂ and WX♂ , resulting in six possible mating combinations . Three extra mating combinations WX♀×F1♂ ( TK♀×WX♂ ) , WX♀×WX♂ and WX♀×TK♂ that would not accompany this population suppression measure were also included in this set to provide more information about the crossing type of F1♂ ( TK♀×WX♂ ) . These nine mating combinations were compared for fecundity . As shown in Table S5 and Figure 6A , the mating combinations F1♀ ( TK♀×WX♂ ) ×F1♂ ( TK♀×WX♂ ) and F1♀ ( TK♀×WX♂ ) ×TK♂ had comparably high hatching rates , while F1♀ ( TK♀×WX♂ ) ×WX♂ produced no larva . These indicate F1♀ ( TK♀×WX♂ ) maintained the crossing type of TK♀ . On the other hand , the average hatching rates of TK♀×F1♂ ( TK♀×WX♂ ) and TK♀×TK♂ were comparably high , while the hatching rates in WX♀×F1♂ ( TK♀×WX♂ ) and WX♀×TK♂ crosses were comparably low , indicating that F1♂ ( TK♀×WX♂ ) maintained the crossing type of TK♂ . These results demonstrate that in TK♀×WX♂ cross , both male and female F1 offspring maintained their maternal crossing type . Reciprocally , the strategy of using TK♂ to suppress WX population was tested for sustainability . To test the crossing type of F1♀ ( WX♀×TK♂ ) and F1♂ ( WX♀×TK♂ ) generated from the cross between WX♀ and TK♂ , nine possible mating combinations were compared for hatching rate . As shown in Table S5 and Figure 6B , the average hatching rates of F1♀ ( WX♀×TK♂ ) ×F1♂ ( WX♀×TK♂ ) and F1♀ ( WX♀×TK♂ ) ×WX♂ were comparably high , while the hatching rate was low for F1♀ ( WX♀×TK♂ ) ×TK♂ group . These indicate F1♀ ( WX♀×TK♂ ) maintained the crossing type of WX♀ . On the other hand , the average hatching rates of WX♀×F1♂ ( WX♀×TK♂ ) and WX♀×WX♂ were comparably high , while the average hatching rates of TK♀×WX♂ and TK♀×F1♂ ( WX♀×TK♂ ) were comparably low . These indicate F1♂ ( WX♀×TK♂ ) maintained the crossing type of WX♂ . In WX♀×TK♂ cross , both male and female F1 offspring maintained their maternal crossing type . Taken together , when using incompatible males to suppress a target Cx . pipiens pallens population , hybrid offspring may be generated if the incompatibility is not absolute . However , these hybrids maintain their maternal crossing types . This phenomenon indicates that both original target population and inadvertently generated hybrids are subject to suppression by the released incompatible males , suggesting that this control strategy is sustainable . The crossing types observed in these Cx . pipiens pallens populations were inherited maternally , suggesting the possibility that Wolbachia was the causal factor . To detect possible infections of Wolbachia , PCR was carried out using total DNA extracted from WX , NJ and TK populations . The primers were selected to amplify the wsp gene [36] , [37] and the ank2 gene [38] according to published studies . For each population , 100% infection rate was detected ( 30 positive out of 30 tested mosquitoes for each population ) . The wsp sequence can distinguish supergroup A , wPip strains of supergroup B and wCauB strains of supergroup B . The ank2 sequence can further distinguish five phylogenetic groups ( wPip-I to wPip-V ) of wPip strains [38] . PCR results show that using the generic wF-wR primer pair a fragment around 600 bp was amplified from all three natural populations of Cx . pipiens pallens ( Figure S2A ) . A fragment around 500 bp was amplified from all three populations using the wPip-specific wBpipF-wR primer pair . Neither wAF-wR primer pair ( specific for supergroup A ) nor wBcauBF-wR primer pair ( specific for wCauB of supergroup B ) generated any amplification product . These results are consistent with previous reports that most mosquito-infecting Wolbachia are wPip strains of supergroup B . This was confirmed by subsequent sequencing analysis , which also revealed that the wsp genes from these three populations are identical ( GenBank accession numbers JX050185 - JX050187 ) . PCR and sequencing analysis of ank2 gene revealed that NJ population was infected by Wolbachia of wPip-III group ( GenBank accession number JX050182 ) , while WX and TK populations were both infected by Wolbachia of wPip-IV group ( GenBank accession numbers JX050183 and JX050184 ) . The ank2 genes from WX and TK populations are identical ( Figure S2B and Figure S3 ) . To confirm that the observed incompatibilities were caused by Wolbachia , the mosquitoes were treated with tetracycline [17] . The elimination of Wolbachia was checked by Hoechst 33342 staining [39] . As shown in Figure S4 , eggs from untreated mosquitoes had strong fluorescence at both anterior and posterior ends , indicating Wolbachia were concentrated at these poles . In contrast , eggs from tetracycline-treated mosquitoes had even distribution of background fluorescence . No strong fluorescence was observed around the micropyle or at the posterior end . These results indicate that Wolbachia were removed from these mosquito populations . In addition , tetracycline-treated strains all tested negative for Wolbachia by PCR using wsp-specific primers wF and wR ( data not shown ) . Crossing experiments were carried out using both Wolbachia-positive and Wolbachia-negative populations . WX females were crossed with Wolbachia-positive TK males ( TK♂ ) , Wolbachia-negative TK males ( TKtet♂ ) and WX males . Similarly , TK females were crossed with Wolbachia-positive WX males ( WX♂ ) , Wolbachia-negative WX males ( WXtet♂ ) and TK males . As shown in Figure 7 , the hatching rate of WX♀×TKtet♂ cross was not significantly different from that of WX♀×WX♂ cross , but was significantly higher than that of WX♀×TK♂ cross . The hatching rate of TK♀×WXtet♂ cross was not significantly different from that of TK♀×TK♂ cross , but was significantly higher than that of TK♀×WX♂ cross . These results indicate that the bi-directional incompatibility between TK and WX populations is dependent on the presence of Wolbachia , i . e . , it is Wolbachia-induced CI . The GenBank accession numbers for sequences mentioned in the paper are ankyrin domain protein ank2 genes of Wolbachia in Nanjing , Tangkou and Wuxi populations of Cx . pipiens pallens ( JX050182 - JX050184 ) , and surface protein precursor wsp genes of Wolbachia in Nanjing , Tangkou and Wuxi populations of Cx . pipiens pallens ( JX050185 - JX050187 ) .
Biological control is a low-pollution component of integrated pest management . A variety of strategies have been developed to suppress populations of insect pests . Taking advantage of many insects' monogamous mating behavior , SIT uses sterile male insects to compete with normal males for their mates , which results in some females producing no or a reduced number of offspring . Two common methods to produce sterile male insects are irradiation and chemical treatments . SIT has been reported to be successful in a number of trials [13] , [40]–[42] . One potential problem with SIT is the dosage of radiation or sterilizing agent used to treat male insects . Insufficient dosages might result in the release of fertile males while overtreatment might damage the males so much that they cannot compete with normal male insects . In addition , the residual chemicals in treated insects could potentially cause pollution to the environment and harm untargeted insect species [43] , [44] . Alternatively , IIT exploits incompatibilities between insect populations and uses incompatible male insects to compete with compatible insects for their mates . Because IIT relies on the genetic traits of incompatible populations , it could be more reproducible than SIT . Most IITs are based on Wolbachia-induced cytoplasmic incompatibility . Infection with Wolbachia has been reported in many species , including some disease vectors such as Aedes albopictus and Culex quinquefasciatus ( = Cx . pipiens fatigans ) [15] , [21] , [30] . For naturally uninfected insects , artificial Wolbachia infection would generate an incompatible strain that could be used to suppress target populations [28] . Introgression using Wolbachia-infected females and Wolbachia-negative target population males can also be used to generate an incompatible strain . For Wolbachia-positive target populations , a common approach is introgression with females from another strain infected with an incompatible Wolbachia . Sometimes , antibiotic treatment to eliminate Wolbachia from target strain is needed to facilitate the introgression process [29] . The advantages of this strategy is the generation of insect population with a desired host genetic background ( often to match that of the target population ) to increase the likelihood of survival in the target environment and successful competition with native males . However , both artificial infection and introgression can be technically difficult . Even with successful establishment of an incompatible strain , because of the placement of Wolbachia in a different host genetic background , the long-term stability of this new symbiosis needs to be tested . In our study , we exploited incompatibilities between naturally existent populations from different geographic locations to suppress mosquito populations . This spares the time and effort to create incompatible populations . Another advantage is that the stability of Wolbachia can be assured since this symbiotic relationship has been naturally selected for a long time . IIT involves repeated release of incompatible males that can competitively inseminate females in a target population . Since these males are not meant to have offspring , any potential adverse effect of their genetic makeup on their fitness in the target environment would not be an issue so long as these males can live long enough to mate and are sexually competitive . We tested the mating behavior of three populations collected from different locations . When mixed with males of a different population , females became inseminated , as spermatozoa were present in the spermathecal capsules of these females . TK females mixed with NJ males or WX males laid comparable number of egg rafts , although these eggs had much lower hatching rates . NJ females or WX females also laid comparable numbers of egg rafts when mixed with TK males compared to when they were mixed with males of their own populations . Egg rafts with low hatching rate also showed clear embryonic development . All these data indicate that these males and females from different populations could successfully mate . Geographical and chronic isolation did not appear to create a mating barrier between these populations . IIT works effectively only if the incompatible males are competitive enough . Although these mosquito populations could mate with each other , we also tested if the females had strong preference of their cognate males in the presence of both cognate and incompatible males using a mating competition experiment . Our results indicated significant mating preference was not observed . In Figure 5 , a linear curve would indicate random mating , while a strong preference of cognate males would result in an arch curve with the median value greater than the average of the two extrema . Both curves are basically linear , supporting that these females chose their mates randomly . With an increasing ratio of incompatible to cognate males , the number of offspring decreased . Our results are consistent with previous reports from other groups that assortative mating usually does not occur in natural populations when numerous Wolbachia strains coexist within those populations [45] . These data suggest that although Wolbachia may cause CI in mosquitoes and skew survival in favor of those embryos that harbor the same or compatible Wolbachia , it does not affect the mate selection significantly . Instead , insects including mosquitoes select their mates based on their own genetic traits [46] , [47] . The selection of suitable populations for mosquito control would depend on finding a population whose genome is compatible with that of target population , while also having a Wolbachia infection that induces cytoplasmic incompatibility in that context . Another factor that influences IIT success is the monogamous mating behavior of females . After female mosquitoes are inseminated , they usually become refractory to re-mating . This phenomenon has been attributed to the effects of proteins in the seminal fluid received from male mosquitoes . But female monogamy is not always absolute . Polyandry has been reported in a number of mosquito species , such as Aedes aegypti and Culex tarsalis . The likelihood of female re-mating increases after these females go through gonotrophic cycles [48] , [49] , possibly due to the waning of seminal fluid proteins . One potential obstacle to the use of mosquitoes from different geographic locations to suppress target populations is the incompatibility between ligands in seminal fluid of one population and receptors in females of the other . In this scenario , insemination would not prevent the female mosquitoes from mating with reproductively compatible males . In our study , TK female mosquitoes became refractory to re-mating after being inseminated by WX or NJ male mosquitoes , even after these females had gone through a gonotrophic cycle , indicating seminal fluid proteins of WX or NJ population can act on receptors of TK population to cause post-coital behavioral changes in TK females . These results suggest that the seminal fluid proteins and their receptors are conserved enough between these populations , or their interactions are flexible enough to tolerate certain mutations . Further studies are needed to reveal the interactions between seminal fluid proteins and their receptors of different mosquito populations . Potentially , there are other factors causing female monogamy . If so , these factors seem to be functional across different populations . Population suppression usually requires repeated effort . It is important for IIT to work for many generations . We tested if hybrids of incompatible populations became compatible with their parental populations . Our data indicate the incompatibility between different mosquito populations is hereditarily stable . This suggests that using males from incompatible natural populations to suppress a target population is sustainable . Although in our study , F1 offspring from incompatible crosses maintained the maternal crossing types , it should be noted that the preservation of maternal crossing type is not always the case . For example , F1♀ from the cross of bi-directionally incompatible Cx . quinquefasciatus strains Bei and Pel became fully compatible with Pel males [50] . This cautions us that when choosing incompatible populations , the crossing types of F1 hybrids should be carefully tested as well . Although the climates at the sites where these mosquitoes originated from are different due to their latitudes , etc . , these mosquitoes did not demonstrate significant behavioral differences in the same artificial environment of our insectary . More trials are needed to determine how they would behave and interact in semi-field and field trials . It remains to be tested whether incompatible male mosquitoes released into different natural environments will remain sexually competitive , and reduce the fecundity of local females . Both maternal inheritance of crossing types and elimination of incompatibility by tetracycline treatment indicate that the observed incompatibility is caused by Wolbachia . To test if these mosquito populations are infected with different strains of Wolbachia , specific DNA fragments were PCR amplified from three mosquito populations . We sequenced wsp and ank2 genes , two specific genes commonly used to type Wolbachia . However , sequence analysis shows that wsp gene is identical in these three mosquito populations . The ank2 gene is identical between TK and WX populations , while ank2 gene of NJ population is different from TK or WX population . These Wolbachia are all wPip strains of supergroup B . It is currently unclear how they differ from each other . The definitive answer would require more sequencing or other typing tests which are beyond the scope of the current study . ANK genes ( including ank2 ) were initially proposed to be involved in CI [50] . This hypothesis was challenged by subsequent analysis that failed to find any association of ANK polymorphism and CI [38] . Our data show NJ and WX populations are bi-directionally compatible , yet their ank2 genes are different . On the other hand , TK and WX populations are bi-directionally incompatible , yet their ank2 genes are identical ( or at least the sequenced segments ) . These provide support to the view that homology between these typing markers does not correlate to the level of CI , suggesting molecular markers that are polymorphic and more closely associated with CI factors are yet to be found [38] , [51] . Nuclear contribution to incompatibility has also been reported [50] . Being isolated from each other in the wild , these mosquitoes might have accumulated enough mutations in their genomes and/or even mitochondrial DNA to make some crosses incompatible . These differences in the host could directly cause incompatibility or more likely cause incompatibility by modulating Wolbachia gene expression . Consequently , although the incompatibilities between these populations are dependent on Wolbachia , contribution of host genomes cannot be ruled out at present . Our observation of embryonic development of incompatible eggs is in accordance with previously reported Wolbachia-mediated CI . It has been reported that incompatible eggs from two Wolbachia-positive populations have higher level of development than incompatible eggs from Wolbachia-negative females ( which are usually undistinguishable from unfertilized eggs ) [34] . We observed stemmata , bristles and segmentation in both TK♀×WX♂ and TK♀×NJ♂ crosses . This would be consistent with the multi-factorial mod resc model: even incompatible Wolbachia provides partial rescue in the eggs [15] . Although incompatibility between insect populations ( including CI ) is not fully understood , its applicability is promising . Our study proves that IIT is a feasible control strategy for Cx . pipiens pallens . The use of naturally occurring populations without genetic manipulation will save time and effort , and require less technical knowhow . This is an advantage for many less developed regions that deserves consideration . The wide distribution of mosquitoes in varied environments may be turned against them because it provides rich diversity in incompatibility; so that it is likely to find naturally incompatible and sexually competitive strains for many target populations . The conclusions from our study on Cx . pipiens pallens might offer reference to control measures of other mosquitoes , as well .
|
Population suppression is an important component of mosquito control measures . The incompatible insect technique exploits the monogamous mating behavior of female mosquitoes to decrease the percentage of females inseminated by compatible males and hence reduce overall fecundity . Previous studies used genetically engineered Wolbachia-infected mosquitoes as the sources of incompatible males . The long-term stability of these mosquitoes is unknown . In this study , we examined naturally occurring incompatibilities between different field populations of Culex pipiens pallens , a vector of West Nile Virus and filarial worms widely distributed in China . We found that bi-directional patterns of incompatibility existed between some Cx . pipiens pallens populations . Incompatible males could compete against cognate males in mating with females and suppress reproduction . The level of suppression depended on the relative population sizes of incompatible and cognate males . We also found that these incompatibilities were preserved in the offspring from incompatible crosses , indicating that this control strategy is likely to be sustainable when applied repeatedly to successive mosquito generations . Our data also indicate that these incompatibilities were caused by endosymbiont Wolbachia , which is also the basis for cytoplasmic incompatibility in a variety of mosquito species . Our results may help to design simpler and more time-effective control strategies for a number of vector-borne diseases .
|
[
"Abstract",
"Introduction",
"Materials",
"And",
"Methods",
"Results",
"Discussion"
] |
[
"biology",
"zoology",
"parasitology"
] |
2013
|
Naturally Occurring Incompatibilities between Different Culex pipiens pallens Populations as the Basis of Potential Mosquito Control Measures
|
The embryonic cuticle is necessary for normal seed development and seedling establishment in Arabidopsis . Although mutants with defective embryonic cuticles have been identified , neither the deposition of cuticle material , nor its regulation , has been described during embryogenesis . Here we use electron microscopy , cuticle staining and permeability assays to show that cuticle deposition initiates de novo in patches on globular embryos . By combining these techniques with genetics and gene expression analysis , we show that successful patch coalescence to form a continuous cuticle requires a signalling involving the endosperm-specific subtilisin protease ALE1 and the receptor kinases GSO1 and GSO2 , which are expressed in the developing embryonic epidermis . Transcriptome analysis shows that this pathway regulates stress-related gene expression in seeds . Consistent with these findings we show genetically , and through activity analysis , that the stress-associated MPK6 protein acts downstream of GSO1 and GSO2 in the developing embryo . We propose that a stress-related signalling pathway has been hijacked in some angiosperm seeds through the recruitment of endosperm-specific components . Our work reveals the presence of an inter-compartmental dialogue between the endosperm and embryo that ensures the formation of an intact and functional cuticle around the developing embryo through an “auto-immune” type interaction .
The Arabidopsis seed is a complex structure composed of three genetically distinct compartments , the maternally-derived seed coat , the embryo , and the endosperm . After fertilization the expansion of the endosperm drives the growth of the seed . However , during later developmental stages the endosperm breaks down , leaving space for the growing embryo . By the end of seed development , only a single endosperm cell layer envelops the embryonic tissues ( reviewed in [1] ) . The endosperm is an angiosperm innovation , thought to have arisen through the sexualisation of the central cell of the female gametophyte [2] . The ancestors of angiosperms probably had seeds more similar to those of gymnosperms , in which tissues of the female gametophyte proliferate independently of egg cell fertilization to produce a nutrient rich storage tissue . However , the endosperm plays not only a nutritional role , but also a role in regulating embryo development . For example , the peptide CLAVATA3/EMBRYO SURROUNDING REGION-RELATED8 ( CLE8 ) may act non-cell autonomously to regulate early Arabidopsis embryogenesis [3] . Recently , maternally-expressed peptides present in the central cell pre-fertilization , and subsequently in the early EMBRYO SURROUNDING REGION ( ESR ) , were shown to regulate Arabidopsis suspensor development . Genetic analysis suggests that this regulation could be mediated by a pathway involving the Receptor-Like Cytoplasmic Kinase SHORT SUSPENSOR [4 , 5] , although the receptor involved remains unidentified . In previous work we showed genetically that the ESR-specific subtilisin protease Abnormal LEaf-shape1 ( ALE1 ) acts in the same genetic pathway as two embryonically-expressed receptor kinases , GASSHO1 [ ( GSO1 ) also known as SCHENGEN3 [6]] and GASSHO2 ( GSO2 ) , to control the formation of the embryonic cuticle in developing seeds [7–10] . Our results indicate that a seed specific inter-tissue signalling event is necessary for the formation of a functional embryonic cuticle [7] . The results of genetic studies have led us to speculate that the role of this pathway is to ensure the robust elimination of apoplastic continuity between the developing embryo and the surrounding endosperm thus gating molecular movement between the two compartments [11 , 12] . The cuticle is the outermost layer of the aerial parts of the plant . It is a highly complex structure mainly composed of a lipid polymer ( cutin ) and waxes , either associated with the polymer ( intracuticular waxes ) or deposited on the top of it ( epicuticular waxes ) ( recently reviewed in [13 , 14] ) . Cutin and waxes are composed of complex mixtures of hydroxylated and very long-chain fatty acid derivatives , respectively . Cuticle structure and composition are highly regulated not only at the tissue level , but also in response to environmental stimuli such as drought , radiation and pollution [13 , 14] . In addition , several reports have highlighted the important role played by the cuticle in biotic interactions , and particularly in protecting plants from attack by bacterial pathogens ( reviewed in [13 , 15 , 16] ) . In Arabidopsis although little , if any , evidence exists for the presence of cutin-like substances in the wall between the mature egg cell and the central cell , by the end of embryogenesis the hypocotyl and cotyledons of the embryo are covered with a continuous cuticle which renders the germinating seedling impermeable to hydrophilic dyes , and resistant to water loss [17] . Cuticle biogenesis is considered to be a unique property of epidermal cells [18] . During plant development , epidermal cells are generated by anticlinal divisions of pre-existing epidermal cells so that each cell inherits an intact external cuticularised cell wall . In this respect the embryonic cuticle is atypical as it is deposited de novo at the interface between the developing embryo and endosperm . Although mutants with defective embryonic cuticles have been described [7–10 , 17] , only very fragmentary evidence about when the embryonic cuticle appears is present in the literature . Furthermore , the structure of the embryonic cuticle , its composition , the mechanisms via which it is deposited and its function during seed development remain unexplored . In this study we aimed to elucidate how the embryonic cuticle is formed , and to investigate how the ALE1 GSO1 GSO2 signalling pathway impacts its biosynthesis and deposition .
An inspection of available in silico data [19–21] showed that many genes encoding enzymes thought to be involved in cutin biosynthesis are expressed during early embryogenesis ( S1 Fig ) . In situ hybridisations confirmed that genes known to affect cuticle production ( LACS2 [22 , 23] , FIDDLEHEAD/KCS10 [24–27] , LACERATA [28] and BODYGUARD [29] ) or export ( LTPG1 [30] and ABCG11 [31] ) have a clear epidermis-specific expression from the mid globular stage onwards ( Fig 1 , S3 Fig , S3 Fig ) . In agreement with published and in silico data [9] ( S1 Fig ) GSO1 and GSO2 were expressed in the embryo from early developmental stages ( Fig 1 , S3 Fig ) . In addition , their expression was mainly restricted to the embryonic epidermis . GSO1 expression in the embryonic epidermis was further confirmed using plants expressing a functional genomic GSO1-mVENUS fusion under the control of the GSO1 promoter ( pGSO1:GSO1-mVENUS ) [6] ( Fig 2 ) . This construction fully complemented the cuticle permeability phenotype of gso1-1 gso2-1 double mutant seedlings , and strongly reduced the misshapen-seed phenotype of gso1-1 gso2-1 mutant seeds when introduced into the gso1-1 gso2-1 mutant background ( Fig 2 ) . Since previous results showed that epidermal identity is not affected in gso1-1 gso2-1 mutants [12] , the expression of cuticle biosynthesis genes was analysed by in situ hybridization in gso1-1 gso2-1 double mutant seeds ( which show a stronger cuticle phenotype than ale1-4 mutants [7] ) . As shown in Fig 1 ( and S2 Fig and S3 Fig ) , no reduction in the expression of any of the cuticle biogenesis genes analysed was detected in the embryonic epidermis of this background , whereas reduced expression of both GSO1 and GSO2 was clearly visible . For these results , we concluded that although many genes involved in cuticle biosynthesis are co-expressed with GSO1 and GSO2 in the embryonic epidermis , their expression is not dependent upon GSO1 and GSO2 . The cutin content of seedling cotyledons was assessed by measuring the quantities of the main cutin monomers released after cutin isolation followed by depolymerisation ( mainly C16 and C18 ωOH ( omega-hydroxy acid ) and DCA ( α , ω-dicarboxylic acid ) ) . As clearly illustrated by the quantification of 18:2-DCA , the major component of Arabidopsis cutin , a slight loss in cutin load was detected in gso1-1 gso2-1 , but not in ale1-4 cotyledons compared to wild-type . In contrast a very clear reduction in cutin load was observed in control plants lacking the acyltransferases GPAT4 and GPAT8 required for cutin biosynthesis , as has previously been reported in rosette leaves [32] ( Fig 3A ) . We therefore investigated the cuticle permeability of etiolated cotyledons by submerging them in the hydrophilic dye toluidine blue , which can only penetrate internal tissues through defects in the cuticle [17] . Surprisingly , we found that the cotyledons of etiolated gpat4 gpat8 seedlings showed a rather similar toluidine blue permeability to ale1-4 seedlings and a considerably reduced permeability compared to gso1-1 gso1-2 double mutants , suggesting that the gpat4 gpat8 cuticle , although quantitatively strongly deficient in cutin monomers , remains partially functional ( Fig 3B ) . Taken together with gene expression analysis , these results suggest that the ALE1 , GSO1 and GSO2-mediated signalling pathway might impact cuticle organisation or integrity rather than the quantity of cuticle components produced by epidermal cells . The process of embryonic cuticle deposition was investigated in more detail in wild-type ( Col-0 ) seeds ( Fig 4A–4D ) . At the two-cell stage the embryo was surrounded by a thick cell wall but no electron dense material was detected at the embryo surface . At the mid-late globular stage , a cutin-like electron-dense material was detected in patches ( Fig 4B and S4 Fig ) . From heart stage onwards , an apparently continuous layer of electron-dense cutin-like material was detected at the surface of the outer epidermal cell wall . Embryonic cuticle production therefore involves the de novo deposition and subsequent coalescence of “patches” of cuticular material at the surface of epidermal cells . Toluidine blue assays with wild-type embryos extruded at different developmental stages indicated that embryonic cuticle permeability started to reduce noticeably at the mid torpedo stage ( after apparent gap closure ) , and that the embryo cuticle continues to become more and more impermeable during embryo development ( S5 Fig ) . These results suggest that the coalescence of visible gaps in the embryonic cuticle precedes a measurable reduction in embryonic permeability , and that cuticle reinforcement continues throughout embryogenesis . In gso1-1 gso2-1 mutant embryos the cuticle still showed discontinuities at the heart and walking stick stage ( Fig 4E and 4F , S4 Fig ) . In this background the cuticle also appeared thicker , but less condensed than that of wild-type embryos . The outer epidermal cell wall was also abnormally thick at later stages ( compare embryonic cell wall thickness in Fig 4D with that in 4F ) . Similar discontinuities were observed , although at a lower frequency , in the ale1-4 background at the heart stage as described previously [10] , but were less frequent at later stages , consistent with the less severe cuticle permeability phenotype observed in the seedlings of this background ( Fig 4G and 4H ) . Consistent with these observations ale1 and gso1-1 gso2-1 mutant embryos remained permeable to toluidine blue throughout embryogenesis ( Fig 5 ) , with permeability phenotypes in fully expanded but immature embryos correlating closely with those observed in etiolated seedlings . Importantly ale1 mutants were more permeable to toluidine blue at embryo maturity despite an apparent lack of visible gaps in the ale1 cuticle after the torpedo stage . Interestingly fully expanded embryos of gpat4 gpat8 mutants were not significantly more permeable than those of wild-type plants , suggesting that permeability defects observed in etiolated seedlings in this background could be due to the rapid expansion of embryonic tissues ( especially the hypocotyl ) after germination . These results are consistent with our hypothesis that the ALE1 GSO1 GSO2 pathway is necessary for generating a continuous cuticle layer and further suggest that it controls “gap closure” during embryonic cuticle maturation . Transcriptional analysis of intact siliques from gso1-1 gso2-1 and ale1-4 mutants and wild-type plants was carried out at globular and heart stages . The results are provided in Fig 6A and 6B , S1 Table , S6 Fig and S7 Fig . The number of differentially down-regulated genes in the mutant backgrounds compared to wild-type was higher than the number of up-regulated genes ( S1 Table ) . A moderate overlap between genes showing higher expression in ale1-4 and gso1-1 gso2-1 mutants than wild-type controls was observed ( S6 Fig , S1 Table ) . In contrast more than three quarters of the genes showing reduced expression at both developmental stages in the gso1-1 gso2-1 background also showed reduced expression at both developmental stages in ale1-4 mutants ( Fig 6A and S1 Table ) , corroborating previously published genetic evidence that ALE1 , GSO1 and GSO2 act in the same genetic pathway [7] . Because ALE1 appears to be expressed exclusively in the ESR region of the endosperm [7 , 8 , 10] , genes mis-regulated in both mutant backgrounds likely comprise bona fide targets ( direct and indirect ) of the ALE1 GSO1 GSO2 pathway , despite the fact that the expression of GSO1 and GSO2 is not restricted to the seed [6 , 9 , 20] . Genes up-regulated in both mutant backgrounds showed a moderate over-representation in GO terms associated with responses to abiotic stress ( S6 Fig ) . In contrast , genes down-regulated in both backgrounds , particularly at the heart stage , showed a very striking overrepresentation for GO terms linked to abiotic and biotic stress responses ( Fig 6B , S7 Fig ) . Mis-regulation of 19 of these genes was validated using additional independent biological samples by qRT-PCR ( S8 Fig ) . The expression levels of these genes in seeds were generally low , and attempts to carry out in situ hybridization were inconclusive . However , for one target , SWI3A [33] , expression in the developing embryo predicted from in silico data was confirmed , and shown to be convincingly reduced in embryos of the gso1-1 gso2-1 double mutant ( S9 Fig ) . Thus , consistent with the embryonic expression of GSO1 and GSO2 , some of the transcriptional regulation downstream of ALE1 GSO1 GSO2 signalling occurs in the embryo . Expression of ALE1 was not reduced in gso1-1 gso2-1 mutants ( S1 Table and S8 Fig ) , suggesting that ALE1 is not a downstream target of GSO1 GSO2- mediated signalling , and could therefore act upstream of GSO1 and GSO2 in mediating embryonic responses necessary for the establishment of an intact embryonic cuticle . The GSO1 and GSO2 receptor kinases belong to family XI of the Leucine-Rich Repeat ( LRR ) -RLKs [34 , 35] , and are closely related to the “danger” peptide receptors PEPR1 and PEPR2 [36 , 37] , which are involved in the amplification of defence responses triggered by pathogen-associated molecular pattern ( PAMP ) perception [38] . A previous study [39] , reported aberrantly shaped seeds , resembling those of ale1-4 mutants , in Arabidopsis mpk6 mutants lacking the MITOGEN ACTIVATED PROTEIN KINASE6 ( MPK6 ) protein , which acts downstream of PEPR signalling . In addition a proportion of mpk6 mutant seeds were reported to rupture [39] . We confirmed these phenotypes in the mpk6-2 mutant background ( S10 Fig ) . A recent article has suggested that some seed defects in mpk6 mutants may depend upon the genotype of the maternal tissues in the seed [40] . Reciprocal crosses were therefore performed , and these confirmed that seed twisting phenotype is dependent upon the genotype of the zygotic compartment and not the maternal compartment ( S11 Fig ) . Consistent with previous reports [39 , 40] mpk6-2 mutants produce embryos with highly variable phenotypes , around 30% of which fail to develop a normal hypocotyl region ( “deformed” embryos Fig 7 ) . We found that 40–50% of both deformed and normal mpk6-2 embryos showed abnormal permeability to the hydrophilic dye toluidine blue , consistent with the presence of embryonic cuticle defects ( Fig 7A ) . This phenotype correlated well with toluidine blue permeability phenotypes observed in etiolated seedlings , confirming that defects were not simply a consequence of abnormal development post-germination ( Fig 7B ) . Auramine O staining of the cotyledons of etiolated seedlings was used to confirm mpk6 cuticle defects ( S12 Fig ) . Using this technique , wild-type cotyledons were found to be covered with a continuous cuticle layer . As previously reported , and consistent with our cutin analysis , gpat4 gpat8 mutants showed drastically reduced cuticle staining . In contrast gso1-1 gso2-1 mutants showed a patchy cuticle , similar to that seen using transmission electron microscopy on the embryo surface . Both ale1-4 and mpk6-2 mutants showed a more weakly stained cuticle than wild-type , which although apparently continuous , showed uneven cutin deposition ( S12 Fig ) . Triple mpk6-2 gso1-1 gso2-1 and double ale1-4 mpk6-2 mutants were generated to investigate further the genetic interactions of ALE1 , GSO1 and GSO2 with MPK6 . Fertility in ale1-4 mpk6-2 double mutants was similar to that in mpk6-2 mutants , while triple mpk6-2 gso1-1 gso2-1 mutant plants were viable but produced very few seeds . In terms of seed shape and cotyledon cuticle permeability , triple mpk6-2 gso1-1 gso2-1 mutants had phenotypes identical to those observed in gso1-1 gso2-1 double mutants ( Fig 7 , S10 Fig ) . Since all gso1-1 gso2-1 mutant seeds are twisted , non-additivity cannot be concluded from this phenotype . However , recent work has shown that additivity of toluidine blue staining phenotypes can be detected in mutant combinations with gso1-1 gso2-1 [41] . The frequency of “twisted” seeds ( including ruptured seeds ) , and toluidine blue stained seedling cotyledons was non-additive in ale1-4 mpk6-2 double mutant plants , consistent with ALE1 , GSO1 , GSO2 and MPK6 acting in the same genetic pathway to control seedling cotyledon permeability ( Fig 7 and S10 Fig ) . MPK6 is involved in a plethora of reproductive and non-reproductive developmental processes and shows functional redundancy with other MPK proteins [39 , 42–52] meaning that global transcriptome analysis in the mpk6-2 background would likely be uninformative for this study . We therefore directly tested a subset of genes mis-regulated in gso1-1 gso2-1 and ale1-4 mutants for misregulation in mpk6-2 mutants at three stages of embryo development . Five out of eight genes tested showed reduced expression in mpk6-2 either at all three stages ( SWI3A , WRKY70 and NIMIN1 ) , or in two out of three developmental stages tested ( SIB1 and NIMIN2 ) ( S13 Fig ) . Unsurprisingly given the relatively weak cuticle phenotype of mpk6 mutants compared with gso1 gso2 mutants , some genes showing strong down-regulation in the gso1-1 gso2-1 mutants ( WRKY33 , WRKY46 and WRKY53 ) did not show any significant reduction in expression in the mpk6-2 mutant background ( S13 Fig ) indicating that their transcriptional regulation downstream of GSO1 and GSO2-mediated signalling could be dependent on signalling components acting redundantly with MPK6 . The expression of ALE1 , GSO1 and GSO2 was not altered in mpk6-2 mutants ( S13 Fig ) , indicating that MPK6 most probably acts downstream of GSO1 and GSO2-mediated signalling . To further confirm this hypothesis , we analysed MPK phosphorylation in developing seeds from Col-0 and gso1-1 gso2-1 double mutants . In seedlings , phosphorylation of MPK6 ( and additional MPKs ) can only be detected after elicitation ( for example with the flg22 peptide ) . The response to flg22 is not attenuated in gso1/gso2 mutant seedlings ( S14 Fig and S15 Fig ) . In contrast , MPK6 phosphorylation ( but not phosphorylation of other MPKs ) could be detected in un-elicited seeds ( Fig 7C and S16 Fig ) . Following quantification , we found that the degree of phosphorylation of MPK6 was reduced by approximately 50% in gso1-1 gso2-1 double mutant seeds compared to wild-type , suggesting that a significant proportion of MPK6 phosphorylation in seeds depends on the activity of GSO1 and GSO2 ( Fig 7D , S16 Fig ) . Intriguingly , in seeds , a band corresponding to a second phosphorylated MPK was detected exclusively in mpk6-2 mutants ( Fig 7C ) , suggesting that the relatively weak mpk6 seedling cuticle phenotype could be due to compensation by an as yet unidentified MPK [53] . The strong expression of GSO1 and GSO2 in the embryonic epidermis , suggests that the activity of GSO1 and GSO2 in cuticle formation is required in the embryo . No promoters confirmed as specifically being expressed only in the embryo or embryo epidermis , have been published . To further confirm the spatial requirement for GSO1/GSO2-dependent signalling in the seed , we therefore complemented the mpk6-2 mutant either with the MPK6 cDNA expressed under the ubiquitously expressed RPS5A promoter , or under the endosperm specific RGP3 promoter [54 , 55] . We were unable to complement either the misshapen seed/seed bursting phenotypes or the toluidine blue permeability phenotypes of mpk6-2 mutants by expressing MPK6 in the endosperm , but obtained full complementation of all phenotypes in plants expressing MPK6 under the RPS5A promoter ( Fig 8 , S17 Fig ) . Together with the results of our reciprocal crosses , these findings indicate that the seedling permeability phenotype of mpk6-2 mutants is most likely due to signalling defects in the embryo . Seed size and seed bursting defects could be caused by lack of MPK6 in the testa , as suggested by reciprocal crosses , although this remains to be investigated in more detail . In order to further confirm the function of MPK6 downstream of GSO1/GSO2 signalling we attempted to express a constitutively active form of MPK6 under the RPS5A promoter in wild type and double mutant plants , but were unable to generate any transformants , potentially due to the critical roles played by MPK6 during early embryogenesis .
It this study , consistent with the similarity between GSO1/2 and PEPR1/2 proteins , we found that stress-associated kinase MPK6 , which has been shown to act downstream of PEPR signalling [56] , shows constitutive phosphorylation in developing seeds , and that this phosphorylation is partially dependent upon GSO1 and GSO2 . In addition , we showed that GSO1/GSO2 , are required for the expression of a set of stress-related genes during early seed development . Our results suggest that GSO1/GSO2 dependent stress response-related signalling pathways are active in developing seeds . Because of the conserved transcriptional targets expressed downstream of GSO1/GSO2 dependent signalling , and in defence responses , this scenario is distinct from previously reported situations in which single pathway components , such as the co-receptor BAK1 , play distinct roles in developmental and defence-related signalling cascades through interaction with multiple receptors [57 , 58] . However , the role of the transcriptional targets of GSO1/GSO2 signalling in seeds remains to be elucidated . Our work also shows that GSO1/GSO2 , ALE1 and MPK6 act in a genetic pathway involved in ensuring embryonic cuticle integrity . We show for the first time that embryonic cuticle biogenesis involves the coalescence of discontinuous patches of cutin-like material that appear on the embryo surface at the globular stage , and that pathway mutants are either incapable of completing , or retarded in the completion of “gap closure” during this process . Interestingly , GSO1 ( also known as SCHENGEN3 [6] ) was recently shown to be involved in ensuring the continuity of another apoplastic diffusion barrier , the Casparian strip , which prevents the apoplastic movement of solutes from the cortex to the stele of the root [6] . GSO1 may therefore form part of a general mechanism employed by plants for monitoring the “integrity” of apoplastic barriers formed during plant development . The role of GSO1 and GSO2 in the closure of gaps in the nascent cuticle implies spatial regulation of signalling outputs at the subcellular level . Cytoplasmic signalling components which , like MPK6 might not be uncovered by transcriptome analysis but instead be modified post-translationally , are therefore likely to be of critical importance in GSO1 GSO2 signalling in the embryonic epidermis . Indeed , although MPK6-mediated signalling has most often been implicated in the control of transcription , particularly via the modulation of the activity of WRKY transcription factors , evidence for potential roles in cytoplasmic responses , for example during funicular guidance of pollen tubes [46] and control of cell division planes [50] , exist in the literature . Cytoplasmic responses downstream of receptor-like kinases include the local production of apoplastic Reactive Oxygen Species ( ROS ) and/or calcium influxes , and indeed localized ROS production has been implicated in Casparian strip formation [6 , 59–61] . However although a plausible model has proposed that ROS release could mediate Casparian strip polymerisation though polymerisation of monolignols [59] , it is less obvious how ROS could directly affect the biosynthesis of an aliphatic cutin-based barrier , although a possible role for ROS in linking the cuticle to the cell wall has been evoked [62] . ROS production has been shown to directly modulate the activation of MAPK signalling , providing a mechanism permitting the reinforcement of localised signalling events [63 , 64] . Another , potentially linked , possibility is that GSO1/GSO2 activity in the embryo could spatially direct the secretion of either cuticle components or enzymes and cell wall components necessary for their integration into the cutin polymer , in a system analogous to the rapid and highly localized deposition of callose observed upon hyphal penetration into epidermal cells ( reviewed in [65 , 66] ) . Interestingly MPK6 has also been shown to be involved in phragmoplast formation during root cell division and therefore could be involved in the localised production/secretion of apoplastic compounds [50] . However , observing these processes in situ , within the living seeds , would require developments in microscopy which are not yet available . Our work highlights several questions which merit further discussion . A first important question is whether the GSO1/GSO2 signalling pathway could play a role in protecting seeds , or more generally plants , against pathogen attack . Cuticle integrity in adult plants has been shown to be required for resistance to Pseudomonas pathovars [67 , 68] . The action of the ALE1 GSO1/GSO2 signalling pathway in ensuring embryonic cuticle integrity is therefore likely to have a significant influence on embryo and seedling susceptibility to bacterial pathogens . However , we have also shown that GSO1/GSO2 , ALE1 and MPK6 are necessary for the expression of known defence marker genes in seeds . Cuticle permeability phenotypes have not been reported in the literature for mutants affected in the defence markers identified in our transcriptome studies . This raises the question of whether the ALE1 , GSO1/GSO2 , MPK6 signalling pathway , in addition to mediating localised apoplastic modifications , could act at a more global level either to protect developing seeds from the ingress of bacterial pathogens ( thus affecting vertical pathogen transmission ) , or to “prime” embryos against pathogen attack upon germination . Exploring this possibility would necessitate functionally separating susceptibility caused by cuticle defects from lack of immune priming , and will be technically very challenging , but could ultimately inform strategies aiming to reduce vertical transmission of plant pathogens . The second question concerns how signalling via GSO1 and GSO2 is triggered in the seed in the absence of pathogens . In this study we consolidate data supporting the function of ALE1 in the same pathway as GSO1 and GSO2 . We previously proposed that the function of ALE1 , GSO1 and GSO2 in ensuring the apoplastic separation of the embryo and endosperm became necessary in angiosperms due to developmental constraints imposed by the sexualisation of the female gametophyte , which led to the simultaneous development of the embryo and surrounding nutritive tissues post-fertilization , rather than their sequential development [11 , 12] . ALE1 expression is endosperm specific and , as previously suggested [69] , the recruitment of ALE1 to a function in reinforcing the embryonic cuticle may have occurred during the emergence of the angiosperm lineage . The fact that the phenotype of ale1 mutants is weaker than that of gso1 gso2 double mutants , particularly at later stages of embryogenesis , may be due to redundancy with other members of the subtilase family in Arabidopsis , several of which are encoded by genes expressed in the developing endosperm [19 , 20] . Subtilases have been shown to be involved in defence responses and immune priming in plants [70 , 71] . It is thus possible that ALE1 acts to produce an as yet unidentified ligand for the GSO1 and GSO2 receptors . In such a scenario the function of ALE1 in the seed could be analogous to the “immune priming” function previously reported for the subtilase SBT3 . 3 [71] . Such a scenario naturally raises a third , and important question , around the identity of the ligand of GSO1 and GSO2 . Two sulfated peptides , CIF1 and CIF2 , which can act as ligands for GSO1 during Casparian strip formation , have recently been identified [72 , 73] . Testing the role of these molecules in developing seeds will be an obvious priority . However Nakayama and colleagues specifically reported that no cuticle defects ( as gauged by cotyledon fusion phenotypes ) were observed in cif1 cif2 double mutants and the possibility that other signalling molecules could be involved in ensuring embryonic cuticle integrity therefore cannot be excluded . In summary , we propose that endosperm-localised factors ( like ALE1 ) may have been recruited to hijack a defence-signalling pathway involving the ancestor ( s ) of GSO1 and GSO2 , and downstream signalling components including MPK6 , and trigger an “auto-immune” type response in the embryo to ensure cuticle integrity . The future identification of further pathway components , and in particular the substrates of ALE1 and ligands of GSO1 and GSO2 , will help to confirm this hypothesis .
The pGSO1:GSO1-mVENUS line was kindly donated by Professor Niko Geldner ( Unil-Sorge , University of Lausanne ) . The mpk6-2 ( SALK_073907 ) mutant and the mpk3-1 ( SALK-151594 ) were kindly provided by Dr Roberta Galletti . Unless otherwise specified , plants were grown for 10 d in sterile conditions on Murashige and Skoog ( MS ) agar plates with 0 . 5% sucrose , 1 month under short-day conditions ( 19°C , 8h light / 17°C , 16h dark ) and then transferred to standard long-day conditions ( 21°C , 16h light/8h dark ) for one more month . To stage material , newly opened flowers were marked each day for two weeks . For bacterial growth assays , plants were grown under controlled conditions in a growth chamber at 21°C , with a 9-hours light period and a light intensity of 190 μmol . m-2 . s-1 . For MPK6 activation analysis , seedlings were grown for 10 d in MS liquid medium supplemented with 0 . 5% sucrose in 100μm cell strainers submerged in 6-well plates ( 5ml of medium per well ) . Cell strainers were transferred to new plates containing MS sucrose 0 . 5% supplemented with 100 nM or 1 nM flg22 or water and incubated for 15 and 60 minutes at room temperature without skaking . Seedlings were then rapidly harvested in liquid nitrogen and stored at -80°C until protein extraction . For cutin analysis of seedlings , seeds were sterilized , plated on MS medium supplemented with 0 . 7% agar , 0 . 7% sucrose and 2 . 5 mM MES-KOH , pH 5 . 7 , and stratified in the dark for 3 days at 4°C . Plates were then transferred to a controlled environment growth chamber at 22°C and with continuous light , and seedlings were grown for 5 days before harvesting the cotyledons . For toluidine blue staining and Auramine O staining , sterilized seeds were spread uniformly on 15 cm MS plates with 0 . 5% sucrose and 0 . 4% Phytagel ( Sigma ) ( pH 5 . 8 ) and stratified for 2 days in the dark at 4°C . After stratification seeds were transferred to a growth chamber and incubated for 6h under continuous light followed by 4 days in the dark . DNA templates for the probes used in in situ hybridizations were amplified using the primers listed in S2 Table . Digoxigenin-labelled RNA probes were produced and hybridized to tissue sections following standard procedures . In brief , siliques were opened , fixed overnight in ice-cold PBS containing 4% paraformaldehyde , dehydrated through an ethanol series , embedded in Paraplast Plus ( Mc Cormick Scientific ) and sectioned ( 8 μm ) . Immobilized sections were dewaxed and hydrated , treated with 2x saline sodium citrate ( 20 min ) , digested for 15 min at 37°C with proteinase K ( 20 mg/ml ) in 50mM Tris-HCl , pH 7 . 5 , 5mM EDTA ) , treated for 2 min with 0 . 2% glycine in PBS , rinsed , post-fixed with 4% paraformaldehyde in PBS ( 10 min , 4°C ) , rinsed , treated with 0 . 25% w/v acetic anhydride in 100mM triethanolamine ( pH 8 . 0 with HCl ) for 10 min , rinsed and dehydrated . Sections were then hybridized under coverslips overnight at 50°C with RNA probes ( produced using DIG RNA labelling kit ( Roche ) ) diluted in DIG easy Hyb solution ( Roche ) following the manufacturer’s instructions . Following hybridization , the slides were extensively washed in 0 . 1x saline sodium citrate and 0 . 5% SDS at 50°C ( 3 h ) , blocked for 1 hour in 1% blocking solution ( Roche ) in TBS and for 30 minutes in BSA solution ( 1% BSA , 0 . 3% Triton-X-100 , 100mM Tris-HCl , 100mM NaCl , 50mM MgCl2 ) , and then incubated in a 1/3000 dilution of in alkaline phosphatase-conjugated antidigoxigenin antibody ( Roche ) in BSA solution for 2h at RT . Sections were extensively washed in BSA solution , rinsed and treated overnight in the dark with a buffered NBT/BCIP solution . Samples were rinsed in water before air drying and mounting in Entellan ( Sigma ) . Embryos were imaged by gently bursting seeds between slide and cover-slip in water and imaging using a dipping lens with a long working distance . Confocal imaging was carried out on a Zeiss LSM700 with a W N-Acroplan 40x/0 . 75 M27 objective . mVENUS was excited using a 488nm diode laser and fluorescence was collected using a 490–555 nm PMT . Light microscopy imaging was carried out using a Zeiss axioimager 2 . Images were acquired using bright field illumination . 4-day-old etiolated seedlings were fixed with 4% PFA ( paraformaldehyde ) ( Sigma-Aldrich ) in 1X PBS with at least 1 hour under vacuum as described in [74] . Fixed seedlings were washed twice for 1 min in 1X PBS before being transferred for at least 6-days to ClearSee solution ( 10% w/v xylitol ( Sigma-Aldrich ) ; 15% w/v sodium deoxycholate ( Sigma-Aldrich ) ; 25% w/v urea ( Sigma-Aldrich ) ; water ) at room temperature with gentle agitation . Cleared etiolated seedlings were stained with a 0 . 5% solution of Auramine O ( Sigma-Aldrich ) in ClearSee . After 12–16 hours of incubation in the dark , seedlings were washed briefly in ClearSee then washed again for 30min and then again for at least 1h before being placed between slide and coverslip with ClearSee as the mounting solution . Confocal imaging was performed using a Zeiss LSM700 with a Plan-Apochromat 63x/1 , 4 oil DIC M27 ( ref 420782–9900 ) objective . Auramine O fluorescence was imaged with a 488nm diode laser excitation and detection of emission from 505-530nm . Images were then processed in the Zeiss LSM Image Browser Program . Cuticle composition and content was analyzed as previously described [75 , 76] . For transmission electron microscopy analysis , seeds were removed from siliques by removal of the replum tissue with attached seeds . Seeds were high-pressure frozen with a Leica EM-PACT-1 system . Three seeds were inserted into a flat copper carrier , fast-frozen , and cryosubstituted into the Leica AFS1 device . The different freeze-substitution steps were as follows: 54 h at −90°C in acetone solution containing 0 . 2% glutaraldehyde , 1% osmium tetroxide , and 0 . 1% uranyl acetate . The temperature was then raised with a step of 2°C/h before remaining for 8 hours at -60°C . The temperature was raised again to -30°C for 8h00 before being increased to 4°C . Samples were washed three times for 10 min in 100% acetone before embedding in Spurr’s resin , which was performed progressively ( 8 h in 25% Spurr’s resin in acetone , 24 h in 50% Spurr’s resin in acetone , 24 h in 75% Spurr’s resin in acetone , and two times for 12 h in 100% Spurr’s resin ) . Polymerization was performed at 70°C for 18 h . Samples were sectioned ( 65 nm sections ) and imaged at 120 kV using an FEI TEM tecnai Spirit with 4 k x 4 k eagle ccd . Microarray analysis was carried out at a Transcriptomic Platform , POPS , at the Institute of Plant Sciences Paris-Saclay ( IPS2 , Orsay , France ) , using a CATMAv7 array based on AGILENT technology [77] . The CATMAv7 array for the Arabidopsis thaliana genome was made using gene annotations included in FLAGdb++ , an integrative database of plant genomes ( http://urgv . evry . inra . fr/FLAGdb , [78] ) . The single high density CATMAv7 microarray slide contains four chambers , each containing 149 916 primers . Each 60 bp primer is present in triplicate in each chamber for robust analysis , and as both strands . The array contains 35 754 probes ( in triplicate ) corresponding to genes annotated in TAIRv8 ( among which 476 probes correspond to mitochondrial and chloroplast genes ) , 1289 probes corresponding to EUGENE software predictions , 658 probes to miRNA/MIRs and 240 control probes . 3 independent biological replicates were produced . For each biological repetition and each point , RNA samples were obtained by pooling RNAs from staged siliques containing embryos at the pre-globular to globular , or the young to late heart stage . Total RNA was extracted using the Spectrum Plant Total RNA Kit ( Sigma-Aldrich ) according to the suppliers’ instructions . For each comparison , one technical replicate with fluorochrome reversal was performed for each biological replicate ( i . e . four hybridizations per comparison ) . The labelling of cRNAs with Cy3-dUTP or Cy5-dUTP was performed as described in the Two-Color Microarray-Based Gene Expression Analysis Low Input Quick Amp Labelling manual ( Agilent Technologies , Inc . ) . The hybridization and washing steps were performed according to the Agilent Microarray Hybridization Chamber User Guide instructions ( ( Agilent Technologies , Inc . ) . Two micron scanning was performed with InnoScan900 scanner ( Innopsys , Carbonne , FRANCE ) and raw data were extracted using Mapix software ( Innopsys , Carbonne , FRANCE ) . Experiments were designed with the statistics group of the Unité de Recherche en Génomique Végétale . For each array , the raw data comprised the logarithm of median feature pixel intensity at wavelengths 635 nm ( red ) and 532 nm ( green ) . For each array , a global intensity-dependent normalization using the loess procedure [79] was performed to correct the dye bias . The differential analysis was based on log-ratio averaging over the duplicate probes and over the technical replicates . Hence the number of available data points for each gene equals the number of biological replicates and is used to calculate the moderated t-test [80] . Analysis was carried out using the R software ( http://www . R-project . org ) . Under the null hypothesis , no evidence that the specific variances vary between probes was highlighted by Limma and consequently the moderated t-statistic was assumed to follow a standard normal distribution . To control the false discovery rate , adjusted p-values found using the optimized FDR approach [81] were calculated . We considered as being differentially expressed , the probes with an adjusted p-value ≤ 0 . 05 . The function SqueezeVar of the library Limma was used to smooth the specific variances by computing empirical Bayes posterior means . The library kerfdr was used to calculate the adjusted p-values . Microarray data from this article were deposited at Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo/ ) , accession no . GSE68048 ) and at CATdb ( http://tools . ips2 . u-psud . fr/cgi-bin/projects/CATdb/consult_project . pl ? project_id=383 ) according to the “Minimum Information About a Microarray Experiment” standards . Intact siliques were frozen in liquid nitrogen and total RNA was extracted using the Spectrum Plant Total RNA Kit ( Sigma ) . Total RNAs were digested with Turbo DNA-free DNase I ( Ambion ) according to the manufacturer’s instructions . RNA was reverse transcribed using the SuperScript VILO cDNA Synthesis Kit ( Invitrogen ) according to the manufacturer’s protocol . PCR reactions were performed in an optical 384-well plate in the QuantStudio 6 Flex System ( Applied Biosystems ) , using FastStart Universal SYBR Green Master ( Rox ) ( Roche ) , in a final volume of 10 μl , according to the manufacturer’s instructions . The following standard thermal profile was used for all PCR reactions: 95°C for 10 min , 40 cycles of 95°C for 10 s , and 60°C for 30 s . Data were analysed using the QuantStudio 6 Flex Real-Time PCR System Software ( Applied Biosystems ) . As a reference , primers for the EIF4A cDNA were used . PCR efficiency ( E ) was estimated from the data obtained from standard curve amplification using the equation E = 10−1/slope . Expression levels are presented as E-ΔCt , where ΔCt = CtGOI-CtEIF4A . Primers are listed in S2 Table . The lids of plates containing etiolated seedlings were removed and plates were immediately flooded with staining solution [0 . 05% ( w/v ) Toluidine Blue + 0 . 4% ( v/v ) Tween-20] for one minutes . The staining solution was poured off and plates were immediately rinsed gently by flooding under a running tap until the water stream was no longer visibly blue ( 1–2 minutes ) . Seedlings were photographed under a Leica MZ12 stereomicroscope . For embryo staining , flowers were marked at anthesis and seeds were dissected from siliques of the same age for all genotypes in parallel . Embryos were removed from seeds by dissection with fine forceps and placed in handling baskets , which were then submerged in toluidine blue solution for one minute prior to rinsing 5 times in an excess of tap water . Embryos were photographed using a Keyence VHX-900 microscope . Seedlings or seeds were quickly frozen in liquid nitrogen and proteins were extracted in buffer containing 50 mM Tris pH 7 . 5 , 200 mM NaCl , 1 mM EDTA pH 8 , 10% glycerol , 0 . 1% tween 20 , 1 mM phenylmethylsulfonyl fluoride , 1 mM dithiothreitol , 1x protease inhibitor cocktail P9599 ( Sigma-Aldrich ) , and 1x MS-Safe protease and phosphatase inhibitor cocktail ( Sigma-Aldrich ) . Equal amounts of proteins ( 20 μg for seedlings and 10 μg for seeds ) were resolved on 10% polyacrylamide gels and transferred onto a nylon membrane ( Schleicher & Schuell ) . For seedlings primary antibodies against phospho p44/42 MAP kinase ( 1:2000 dilution ) ( Cell Signaling Technologies ) and then against MPK6 ( 1:10000 dilution ) ( Sigma-Aldrich ) were used with horseradish peroxidase-conjugated anti-rabbit as secondary antibody . Signal detection was performed using the SuperSignal West Femto Maximum Sensitivity Substrate kit ( Pierce ) . For seeds primary antibodies against phospho p44/42 MAP kinase and then against MPK6 were used with IRDye 800CW Donkey anti-Rabbit IgG ( H + LI-COR , 1:10000 dilution ) , and the bound complex was detected using the Odyssey Infrared Imaging System ( Li-Cor; Lincoln , NE ) . The images were analysed and quantified with ImageJ . Background was subtracted for each band . To test the linearity of the detection , 5–15 μg protein from heart stage developing seeds were treated as previously . To detect the antibody against phospho p44/42 MAP kinase an anti-Rabbit IgG , HRP conjugate ( Amersham , 1:30000 ) was used . Anti-alpha-tubulin ( Sigma , 1:2000 ) was used with an anti-mouse IgG , HRP conjugate ( GE HealthCare , 1:10000 ) . Signal detection was performed using Clarity Max Western ECL Substrate ( Biorad ) with a ChemiDoc Touch ( Biorad ) instrument . The images were quantified with ImageJ . Background was subtracted for each band .
|
Plant embryogenesis occurs deep within the tissues of the developing seed , and leads to the production of the mature embryo . In Arabidopsis and many other plant species embryo-derive structure ( such as the cotyledons ) are suddenly exposed to environmental stresses such as low humidity . In these species the embryonic cuticle provides a primary defence against environmental stress , and particularly dehydration , at germination . The formation of an intact and functional cuticle during embryogenesis is thus of key importance for seedling survival . Our work shows that a signalling pathway involving receptor-kinases expressed in the embryo epidermis , and a protease expressed in the endosperm tissue surrounding the embryo , is critical for ensuring the production of an intact cuticle . Furthermore , we show that a component of stress-related MAP-Kinase signalling in plants acts downstream in this pathway , possibly to mediate transcriptional responses characteristic of responses to stress . We propose that plants have redeployed a signalling pathway associated with stress resistance to ensure the formation of an intact embryonic cuticle prior to germination , and thus ensure seedling survival at germination .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[] |
2019
|
A stress-response-related inter-compartmental signalling pathway regulates embryonic cuticle integrity in Arabidopsis
|
Weight control diets favorably affect parameters of the metabolic syndrome and delay the onset of diabetic complications . The adaptations occurring in adipose tissue ( AT ) are likely to have a profound impact on the whole body response as AT is a key target of dietary intervention . Identification of environmental and individual factors controlling AT adaptation is therefore essential . Here , expression of 271 transcripts , selected for regulation according to obesity and weight changes , was determined in 515 individuals before , after 8-week low-calorie diet-induced weight loss , and after 26-week ad libitum weight maintenance diets . For 175 genes , opposite regulation was observed during calorie restriction and weight maintenance phases , independently of variations in body weight . Metabolism and immunity genes showed inverse profiles . During the dietary intervention , network-based analyses revealed strong interconnection between expression of genes involved in de novo lipogenesis and components of the metabolic syndrome . Sex had a marked influence on AT expression of 88 transcripts , which persisted during the entire dietary intervention and after control for fat mass . In women , the influence of body mass index on expression of a subset of genes persisted during the dietary intervention . Twenty-two genes revealed a metabolic syndrome signature common to men and women . Genetic control of AT gene expression by cis signals was observed for 46 genes . Dietary intervention , sex , and cis genetic variants independently controlled AT gene expression . These analyses help understanding the relative importance of environmental and individual factors that control the expression of human AT genes and therefore may foster strategies aimed at improving AT function in metabolic diseases .
Obesity is characterized by an excess of fat deposited in adipose tissue ( AT ) as triglycerides . An increase in adiposity is associated with increased risk of cardiovascular disorders and metabolic abnormalities , including hypertension , insulin resistance , type 2 diabetes , obstructive sleep apnea and cancers . Diet-induced weight loss prevents risk for type 2 diabetes and metabolic syndrome [1] , [2] , emphasizing the pivotal role of AT in obesity-related complications . As a key target tissue of dietary intervention and a node of integration between metabolism and immunity , adaptations occurring in AT are likely to have a profound impact on the whole body response [3] , [4] . Obesity is a complex disorder with numerous contributing environmental and genetic factors . A multidisciplinary research effort involving a combination of clinical , biochemical and omics approaches appears mandatory to increase knowledge in the complexity of biological traits and processes associated with obesity [5] . Through probing of the transcriptional activity of tissues , the techniques allowing systematic analysis of AT gene expression have proved useful at identifying master genes [6] and regulatory networks involved in human obesity and related disorders [7] . Moreover , mRNAs are molecular species easily and evenly amplified . Hence , mRNA profiling remains one of the most powerful methods to comprehensively explore minute amounts of tissue . Real-time PCR , which provides great dynamic range and sensitivity , is a low throughput and time-consuming technology . DNA microarray analysis allows genome-wide profiling often applied to small subsets of samples . Combining benefits of both approaches recently became possible through the emergence of microfluidic-based technologies that use very limited sample and reagent quantities [8] . Moreover , large-scale investigation of gene expression in small AT samples obtained from microbiopsy has been impaired by poor yield of total RNA due to the richness in lipid . Optimization of AT biopsy handling and total RNA extraction is thus an essential step to profitably use AT samples for gene profiling applications . The DiOGenes trial is one of the largest longitudinal dietary interventions worldwide consisting in an 8-week weight loss diet and a 26-week weight control phase with different dietary regimes [9] , [10] . The prospective long-term , randomized , controlled study design offered a unique opportunity to apply genomics technology to dietary intervention aimed at maintaining weight loss . In this study , we applied an improved total RNA preparation from AT to the thousands of samples available during the DiOGenes study . Using a novel microfluidic technology , quantitative expression analysis of AT genes was performed in individuals from this cohort . The relationship between mRNA levels and bio-clinical and genetic data was investigated . These integrative analyses provide evidence of composite control of AT gene expression by nutrition , metabolic syndrome , body mass index ( BMI ) , sex and genotype .
Despite recent development in single-step techniques dedicated to lipid-enriched samples , total RNA extraction from AT had to be improved before application to AT analysis in the DiOGenes clinical trial . Each step of total RNA extraction from small amounts of human AT samples was optimized in order to prevent the loss of precious samples ( Table S1 , Figure S1 ) . In the context of large scale clinical programs , we also investigated whether long term storage of fat samples may have negative impact on total RNA integrity . AT samples frozen in liquid nitrogen can be stored at −80°C up to 3 years without affecting total RNA yield ( Figure S1 ) or quality ( data not shown ) . Flash freezing in liquid nitrogen before storage proves as efficient as soaking the samples in preservative solutions . This is a critical point as it allows use of fat samples for other applications than transcriptomics . Different approaches were used for real time qPCR data normalization . Use of the simple 2−ΔCt method with GUSB as a reference transcript proved to be the best for normalization in human subcutaneous AT ( Figure S2 ) . The DiOGenes dietary intervention consisted of two phases [9] , [10] . The first phase was an 8-week low-calorie diet ( LCD ) with the objective of ≥8% weight loss . In the second phase , the successful patients were randomized into one of five ad libitum weight maintenance diets ( WMD ) : four diets combining high and low protein content with high and low glycemic index of carbohydrates and a control diet according to National dietary guidelines on healthy diets . Clinical investigations including subcutaneous AT microbiopsies were performed before and at the end of each phase . Five hundred sixty eight obese individuals , age 24 to 63 ( mean weight: 99 . 6±17 . 1 kg ) had clinical data available and good quality AT RNA samples . Two groups of patients were defined ( Figure S3 ) . The first group , group A , included 311 obese individuals ( 107 men and 204 women ) with gene expression data available at each clinical investigation day . The second , group B , had 204 individuals with gene expression data available at baseline and after LCD . Subjects were also categorized according to the occurrence of metabolic syndrome at baseline [11] . Group A had 125 metabolic syndrome and 186 non-metabolic syndrome individuals at baseline . Group B had 81 metabolic syndrome and 123 non-metabolic syndrome individuals at baseline . All baseline anthropometric and plasma characteristics are described in Table S2 . In both men and women , blood pressure , triglycerides , HDL-cholesterol , C reactive protein , adiponectin , fasting glucose and insulin were significantly different in metabolic syndrome compared to non-metabolic syndrome individuals . In addition , women with metabolic syndrome had higher weight , BMI , fat mass and waist circumference . Massive parallel reverse transcription quantitative PCR ( RT-qPCR ) was performed on AT from the DiOGenes study using a microfluidic qPCR device [8] . AT expression data from 271 genes of interest ( Table S3 ) was investigated on 1341 samples from 515 subjects . The genes were selected from our previous published and unpublished DNA microarray analyses on a limited number of individuals . The choice was made using the following criteria: regulation during dietary weight loss programs [12]–[14] , including the DiOGenes trial [14] , and differential expression according to the presence or absence of obesity and metabolic syndrome [15] , [16] . Forty percent of these genes encoded proteins involved in metabolism and 23% participated in immune response . This list encompassed 38 AT macrophage [15] , [16] and 84 adipocyte markers [12] , [15] , i . e . genes expressed in these cell types at much higher levels than in any other AT cell type . Controlling for weight variation , a majority of genes were regulated in both men and women by the dietary weight management program . The main pattern observed on 175 genes was an opposite regulation of AT gene expression between LCD and WMD phases ( Figure 1 , Table S4 ) . Genes downregulated during LCD and upregulated during WMD ( n = 158 ) were mostly associated with metabolic functions ( n = 110 ) , including 72 genes defined as adipocyte markers . The top ranking genes included SCD , FADS1 and FADS2 encoding enzymes involved in unsaturated fatty acid synthesis . An inverse trend was seen for 17 genes including 9 immunity-related genes . Four of those genes were AT macrophage markers . Most of the genes had similar expression at the end of the intervention compared to baseline . Forty three genes showed variations in expression at the end of WMD compared to baseline ( Table S5 ) . The majority showed decreased expression compared to baseline . LEP showed a downregulation during calorie restriction that persisted until the end of the intervention . As shown in Figure 1 , this pattern is superimposable with the evolution of HOMA-IR , an index of insulin resistance . The macronutrient composition of the diet during the WMD phase had no effect on AT gene expression . We also looked at genes related to weight changes during the ad libitum WMD by comparing changes in mRNA levels between the end of LCD and the end of WMD in women who lost and those who regained at least 50% of the weight lost during calorie restriction . The changes in mRNA levels of 16 genes differed between the two groups of women ( Table S6 ) . CIDEA which is involved in fat cell lipid droplet metabolism was the best marker for weight loss ( Figure 2 ) . FADS1 , encoding a fatty acid desaturase , and BCAT1 , encoding a branched-chain amino acid aminotransferase , were the best markers for weight regain . Principal component analysis of gene expression data from group A and group B subjects at baseline indicated that the major component explaining AT gene expression data distribution was sex . Figure S4a depicts partial least square-discriminant analysis of AT genes with sex specificity . To list the AT genes with sex-biased expression , a mixture model controlling for centre was first built with data from group A . One hundred and eighty six genes exhibited sex specificity . The same model was then run with data from group B , giving a list of 158 genes . Sex specificity for 109 transcripts persisted during the dietary intervention ( Table S7 ) . Higher expression in female AT was found for all genes except for CCL19 , which showed higher expression in male AT ( Figure 3 ) . Fat mass being higher in women than in men could possibly explain this marked sexual dimorphism . However , 88 genes remained different when controlling for fat mass . Only 5 genes were located on sex chromosomes ( Table S7 ) . SAA4 , AZGP1 , CDKN2C and CES1 were the highest ranked genes with a more than two-fold higher expression level in female than in male AT ( Figure 3 ) . Exploratory analysis of AT gene expression also indicated a discriminatory effect according to the presence or absence of metabolic syndrome ( Figure S4b ) . Because the clinical presentation of metabolic syndrome is different in men and women and might be at least in part originating in the AT [16] , [17] , we separately analysed the 2 populations to assess the molecular characteristics of AT from patients with metabolic syndrome . A metabolic syndrome signature was found for 22 genes ( Table S8 ) . CCL3 and AZGP1 showed two-fold higher and lower expression , respectively , in women with metabolic syndrome compared to women without metabolic syndrome ( Figure 4 ) . The difference , albeit less pronounced , was also present in men . To assess the contribution of obesity to AT gene expression , the impact of BMI was studied in men and women separately at baseline and along the dietary intervention . In women , 51 genes showed significant BMI dependency that persisted during the whole dietary intervention ( Table S9 ) . In men , a single gene , AZGP1 , was dependent on BMI at each time point of the intervention ( data not shown ) . To elucidate the relationship between AT gene expression and related phenotypes at a greater depth , we used a co-correlation network-based approach . Interactions between the two matching bio-clinical and AT gene expression data sets were modeled using partial correlations . By eliminating over-estimation of co-correlations due to correlation with a third variable , partial correlations measure direct correlation between two variables with control for confounding variables . First , in addition to bio-clinical parameters ( Table S2 ) , we selected 38 metabolism and immunity related genes that showed regulation in the present analysis ( Table S10 ) . Since sex appeared to control AT gene expression variance more than other parameters , we used the network approach on the 180 men and 335 women separately . Figure S5 displays the regulatory networks in men and women . Edges represent direct and strong correlations and thickness connection strength between variables . Each node represents a variable . Node degree refers to the number of edges attached to the node . High degree indicates hubs which are the most connected variables . Betweenness centrality quantifies the importance of a variable within network . Nodes with highest betweenness centrality are those providing the strongest network connection and show key variables in the network . The topology of the male and female networks was similar with 75% of edges in common between men and women . A majority of highly connected co-expression networks consisted of the same clinical parameters and genes that clustered together in both men and women . Several macrophage markers ( Table S3 ) showed strong connection . A module consisting of the same group of correlated genes in both the male and female networks encompassed genes involved in de novo lipogenesis such as FASN , SCD , FADS1 , FADS2 and ELOVL5 . In men and women , CIDEA and AZGP1 , two cachexia markers , were connected . Next , the network-based approach was used on 204 women from group A ( Table S2 ) to expand the analyses to functional interconnections during the dietary intervention . The list of genes consisted of the lipogenesis module observed at baseline ( Figure S5 ) extended to related glycolysis and glucose metabolism genes ( Table S11 ) . Networks were built during LCD ( Figure 5a ) and from baseline to the end of the WMD ( Figure 5b ) . The two data-driven dependency networks in AT showed 62% of shared edges . HK1 appeared as an important hub , with 10 connections , including another hub , SCD , ACACB , ELOVL5 and fasting glucose ( Figure 5a ) . Notably , two components of the metabolic syndrome , waist and triglycerides , were highly connected to the most important node , SCD , which was also highly connected to glycolytic genes ( SLC2A4 and PCK1 ) during LCD . From baseline to the end of the WMD ( Figure 5b ) , fat mass , waist and triglycerides were associated with the network key node ALDOC , which was connected to several lipogenic genes ( FASN , ACACB , SCD , FADS2 and ELOVL5 ) . We identified 2953 single nucleotide polymorphisms ( SNPs ) which were in the close proximity of 252 genes . At baseline , 118 SNPs representing 46 genes showed association with AT gene expression ( Table S12 ) . The strongest associations ( P<10−10 ) were found for ALDOB , MARCO , MMP9 and HLA-A ( Figure 6 ) . Four SNPs located in the intronic regions of MARCO , which encodes an AT macrophage-specific marker regulated by obesity [15] and dietary intervention ( Table S5 ) , showed associations with P<10−20 . A moderate effect of sex and BMI was observed for 3 and 13 SNPs , respectively . However , these effects were not consistent among SNPs with significant associations with AT gene expression in the corresponding genes . The majority of the associations observed at baseline remained significant when expression after LCD and WMD were considered ( Table S13 ) . Of note , no SNP showed association with diet-induced variations in mRNA levels ( P>0 . 5 ) .
Carefully monitored weight-control diets favorably affect parameters of the metabolic syndrome and delay the onset of diabetic complications [1] . AT is a key player in the homeostatic control of whole body metabolism . Besides the more recent DNA methylome and microRNA arrays , gene expression profiling is one of the most comprehensive omics technologies , which permits parallel quantitative measurements of a large number of targets . However , a minimal amount of high quality total RNA is required . To study large intervention programs , we optimized the entire process , from needle biopsy of human subcutaneous AT to long term storage of samples . A single needle biopsy allows fast and painless AT sampling that may be easier to perform than blood sampling in morbidly obese subjects . Biopsies of about 200 mg are easily obtained in large scale intervention studies and can also be obtained from lean individuals . From this amount of tissue up to 34 , 100 different transcripts can be quantified using the Fluidigm Biomark Dynamic Array technology . This high-throughput technology has been successfully applied to many biological fields [18]–[20] . It allows reduction in cost and time , and improvement of accuracy , throughput and performance compared with conventional instruments . We show here its potencies to study AT expression of multiple genes in large-scale population-based interventions . Normalization is an essential step to correct for systematic bias in transcriptomic data . In microfluidic RT-qPCR assays , systematic errors due to sampling , reverse transcription and preamplification steps as well as set and plate spreading must be eliminated . Numerous different methods can be used for data normalization , including those used for transcriptome analyses [21] . When analyzing real-time PCR data , the most widely used is the 2−ΔCt method using a reference gene [22] . The ideal reference , also referred to as a house-keeping gene , should be constantly transcribed in all cell types and tissues regardless of internal and external influences . However , the expression of house-keeping genes may vary considerably . GAPDH is one of the most commonly used house-keeping genes . However , GAPDH expression was shown to be regulated during dietary intervention in this study and previous work [12] . The ribosomal 18S is another common reference gene . However , because of its very high expression level , this transcript shows such a strong expression after the preamplification step that it overflows the detection system . Here , the use of GUSB as a reference gene in the easy-to-use 2−ΔCt method proved to deliver the best normalization for human subcutaneous AT . AT gene expression profiling combined with clinical investigations has opened a novel approach to decipher physiological and pathophysiological processes [5] . Most previous studies have aimed at comparing obese and lean individuals and observing the effects of altered body weight during dynamic weight change . The highly clinically relevant weight stabilization phase has rarely been studied . We have previously investigated AT gene expression during multiple phase dietary interventions on a limited number of individuals [12] . The DiOGenes intervention study was designed to investigate diet-induced changes on a much larger scale [10] . This trial focused on identifying key dietary and genetic factors as a basis for predicting whether individuals may reach and maintain healthy weight . The number of participants and the complete bioclinical characterization combined with AT biopsy offered a unique opportunity to study the interactions between sex , metabolic status , dietary phases and genetic factors . Unexpectedly , a striking sex effect on AT gene expression that persisted during the dietary intervention for most of the sex-specific genes was found . As a prototypical example , it has repeatedly been shown that leptin AT mRNA level is higher in females than in males even after correction for the degree of body fat mass [23] . Similar sexual dimorphism has previously been reported for 45 of the genes described in the present study in mouse AT [24] . Higher expression in female than in male AT was found for all but 1 gene . However , to ascertain that the majority of human AT genes showing sex differences have female-biased expression , a systematic analysis would be required . Higher fat mass in women than in men could possibly explain this marked sexual dimorphism . However , 80% of the genes remained different when controlling for fat mass . Indeed , SAA4 and CDKN2C AT expression are increased and decreased , respectively , in morbidly obese subjects whereas the two genes show higher expression in women than men [25] , [26] . The contribution of gonadal hormones and sex chromosomes has been investigated in mouse models . Sexual steroids play a strong role in sex-biased gene expression in various tissues with minor sex differences explained by direct effects of the sex chromosome in liver genes [27] . Here , less than 5% of the human AT sex-related genes were located on sex chromosomes . The large number of subjects with a wide range of adiposity allowed testing the effect of BMI on AT gene expression at baseline and during the dietary intervention . In women , 51 genes show BMI dependency which persists at each time point of the dietary intervention indicating adiposity-dependent control of gene expression that is not influenced by diet-induced changes in weight . AZGP1 was the only common gene to men and women and among the top ranking less expressed genes in the morbidly obese patients ( BMI>40 ) . It was recently shown as down regulated with fat mass expansion in obesity in both visceral and subcutaneous fat with positive association with adiponectin [28] . Of note , the network approach also showed such connection in the male network . The lower number of significant genes in men may be related to real sex differences but may also be due to the lower number of men in the cohort . A metabolic syndrome signature was also found in AT . As a top ranking gene , CCL3 encodes macrophage inflammatory protein 1α , a CC chemokine involved in the interactions between immune cells and regulated by insulin resistance in AT [29] . AZGP1 appears as a marker of sexual dimorphism , obesity and metabolic syndrome encodes an adipokine with putative antidiabetic properties [30] . When looking along the DiOGenes dietary program , the main pattern was an opposite regulation of AT gene expression between LCD and WMD phases . Genes downregulated during LCD and upregulated during WMD were mostly adipocyte genes associated with metabolic functions [12] . The top ranking genes encoded enzymes involved in fatty acid desaturation [31] . An inverse trend was seen for immunity-related genes . As a result of the opposite regulation between LCD and WMD , most of the genes had similar expression at the end of the intervention compared to baseline . However , a subset of genes showed downregulation at the end of the dietary intervention . The list included several genes previously characterized as human AT macrophage-specific markers ( CD68 , CD163 , CD209 , IL10 , LIPA , MARCO , MS4A4A , PLA2G7 , SPP1 ) [12] , [15] . This coordinated downregulation most likely reflects a decrease in AT macrophage number as observed in a 6-month weight reducing intervention [32] . Leptin mRNA levels were also lower at the end of the dietary intervention . The superimposition of LEP and HOMA-IR data lend support to our hypothesis that variation in LEP expression contributes to the improvement in insulin sensitivity observed during diet-induced weight loss [3] . Genes related to weight changes during WMD include AGPAT9 , presumably involved in the biosynthesis of triacylglycerol and phospholipid , ALOX12 , encoding the arachidonate12-lipoxygenase involved in production of inflammatory and adipogenesis mediators , and the proangiogenic VEGFA , which all showed robust overexpression in individuals who regained weight . PKM2 , AP2M1 , ACTR3 and CES1 had been shown in microarray experiments to have higher expression in individuals who failed to control their weight [14] . FADS1 , encoding a fatty acid desaturase , and BCAT1 , encoding a branched-chain amino acid aminotransferase , were the best markers for weight regain . Interestingly , branched-chain amino acid catabolism is down-regulated in obese individuals [33] . CIDEA , which plays a critical role in fat cell lipid droplet metabolism [34] , showed decreased expression in individuals continuing to lose weight after LCD , supporting a role for the encoded protein in the adaptation of subcutaneous AT to body weight changes as over-feeding-induced weight gain induces its expression [35] . The macronutrient composition of the diet ( i . e . , protein content and glycemic index ) during the WMD phase had no effect on AT gene expression in agreement with previous data on energy-restricted diets differing in fat and carbohydrate content [13] . Therefore , during dietary weight management programs , energy balance and fat mass variations rather than the composition of the diet is a determinant of AT gene expression . During LCD , network analysis of gene expression and clinical parameters showed that the top associations function as part of a major hub gene , the stearoyl CoA desaturase SCD , which is highly connected to components of the metabolic syndrome and the gene encoding a glycolytic enzyme , hexokinase HK1 , connected to glycemia . These genes are targets of ChREBP , a transcription factor involved in glucose-mediated control of de novo lipogenesis gene expression [36] , [37] . These connections suggest that ChREBP target genes are regulated during LCD . This transcription factor was not studied here . The link between ChREBP and metabolic improvements along the dietary intervention requires further investigation [38] . Along the dietary intervention , clinical data and gene co-expression network analysis also revealed ALDOC , an aldolase involved in glycolysis , and fat mass as key nodes . Both hubs were connected to components of the metabolic syndrome . The ALDOC-centered module included key genes for de novo lipogenesis , illustrating the common transcriptional control of glycolysis and fatty acid synthesis [37] . The fat mass centered module was composed of glycolytic genes , indicating a direct link between change in fat mass and aerobic glycolysis , which seems to be related to the connection between de novo lipogenesis gene expression and metabolic features [39] . Human AT gene expression is under strong genetic control [40] . Recent genome-wide gene expression and genotyping analysis identified 10 , 000 cis SNPs associated to gene expression in subcutaneous AT [41] . The number of cis expression SNPs ( eSNPs ) was much higher than the number of trans eSNPs . In the present study , more than 80% of the genes with eSNPs had not previously been reported [41] . This high level of detection was related to several factors . First , we selected SNPs located in the immediate vicinity of the genes that allow capture of significant associations with our sample size [42] . Second , we investigated a carefully selected population enrolled in a multicentric dietary intervention [9] , [10] . Thereby , we could control for biological and non-biological confounders such as center , sex , fat mass and diet . Highly significant associations were found for MARCO and MMP9 . MARCO encodes a class A scavenger receptor shown to be specific of AT macrophages compared to other human AT cell types [15] . AT macrophages also specifically produce metalloproteinase 9 , a key enzyme involved in remodeling processes [43] . Of 46 genes with eSNPs , 19 were directly related to immunity and inflammation and were highly expressed in human AT , in agreement with the existence of an AT macrophage gene network module with tight cis genetic control [40] . Strikingly , the eSNPs identified here were not influenced by sex and diet-induced changes in AT gene expression . We found no evidence of association between cis SNPs and variations in mRNA levels during the dietary protocol , suggesting that cis genetic control operates at baseline and is preserved during the dietary intervention but does not influence the response to the diet . As prototypical examples , ACSL1 , ECHDC3 and HSDL2 mRNA levels were influenced by all the investigated factors ( Figure 7 ) . SNPs did show associations with AT mRNA levels as transcript abundance varied during the dietary intervention . It also differed according to sex and metabolic syndrome . Dietary intervention did not alter the sexual dimorphism in gene expression . ACSL1 , which catalyzes the conversion of long chain fatty acids into acyl-CoAs , is the most abundant ACSL isoform expressed in AT . AT-specific ablation of Acsl1 in mice shows that the enzyme plays a crucial role in directing acyl-CoAs towards β-oxidation in fat cells [44] . Here , ACSL1 gene expression was lower in individuals with metabolic syndrome . An association between ACSL1 gene polymorphisms and the metabolic syndrome has recently been reported [45] . These data suggest that impaired adipocyte fatty acid oxidation due to ACSL1 defect may constitute a feature of the metabolic syndrome . The present data provide evidence for control of AT gene expression by nutrition , sex , metabolic status and genotype . A main feature was a major effect of sex , which was independent of sex chromosomes , fat mass and dietary intervention . Another characteristic was that the control of gene expression by genetic elements appeared unaffected by nutritional status . Altogether , the effects of the investigated factors were most often independent of each other . Understanding the relative importance of environmental and individual factors that control the expression of human AT genes may help in deciphering strategies aimed at improving AT function in metabolic diseases .
Fat samples used for the optimization of total RNA extraction were obtained using abdominal dermolipectomy from the plastic surgery department of the Toulouse University Hospitals . The patients were not included in a weight reduction program . DiOGenes was registered in ClinicalTrials . gov ( NCT00390637 ) . Briefly , 932 overweight and obese adults in 8 European centers participated in a dietary program with a 8-week 3 . 3 MJ/day LCD ( Modifast , Nutrition et Santé , France ) . Subjects achieving 8% of initial body weight loss were randomized to a 6-month ad libitum WMD consisting in one of four low-fat diets that differed in glycemic index and protein content , or a control diet as described in [9] . Abdominal subcutaneous AT biopsies from the DiOGenes protocol were obtained by needle aspiration under local anesthesia after an overnight fast at baseline , at the end of LCD and at the end of WMD [14] . All clinical investigations were performed according to standard operating procedures . Analysis of blood samples was performed at the Department of Clinical Biochemistry , Gentofte University Hospital , Denmark as described in [9] . The study was performed according to the latest version of the Declaration of Helsinki and the Current International Conference on Harmonization ( ICH ) guidelines . All subjects gave verbal and written informed consent . Applications were submitted to the regional Ethics Committees from the participating centres and the study was not undertaken without a positive statement from the committee regarding the study . Five single-step methods for total RNA extraction were evaluated on 6 human subcutaneous AT samples ( Table S1 ) . Among the 5 extraction methods , 2 yielded very few or partly degraded total RNA and one low purity RNA . The QIAGEN methods ( RNeasy Mini and RNeasy Lipid Tissue Mini ) yielded sufficient amount of good quality total RNA that appeared to be free of genomic DNA contamination . Based on the QIAGEN RNeasy Lipid Tissue Mini Kit , an in-house optimized total RNA preparation protocol that uses chloroform delipidation and phenol/guanidine isothiocyanate-based ( QIAzol ) extraction , silica-gel membrane purification and microspin technology was set up ( see below ) . This adapted protocol provided a higher mean total RNA supply with more consistent yield than other methods . Human AT samples of weights ranging from 0 . 04 g to 1 . 5 g were collected , flash frozen in liquid nitrogen and stored at −80°C . Figure S1a was drawn from 84 AT samples . It shows a positive correlation between the amount of total RNA extracted and the weight of the fat biopsies up to 0 . 2 g . Above 0 . 5 g of fat , the amount of total RNA per g of tissue becomes more variable . The data reveal that 0 . 3 to 0 . 5 g of fat are enough for substantial total RNA recovery . Such an amount may yield a minimum of 5 µg of total RNA which is sufficient for both microarray application and RT-qPCR . Besides sample preparation , storage conditions are a major concern because of the instability of mRNA due to contaminating RNases . In order to prevent total RNA degradation , commercial RNA stabilization reagents are available . Figure S1b shows adipose tissue total RNA yield and quality using alternative protocols . Samples of about 0 . 5 g of human fat tissue were collected and stored at −80°C following 5 different protocols: 1 ) immediate storage of the freshly cleaned fat sample at −80°C 2 ) flash-frozen in liquid nitrogen and stored at −80°C 3 ) stored overnight in RNAlater RNA Stabilization Reagent ( QIAGEN ) at 2–8°C , then removed from the RNAlater and stored at −80°C 4 ) freshly cleaned fat sample stored at −80°C in QIAzol Lysis Reagent ( QIAGEN ) 5 ) homogenized in QIAzol Lysis Reagent with ultra-turax homogenizer and stored at −80°C . These samples were extracted after short term , 1 month , and after long term , 1 year , storage . The 5 protocols gave similar total RNA yield and quality . The frozen AT sample was homogenized in QIAzol ( QIAGEN ) ( 2 . 5 ml of QIAzol for 500 mg of tissue , 5 ml for 1 g of tissue , 1 ml for ≤200 mg of tissue ) using a rotor-stator homogenizer until homogeneity ( 20–40 s; longer time may lead to overheating ) then incubated at room temperature for 5 min . Two hundred µl of chloroform was added for 1 ml of QIAzol ( otherwise the volume of chloroform was adjusted to QIAzol volume with a 1∶5 ratio ) and vigorously shaked for 15 s using a vortex then incubated at room temperature for 3 min . After centrifugation at 4000 rpm for 15 min at 4°C the upper aqueous phase was transferred to a new tube . One volume of 70% ethanol was added and mixed by vortexing . Seven hundred µl of this sample was pipetted onto an RNeasy Mini Spin Column ( QIAGEN ) in a 2 ml tube and centrifuged at ≥10 , 000 rpm for 15 s at 25°C . Flow-through was discarded . This step was repeated using the reminder of the sample . Seven hundred µl of Buffer RW1 ( QIAGEN ) was added to the column and centrifuged at ≥10 , 000 rpm for 15 s at 25°C . Flow-through was discarded . The column was transferred into a new 2 ml tube . Five hundred µl of Buffer RPE ( QIAGEN ) was added to the column and centrifuged at ≥10 , 000 rpm for 15 s at 25°C . Flow-through was discarded . Another 500 µl of Buffer RPE was added to the column and centrifuged at ≥10 , 000 rpm for 2 min at 25°C . The column was transferred into a new 2 ml tube and centrifuged at ≥10 , 000 rpm for 1 min at 25°C . To elute , the RNeasy Mini Spin Column was transferred to a 1 . 5 ml tube and 30 µl of preheated RNAase-free water at 50°C directly pipetted onto the column then centrifuged at ≥10 , 000 rpm for 1 min at 25°C . For the optimization of total RNA extraction from AT samples ( Table S1 ) , total RNA quality was checked using ethidium bromide-stained agarose gels . Concentration was determined using a Nanodrop spectrophotometer , Illkirch , France ) . For AT biopsies from the dietary program , total RNA concentration and quality were estimated by capillary electrophoresis using the Experion analyzer ( BioRad , Marnes-la-Coquette , France ) . The amount of total RNA from the DiOGenes study was 25 . 3±9 . 3 µg/g of AT ( n = 1363 ) validating the RNA extraction and purification method in a large multicenter study . Total RNA was of good quality and free of genomic DNA . Genomic DNA was extracted from the buffy coats with a salting out method . Genomic and amplified DNA samples were quality-checked , quantified and normalized to approximately 100 ng/ml and 2 . 0 µg before genotyping . High throughput SNP genotyping was carried out using the Illumina iScan Genotyping System ( Illumina , San Diego , CA , USA ) . Seven hundred forty eight individuals were genotyped using the Illumina 660W-Quad SNP chip . SNP genotyping was done in accordance with manufacturer's protocols . The Integrated mapping information is based on NCBI's build 37 . The coding sequences were investigated 15 kb downstream and 10 kb upstream . All SNP with a genotype frequency ≥95% and in Hardy-Weinberg equilibrium ( P>0 . 05 ) were selected for further analyses . Among 3965 SNPs related to 252 genes , 2953 SNPs remained after data filtering .
|
In obesity , an excess of adipose tissue is associated with dyslipidemia and diabetic complications . Gene expression is under the control of various genetic and environmental factors . As a central organ for the control of metabolic disturbances in conditions of both weight gain and loss , a comprehensive understanding of the control of adipose tissue gene expression is of paramount interest . We analyzed adipose tissue gene expression in obese individuals from the DiOGenes protocol , one of the largest dietary interventions worldwide . We found evidence for composite control of adipose tissue gene expression by nutrition , metabolic syndrome , body mass index , sex , and genotype with two main novel features . First , we observed a preeminent effect of sex on adipose tissue gene expression , which was independent of nutritional status , fat mass , and sex chromosomes . Second , the control of gene expression by cis genetic factors was unaffected by sex and nutritional status . Altogether , the effects of the investigated factors were most often independent of each other . Comprehension of the relative importance of environmental and individual factors that control the expression of human adipose tissue genes may help deciphering strategies aimed at controlling adipose tissue function during metabolic disorders .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genome-wide",
"association",
"studies",
"genome",
"expression",
"analysis",
"medicine",
"nutrition",
"obesity",
"gene",
"expression",
"regulatory",
"networks",
"biology",
"molecular",
"biology",
"genetics",
"genomics",
"molecular",
"cell",
"biology",
"computational",
"biology",
"genetics",
"and",
"genomics",
"human",
"genetics"
] |
2012
|
Determinants of Human Adipose Tissue Gene Expression: Impact of Diet, Sex, Metabolic Status, and Cis Genetic Regulation
|
We have developed the first computational model of solute and water transport from Bowman space to the papillary tip of the nephron of a human kidney . The nephron is represented as a tubule lined by a layer of epithelial cells , with apical and basolateral transporters that vary according to cell type . The model is formulated for steady state , and consists of a large system of coupled ordinary differential equations and algebraic equations . Model solution describes luminal fluid flow , hydrostatic pressure , luminal fluid solute concentrations , cytosolic solute concentrations , epithelial membrane potential , and transcellular and paracellular fluxes . We found that if we assume that the transporter density and permeabilities are taken to be the same between the human and rat nephrons ( with the exception of a glucose transporter along the proximal tubule and the H+-pump along the collecting duct ) , the model yields segmental deliveries and urinary excretion of volume and key solutes that are consistent with human data . The model predicted that the human nephron exhibits glomerulotubular balance , such that proximal tubular Na+ reabsorption varies proportionally to the single-nephron glomerular filtration rate . To simulate the action of a novel diabetic treatment , we inhibited the Na+-glucose cotransporter 2 ( SGLT2 ) along the proximal convoluted tubule . Simulation results predicted that the segment’s Na+ reabsorption decreased significantly , resulting in natriuresis and osmotic diuresis .
The parenchyma of a kidney is divided two major structures: the medulla and the outer renal cortex . In the multi-lobed human kidney , these structures take the shape of 8–18 cone-shaped renal lobes , with each resembling a uni-lobed rodent kidney , The outer region is the cortex , in which are clusters of capillaries , and convoluted segments of renal tubules . The inner region is the medulla , which further divides into the outer and inner medulla . Within the medulla one finds almost parallel arrangement of tubules and vessels [1] . Each human kidney is populated by about a million nephrons . Each nephron consists of an initial filtering component called the glomerulus and a renal tubule specialized for reabsorption and secretion . The renal tubule is the portion of the nephron in which the glomerular filtrate circulates before being excreted as urine . The functional role of the nephron is to adjust the composition of the urine so that wastes are excreted and that daily intake roughly equals urinary excretion . The renal tubule consists of a number of segments . Given in an order consistent with fluid flow direction , the segments are: the proximal tubule , which consists of two segments , the proximal convoluted tubule ( or , the S1-S2 segments ) and the S3 segment; the loop of Henle , which in turn consists of a descending limb and an ascending limb; the distal convoluted tubule , the connecting tubule , and the collecting duct . Each tubular segment is lined by a single layer of epithelial cells . The ultrastructure and transport properties of the epithelial cells vary widely among different tubular segments , so that different tubular segments specialize in different roles in renal water and solute transport . Generally , the proximal tubule reabsorbs the largest fraction of the glomerular filtrate , including about two-thirds of the water and NaCl , in addition to filtered nutrients like glucose and amino acids . The thick ascending limb of the loop of Henle that follows actively pumps NaCl into the interstitium of the medulla , without water following . As a result , the fluid that reaches the distal tubule is dilute relative to blood plasma . Depending on the hydration status of the body , the collecting duct exploits this hypotonicity by either allowing ( anti-diuresis ) or not allowing ( diuresis ) water to return to general circulation via osmosis [1] . To represent physiological processes and function changes of the kidney in diseases , one may employ a useful and non-invasive approach: computational modeling . Detailed models of solute and transport have been developed for renal epithelial cells [2 , 3] , tubular segments [4–6] , and populations of nephrons [7 , 8] . All these models , and other published ones , are formulated for the rat , due to the relatively plentiful anatomic , micropuncture , and electrophysiologic data available in rodents . It goes without saying that significant differences exist between the rat kidney and the human kidney , in terms of anatomy and hemodynamics . Consequently , while results obtained using a rat kidney model may shed insights into human kidney function , those results don’t always or entirely translate . To investigate human kidney function under physiological , pathophysiological , and pharmacological conditions , we have developed the first computational model of epithelial solute and water transport of the human nephron . Starting with our published computational model of the rat nephron [9] , we incorporated anatomic and hemodynamic data from the human kidney , and we adjusted key transporter data so that the predicted urine output is consistent with known human values . Due to the relative sparsity of data on the renal transporter expression levels in humans , we identified a set of compatible transport parameters that yielded model predictions consistent with human urine and lithium clearance data . Using the resulting model , we then explored the effects of two renal transporter inhibitors on kidney function . First , we considered an inhibitor of the sodium-glucose cotransporter 2 ( SGLT2 ) cotransporter , which is expressed on the apical membrane of the proximal convoluted tubule and is a novel target of diabetes drugs [10] . Under normoglycemic conditions , how does the drug impact segmental Na+ transport and urine excretion ? We also simulated inhibition of the Na+-K+-Cl− cotransporter ( NKCC2 ) , which is expressed on the apical membrane of the thick ascending limbs of the loops of Henle and aids in the active transport of Na+ , K+ , and Cl− into the cell . How substantial are the compound’s diuretic , natriuretic , and kaliuretic effects ?
The typical volume of a kidney from a 340-g rat is ∼1 . 32 ml , with the cortex , outer medulla , and inner medulla occupying 63 , 34 , and 3% , respectively , of that volume [20] . The volume of a kidney from a 30–40-year old adult human was reported to be 285 ml , with the cortex , outer medulla , and inner medulla occupying 63 , 27 , and 10% , respectively , of that volume [21] . Thus , the human kidney is >200 times bigger than a rat kidney . That large difference in volume can be attributed , in part but not solely , to the ∼30-times larger glomerulus population in a human kidney ( 1 million versus 36 , 000 ) . The model medulla is taken to be 17 mm [22] , with the outer and inner medulla taken to be 5 and 12 mm , respectively , based on Ref . [23] and personal communication from Tom Pannabecker . Tubular length and luminal diameter values are given in Table 1 . Tubular lengths were gleaned from Refs . [22–24] . Tubular luminal diameters are estimated from renal biopsy studies of health individuals [25–28] . The composition of the interstitial fluid in the cortex at the boundary between the outer and inner medulla , and at the papillary tip is specified in Table 2 . We assume that interstitial concentrations vary linearly between the cortico-medullary junction and the inner-outer medullary boundary , and between the inner-outer medullary boundary and the papillary tip . The interstitial fluid in the cortex is taken to be homogeneous except for ammonia ( see Ref . [9] ) . Tubular fluid concentrations at the proximal tubule inlet are equal to those in the cortical interstitium , except for the absence of protein . The model represents only a superficial nephron because in human , the vast majority of the nephrons ( ∼85% ) are superficial [24] . Normal human glomerular filtration rate ( GFR ) falls within the range of 90–120 ml/min/1 . 73m2 [29] . Single-nephron GFR ( SNGFR ) is taken to be 100 nl/min , assuming 1 million nephrons [30] . Proximal tubule inflow fluid pressure is taken to be 21 mmHg [31] . Unless otherwise specified , model parameters specifying transporter density and water and solute permeability are taken to be the corresponding rat values [5 , 9 , 17 , 32]; see below for exceptions .
The proximal tubule of the rat kidney is known to exhibit glomerulotubular balance [35] , such that tubular Na+ reabsorption changes in proportion to SNGFR to maintain approximately the same fractional reabsorption . Glomerulotubular balance has not been demonstrated experimentally in humans . To assess the relationship between segmental transport to SNGFR in the human nephron , we conducted simulations where SNGFR was varied by ±10% of baseline values . We assumed that changes in SNGFR are due to changes in renal blood flow , as opposed to changes in filtration fraction which are linked to changes in peritubular oncotic pressure . Thus , as was done in a previous rat modeling study [11] , we assumed no change in proximal tubule transport expression levels , except in response to changes in torque . ( Model proximal tubule transepithelial transport is assumed to be flow dependent . Details can be found in Ref . [11] ) . The predicted Na+ , K+ , Cl− , and volume deliveries to individual segments under differing SNGFR are shown in Fig 4; the corresponding solute and fluid transport are shown in Fig 5 . When SNGFR was increased by 10% , the higher luminal flow along the proximal tubule raised transcellular transport via the torque-dependent scaling . That was followed by an increase in paracellular transport , the driving force of which depends on transcellular transport . Consequently , proximal tubule Na+ reabsorption was predicted to increase by 11%; that was accompanied by a 10% increase in Cl− reabsorption . Conversely , when SNGFR was reduced by 10% , proximal tubule Na+ and Cl− transport decreased by 11 and 10% , respectively . Thus , the model proximal tubule exhibits glomerulotubular balance , albeit not 100% . As SNGFR and Na+ delivery to the thick ascending limb increased , luminal [Na+] decreased more slowly along this segment . This trend continued through the initial segment of the connecting tubule . With the higher luminal [Na+] , transcellular Na+ reabsorption increased and paracellular secretion decreased . Similar trends were observed for Cl− transport . When SNGFR increased by 10% , urine Na+ and Cl− excretion increased by 38 and 95% , respectively; when SNGFR decreased by 10% , urine Na+ and Cl− excretion decreasde by 17 and 44% , respectively . Water transport also exhibits similar trends along the water-permeable segments ( e . g . , proximal tubule , connecting tubule , collecting duct ) . See Fig 5 , panels A and D . As SNGFR and Na+ delivery to the thick ascending limb increased , K+ reabsorption via the NKCC2 increased; see Fig 5B . On the other hand , the increased Na+ delivery to the connecting tubule and the resulting higher Na+ reabsorption yielded substantially higher K+ secretion ( Fig 5B ) . Taken together , when SNGFR increased by 10% ( decreased by 10% ) , urine K+ excretion increased by 55% ( decreased by 33% ) ; see Fig 4B . The proximal convoluted tubule ( a . k . a . S1-S2 segment ) expresses the Na+-glucose cotransporter 2 ( SGLT2 ) on its apical membrane , whereas the S3 segment expresses SGLT1 . Under baseline conditions , the SGLT2 and SGLT1 mediate the uptake of approximately 90 and 10% of the filtered glucose . A new diabetic drug inhibits SGLT2 and thus renal reabsorption of glucose [10] , and has been shown to be effective in lowering blood glucose level . Because SGLT2 inhibition affects both Na+ and glucose transport , a question is: To what extent do SGLT2 inhibitors shift Na+ transport to downstream nephron segments , and to what extent do these compounds elevate Na+ excretion ? To simulate acute SGLT2 blockade , we reduced SNGFR by 3% , based on observations in non-diabetic humans receiving canagliflozin or dapagliflozin for 4 days ( personal communication by Volker Vallon ) . SGLT2 expression was reduced by 90% . These changes resulted in urine glucose excretion of 242 μ/min/kidney , which corresponds to a fractional excretion of 54% . Fractional glucose flow along the proximal tubule is shown in Fig 6 for baseline and SGLT2 inhibition . The model predicted that 17% of the filtered glucose was reabsorbed along the S1-S2 segments by the remaining SGLT2 and via the paracellular route , and 29% along the S3 segment across the SGLT1 and the tight junctions . The deliveries of Na+ , K+ , Cl− , and water to individual nephron segments are shown in Fig 7 . With SGLT2 inhibition , significantly more glucose was retained in the proximal tubule luminal fluid , thereby increasing its osmolality and inhibiting water reabsorption . In other words , SGLT2 inhibition elicited osmotic diuresis , thereby lowering proximal tubular fluid Na+ concentration and reducing passive Na+ transport via the paracellular route in that segment . Even though the higher luminal flow conversely stimulated active Na+ transport ( via torque-induced increases in transcellular transporter expression [11] ) , the reduction in passive transport was greater . Consequently , net Na+ reabsorption decreased in the proximal tubule by 17% . As can be seen in Fig 7 , SGLT2 inhibition elevated Na+ , K+ , Cl− , and volume flow in all nephron segments . Taken together , SGLT2 in euglycemic conditions resulted in diuresis , natriuresis , and kaliuresis , with urine flow , Na+ and K+ excretion increased by 144 , 46 , and 77% , respectively ( see Table 3 , column “SGLT2 inhibition” ) . Next , we simulated the effect of a diuretic ( furosemide ) that may be used in treatment of hypertension . Specifically , we simulated 80% inhibition of NKCC2 , the Na+-K+-2Cl− cotransporter that is expressed on the apical membrane of the thick ascending limbs of the loops of Henle . Inhibiting NKCC2 impairs the kidney’s ability to generate an axial osmolality gradient , with the assumption that the NKCC2 inhibitor was administered for long enough for this washout to occur . Thus , following the approach in our previous study [8] , the interstitial fluid concentrations of selected solutes were lowered . Specifically , we assumed that ( i ) cortical interstitial concentration profiles were unaffected; ( ii ) the concentrating mechanism of the outer medulla was significantly impaired , so that at the outer-inner medullary boundary , the interstitial concentrations of Na+ , K+ , Cl− , and urea were reduced to 204 , 5 . 6 , 193 , and 28 mM , respectively; ( iii ) at the papillary tip , the interstitial concentrations of Na+ , K+ , Cl− , and urea were reduced to 160 , 7 . 0 , 149 , and 56 mM , respectively . With these values , the interstitial fluid osmolality at the papillary tip was 416 mosm/ ( kg H2O ) . Segmental delivery of key solutes and fluid is shown in Fig 7 . Inhibition of NKCC2 had no direct impact on proximal convoluted tubule transport . However , because medullary interstitial fluid osmolality was assumed to be lower , water reabsorption from the S3 segment and other medullary segments was reduced , resulting in increased volume delivery to downstream segments . Urine flow increased by 2 . 8 fold , consistent with observations reported in Ref . [36] . See Table 3 , column “NKCC2 inhibition . ” Solute delivery to the thick ascending limb was largely minimally affected by 80% NKCC inhibition ( see Fig 7 , compare “Baseline” and “NKCC2 inhib” ) . Profiles of fractional Na+ flow along the thick ascending limb under baseline conditions and with NKCC2 inhibition are shown in Fig 8 . Along the thick ascending limb , Na+ transport fell by 20% . That was accompanied by a substantial reduction in the reabsorption of K+ and Cl− along the thick ascending limb ( results not shown ) . Consequently , urinary Na+ , K+ , and Cl− excretion were predicted to increase by 1 . 8 , 2 . 4 , and 3 . 6 fold with NKCC2 inhibition .
Our understanding of the kidney has been vastly improved by studies employing techniques that evaluate renal function at the single nephron level . Particularly instrumental and indispensable is the micropuncture technique [37] , which has facilitated studies of glomerular filtration and hemodynamics , and tubular epithelial activity in animal models . However , such experiments are considered too invasive to be conducted in the human kidney . To reveal microscopic processes and physiological function in the human kidney , one may utilize functional MRI [38] , a non-invasive technique that could facilitate translation of many studies performed in controlled animal models using technologies that are invasive to humans . Another alternative is the computational modeling technique . One notable application of computational models is the simulation of “clean” knockout experiments . Because unlike an animal , a computational knockout model does not need to live , it will not attempt to compensate by adjusting other transport mechanisms . In this study , we developed the very first computational model of solute and water transport along a human nephron . A number of studies have suggested that a similar set of transporters are expressed along the human and rodent nephrons ( e . g . , [39–41] ) . Thus , we based the human nephron model on our published rat model [9 , 11] . We first incorporated anatomic and hemodynamic data from humans , and then determined what additional transport parameters need to be adjusted to ensure that model predictions are consistent with known human data . Even though we did not expect a human to be a big rat , a rat model using human anatomic and hemodynamic data , without any change in transport expression levels , generate predictions that are largely ( albeit not perfectly ) consistent with human data . See Table 3 , column “GFR & dimensions . ” This result suggests general similarity between the electrophysiology of mammals that survive in similar living environment . With additional adjustments in transporter expression ( see below ) , the model predicts key tubular transport and urine output that are consistent with human values ( Table 3 , column “Baseline” ) . Model simulations were validated by comparing predicted urine output , and possibly other predictions , with measurements in humans . However , it is important to note that the model does not merely recapitulate experimental observations . The model predicts , at every single point along the nephron , luminal fluid flow , luminal solute concentrations , cytosolic solute concentrations , epithelial membrane potential , transcellular solute and water fluxes , and paracellular solute and water fluxes—most of which are virtually impossible to determine in humans in vivo under current ethical guidelines . Hence , the model suggests , under various physiological or pharmacological conditions , what transport processes might be taking place within the nephron in order to produce the urine that we observe . Key model predictions are summarized below: The present model simulates solute and water transport along the nephron of a healthy adult . The model can be used to simulate the nephron of a diabetic patient , if model parameters are appropriately adjusted to capture pathophysiological changes in hemodynamics ( to represent diabetes-induced glomerular hyperfiltration ) , anatomic ( tubular hypertrophy ) , transport and other relevant model parameters ( see Table 2 in Ref . [9] or Na+ transport expression changes in a diabetic rat ) . It must be acknowledged that quantifying these pathophysiological changes is a challenging task . Nonetheless , such a model . if successfully formulated , can be used to assess the actions of SGLT2 inhibitors in a diabetic kidney , as was done in a rat [9 , 42] . By further adjusting model parameters to simulate a remnant kidney ( using our previous approach for the rat [12 , 43] ) , we can simulate the administration of SGLT2 inhibitors to a kidney with diabetic nephropathy . To the extent that renal fluid and Na+ excretion can determine blood pressure and heart failure , model results can be used to assess the degree to which cardiovascular benefits SGLT2 inhibitors persist in patients with reduced GFR , and why . To directly predict blood pressure , however , one would need a more comprehensive model such as Ref . [44] . By adjusting SNGFR ( to represent changes in renal blood flow regulation ) and key transport activity levels ( e . g . , NHE3 [45] ) , one can simulate a hypertensive kidney and assess the relative effectiveness of diuretics , and why . A major motivation for developing a computational model of the human nephron is to to provide a platform for pharmacokinetics and pharmacodynamics simulations; that platform can be applied to predict the effects of a new drug , or to explain the underlying mechanisms of observed effects . In this regard , it is noteworthy that important species differences have been reported in relation to the expression of various membrane transporters that mediate transport of organic anions and cations in the mammalian kidneys and other organs [46] . ( Organic anions and cations are not represented in the present model . ) A computational model that includes the correct expression of the relevant organic anions/cations ( in humans ) would prove useful in translating drug test results in rodents to humans . The present nephron model predicts cellular solute concentrations , tubular flow , and luminal fluid solute concentrations . Except for the proximal tubule , nephron segments are represented as rigid tubules , and cell volume regulation [3] is not represented . Also , SNGFR is assumed known a priori . To model autoregulation , SNGFR can be set as a function of downstream tubular flow composition [47–51] . Interstitial fluid composition is assumed known a priori ( Table 2 ) . Additionally , the model does not represent the vasculature . As a result , the model does not represent the interactions among the nephron segments , or the interactions between nephrons and the renal vasculature . To properly simulate renal handling of a given drug compound , a model that represents the interactions among renal tubule and vessels is needed . Such a model can be formulated by embedding the nephron model into a human renal medullary model ( e . g . , Ref . [52–54] ) , similar to the recent rat kidney model by Weinstein [55] . Indeed , the present model can be used as an essential component in an integrated model of kidney function in humans for studying clinically relevant questions such as K+-induced natriuresis [55] .
|
In addition to its well-known function of waste removal from the body , the kidney is also responsible for the critical regulation of the body’s salt , potassium , acid content , and blood pressure . The kidneys perform these life-sustaining task by filtering and returning to blood stream about 200 quarts of blood every 24 hours . What isn’t returned to blood stream is excreted as urine . The production of urine involves highly complex steps of secretion and reabsorption . To study these processes without employing invasive experimental procedures , we developed the first computational model of the human nephron ( which is the functional unit of a kidney ) . The model contains detailed representation of the transport processes that take place in the epithelial cells that form the walls of the nephron . Using that model , we conducted simulations to predict how much filtered solutes and and water is transported along each individual and functionally distinct nephron segment . We conducted these simulations under normal physiological conditions , and under pharmacological conditions . The nephron model can be used as an essential component in an integrated model of kidney function in humans .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"classical",
"mechanics",
"chemical",
"compounds",
"fluid",
"mechanics",
"vertebrates",
"carbohydrates",
"plant",
"physiology",
"organic",
"compounds",
"glucose",
"animals",
"mammals",
"urine",
"animal",
"models",
"physiological",
"processes",
"plant",
"science",
"solute",
"transport",
"model",
"organisms",
"experimental",
"organism",
"systems",
"nephrons",
"amniotes",
"kidneys",
"research",
"and",
"analysis",
"methods",
"animal",
"studies",
"chemistry",
"fluid",
"dynamics",
"continuum",
"mechanics",
"fluid",
"flow",
"physics",
"rodents",
"eukaryota",
"anatomy",
"organic",
"chemistry",
"physiology",
"monosaccharides",
"biology",
"and",
"life",
"sciences",
"renal",
"system",
"physical",
"sciences",
"excretion",
"organisms",
"rats"
] |
2019
|
A computational model of epithelial solute and water transport along a human nephron
|
Organisms are continuously exposed to a myriad of environmental stresses . Central to an organism's survival is the ability to mount a robust transcriptional response to the imposed stress . An emerging mechanism of transcriptional control involves dynamic changes in chromatin structure . Alterations in chromatin structure are brought about by a number of different mechanisms , including chromatin modifications , which covalently modify histone proteins; incorporation of histone variants; and chromatin remodeling , which utilizes ATP hydrolysis to alter histone-DNA contacts . While considerable insight into the mechanisms of chromatin remodeling has been gained , the biological role of chromatin remodeling complexes beyond their function as regulators of cellular differentiation and development has remained poorly understood . Here , we provide genetic , biochemical , and biological evidence for the critical role of chromatin remodeling in mediating plant defense against specific biotic stresses . We found that the Arabidopsis SWI/SNF class chromatin remodeling ATPase SPLAYED ( SYD ) is required for the expression of selected genes downstream of the jasmonate ( JA ) and ethylene ( ET ) signaling pathways . SYD is also directly recruited to the promoters of several of these genes . Furthermore , we show that SYD is required for resistance against the necrotrophic pathogen Botrytis cinerea but not the biotrophic pathogen Pseudomonas syringae . These findings demonstrate not only that chromatin remodeling is required for selective pathogen resistance , but also that chromatin remodelers such as SYD can regulate specific pathways within biotic stress signaling networks .
In eukaryotic organisms genomic DNA is packaged into chromatin , which can repress transcription by blocking the access of regulatory proteins to DNA . Dynamic changes in chromatin structure are now recognized as a robust mechanism of transcriptional control [1]–[3] . Changes in chromatin structure are brought about by a number of different mechanisms including: chromatin modifications , which covalently modify histone proteins; incorporation of histone variants; and chromatin remodeling , which utilizes ATP hydrolysis to alter histone-DNA contacts [1] , [3]–[5] . ATP-dependent chromatin remodeling complexes are present in all eukaryotic organisms and can be grouped into three main classes: the SWI/SNF ATPases , the imitation switch ( ISWI ) ATPases , and the chromodomain and helicase-like domain ( CHD ) ATPases [2] , [3] . Significant advances have been made in understanding the mechanism of ATP-dependent chromatin remodeling complex action [1] , [4] . However , the biological role of chromatin remodeling complexes remains poorly understood , particularly in multicellular organisms where null mutations tend to be lethal [3] , [6] . Studies that have investigated the biological role of chromatin remodeling complexes in multicellular organisms have largely focused on their role as regulators of cellular differentiation and development [2] , [3] . In particular , Arabidopsis has served as a valuable model due to the fact that mutants in genes encoding a number of chromatin remodeling complex proteins are viable . One of the most well characterized chromatin remodeling complex proteins in Arabidopsis is the SWI/SNF class chromatin remodeling ATPase SPLAYED ( SYD ) . Loss of SYD activity causes defects in many different developmental pathways including stem cell maintenance , patterning , developmental transitions and growth [3] , [7]–[9] . The biological role of altering chromatin structure in response to stress via chromatin modifications and incorporation of histone variants has been documented [10]–[14] . However , the biological role of chromatin remodeling complexes or their specificity remains poorly understood . The role of chromatin remodeling in response to stress has been best studied in yeast where it has been shown that chromatin remodeling complexes are required for stress tolerance and are recruited to specific promoters upon stress [15]–[19] . However , few studies performed in multicellular organisms have investigated the role of chromatin remodeling in mediating stress responses . One study conducted in the human cell culture line SW480 demonstrated that chromatin remodeling complexes are recruited to specific promoters upon oxidative stress , which suggests that chromatin remodeling plays a role in the stress tolerance of multicellular organisms [20] . Additionally , it is unknown in any eukaryotic organism whether reduced stress tolerance in chromatin remodeling mutants is stress specific or indicative of decreased overall fitness due to non-specific global mis-regulation of gene expression . In this study we examine the role chromatin remodeling plays in biotic stress responses . We found that SYD is required for expression of specific genes within biotic stress signaling networks . This requirement is likely both direct and indirect as SYD is recruited to the promoter of some , but not all , of the genes for which it is required for expression . We show that SYD is required for resistance against the necrotrophic pathogen B . cinerea but not the biotrophic pathogen P . syringae . These findings demonstrate not only that chromatin remodeling is required for selective pathogen resistance , but also that chromatin remodelers , such as SYD can regulate specific pathways within biotic stress signaling networks .
Our results show that ATP-dependent chromatin remodeling is required for expression of specific genes within stress signaling networks . Additionally , this requirement is likely both direct and indirect as the chromatin remodeling ATPase SYD binds several , but not all , of the stress responsive promoters examined in vivo . Loss of chromatin remodeling activity also results in increased susceptibility to B . cinerea but not P . syringae . These results provide biological evidence that chromatin remodeling complexes , which are evolutionarily conserved within eukaryotes , are required for stress tolerance not only in yeast but also multicellular organisms . Furthermore , the requirement of ATP-dependent chromatin remodeling complexes is pathogen-specific and not a result of a general reduction in fitness .
Arabidopsis thaliana plants were grown in a 16 h light/8 h dark photoperiod at 22°C; except plants for pathogen treatments , which were grown in a 12 h light/12 h dark photoperiod . Wounding was performed as previously described [21] . All experiments were performed on 4 to 5-wk-old plants , which exhibited no disease symptoms or insect herbivory prior to treatment . Detached leaf assays were performed using the B . cinerea isolates DN , Grape , B05 . 10 and 83-2 [38] . Arabidopsis leaves were inoculated with 5 µl of spores at a concentration of 50 , 000 spores/ml [38] , [39] . For P . syringae bacterial growth assays Arabidopsis leaves were inoculated with 2×104 CFU/ml P . syringae pv . tomato ( Pst ) DC3000 by hand injection . Total RNA from rosette leaves was isolated by TRIzol extraction ( Life Technologies , Grand Island , NY ) and treated with DNAaseI to control for DNA contamination . RNA was reverse transcribed using Superscript III ( Invitrogen , Carlsbad , California ) . PCR for RT-PCR were conducted in 25 µl reactions containing 20 ng cDNA , 1 . 5 mM MgCl2 , 0 . 2 mM each dNTP , 0 . 05 µM each primer , and 1 U Choice-Taq Blue ( Denville Scientific , Metuchen , NJ ) and amplified for 29 cycles except for PDF1 . 2a in Figure 1B and ERF1 in Figure 2A , which were amplified for 34 cycles . Quantitative RT-PCR was conducted in 50 µl reactions containing 10 ng cDNA , 1× iQ SYBR Green supermix ( Bio-Rad Laboratories , Hercules , CA ) , and 200 or 250 nM each primer . Amplification and data analysis were carried out as previously described [21] . The internal controls At4g34270 and At4g26410 previously described were used for transcript normalization [40] . Primers are listed in Table S2 . Extraction of JAs ( MeJA and JA ) were carried out as previously described [41] and further analyzed by GC-MS using a Hewlett and Packard 6890 series gas chromatograph coupled to an Agilent Technologies 5973 network mass selective detector operated in electronic ionization ( EI ) mode . Camalexin and glucosinolates were measured 72 h after mock or B . cinerea inoculation as previously described [42] . Briefly , individual leaves were collected into deep 96-well plates containing 0 . 5 ml 90% methanol in each well . Following tissue disruption and centrifugation , 150 µl of leaf extract was removed for camalexin measurement . De-sulfo glucosinolates were extracted from an additional 150 µl of the same sample by passing the methanolic extract over a column of DEAE Sephadex A-25 ( Sigma-Aldrich ) and , after methanol and water washes , incubating the samples overnight with an excess of sulfatase before eluting with 150 µl H20 . Extractions were performed largely as previously described , but using centrifugation rather than vacuum to remove liquid from the Sephadex columns [43] . Separation of 50 µl of aqueous extracts was performed on a 5-µm column ( Lichrocart 250-4 RP18e , Hewlett-Packard , Waldbronn , Germany ) attached to a Hewlett-Packard 1100 series HPLC , using the following series of solvent gradients: 6-min 1 . 5% to 5 . 0% ( v/v ) acetonitrile , 2-min 5% to 7% ( v/v ) acetonitrile , 7-min 7% to 25% ( v/v ) acetonitrile , 2-min gradient from 25% to 92% ( v/v ) acetonitrile , 6 min at 92% ( v/v ) acetonitrile , 1-min 92% to 1 . 5% ( v/v ) acetonitrile , and a final 5 min at 1 . 5% ( v/v ) acetonitrile . Compounds were detected at 229 nm using a diode array detector , identified by comparison with retention time and absorption spectra of purified references , and quantified using response factors as previously published ( Table S1 ) [44] , [45] . ChIP-qPCR assays were performed as previously described [8] with the following modifications . Each ChIP was conducted using 500 mg of Ler rosette leaf tissue . DNA was sonicated to a size range of 0 . 3–1 . 5 kb . For the IgG control ChIP 2 µg of IgG from rabbit serum ( Sigma , St . Louis , MO ) was used . Following reverse cross-linking of the immunoprecipitation reactions the samples were treated with RNase A solution ( CalBiochem , La Jolla , CA ) and Proteinase K ( Sigma , St . Louis , MO ) . qPCR of the ChIP eluates was performed with iQ SYBR Green supermix according to manufacturer . ChIP-qPCR results were calculated based on the ΔΔCt method using the SuperArray ChIP-qPCR Data Analysis Template ( Frederick , MD ) according to the SuperArray manual , as described [46] . Briefly , ChIP DNA fractions were first normalized to input DNA ( ΔCt ) to account for chromatin sample preparation differences . Input normalized SYD and POLII ChIP fractions were then adjusted for the normalized non-specific background ( IgG ) giving the ΔΔCt value . Fold differences relative to the IgG reference were then calculated by raising 2 to the ΔΔCt power . The primers used in this study are listed in Table S2 . To determine statistical significance of treatment effects comparing WT versus syd t-tests were performed using Sigma Stat v3 . 5 ( San Jose , CA ) . For comparison of WT versus myc2/jin1 factorial ANOVA performed within SAS ( Cary , NC ) was used to analyze the effects of genotype and treatment on measured phenotypes , with significance of differences determined via t-tests of pre-selected comparisons . PR1: At2g14610 , PDF1 . 2a: At5g44420 , UBQ10: At4g05320 , ERF1: At3g23240 , PAD4: At3g52430 , ICS1: At1g74710 , NPR1: At1g64280 , WRKY70: At3g56400 , ERF#18: At1g74930 , CAF1-like: At3g44260 , AOS: At5g42650 , ERF2: At5g47220 , MYC2: At1g32640 , VSP2: At5t24770 , Oleo1: At4g25140 , AT2S3: At4g27160
|
In eukaryotes , genomic DNA is organized into a complex DNA-protein structure termed chromatin . The organization of chromatin serves to compact DNA within the nucleus and plays a central role in regulating transcription by controlling the access of DNA to transcriptional machinery . Chromatin structure can be altered through several mechanisms , one of which is chromatin remodeling , a process that disrupts DNA–protein interactions resulting in altered accessibility of specific DNA regions to regulatory proteins in the transcriptional machinery . In this study , we investigated the biological role of chromatin remodeling in defense responses to biotic stresses using the model plant Arabidopsis . We found that a chromatin remodeling protein , SPLAYED , is required for gene expression within specific biotic stress signaling networks . Consistent with this observation , loss of SPLAYED chromatin-remodeling activity resulted in increased susceptibility to a fungal pathogen , Botrytis cinerea , but not to a bacterial pathogen , Pseudomonas syringae . These results demonstrate that reduced stress tolerance in a chromatin-remodeling mutant plant can be stress specific , and is not simply due to a decrease in overall fitness as a result of non-specific global mis-regulation of gene expression .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"plant",
"biology/plant-biotic",
"interactions",
"plant",
"biology/plant",
"genetics",
"and",
"gene",
"expression",
"molecular",
"biology/chromatin",
"structure"
] |
2008
|
The Chromatin Remodeler SPLAYED Regulates Specific Stress Signaling Pathways
|
Both brucellosis and tuberculosis are chronic-debilitating systemic granulomatous diseases with a high incidence in many countries in Africa , Central and South America , the Middle East and the Indian subcontinent . Certain focal complications of brucellosis and extrapulmonary tuberculosis are very difficult to differentiate clinically , biologically and radiologically . As the conventional microbiological methods for the diagnosis of the two diseases have many limitations , as well as being time-consuming , multiplex real time PCR ( M RT-PCR ) could be a promising and practical approach to hasten the differential diagnosis and improve prognosis . We designed a SYBR Green single-tube multiplex real-time PCR protocol targeting bcsp31 and the IS711 sequence detecting all pathogenic species and biovars of Brucella genus , the IS6110 sequence detecting Mycobacterium genus , and the intergenic region senX3-regX3 specifically detecting Mycobacterium tuberculosis complex . The diagnostic yield of the M RT-PCR with the three pairs of resultant amplicons was then analyzed in 91 clinical samples corresponding to 30 patients with focal complications of brucellosis , 24 patients with extrapulmonary tuberculosis , and 36 patients ( Control Group ) with different infectious , autoimmune or neoplastic diseases . Thirty-five patients had vertebral osteomyelitis , 21 subacute or chronic meningitis or meningoencephalitis , 13 liver or splenic abscess , eight orchiepididymitis , seven subacute or chronic arthritis , and the remaining seven samples were from different locations . Of the three pairs of amplicons ( senX3-regX3+ bcsp3 , senX3-regX3+ IS711 and IS6110+ IS711 ) only senX3-regX3+ IS711 was 100% specific for both the Brucella genus and M . tuberculosis complex . For all the clinical samples studied , the overall sensitivity , specificity , and positive and negative predictive values of the M RT-PCR assay were 89 . 1% , 100% , 85 . 7% and 100% , respectively , with an accuracy of 93 . 4% , ( 95% CI , 88 . 3—96 . 5% ) . In this study , a M RT-PCR strategy with species-specific primers based on senX3-regX3+IS711 sequences proved to be a sensitive and specific test , useful for the highly efficient detection of M . tuberculosis and Brucella spp in very different clinical samples . It thus represents an advance in the differential diagnosis between some forms of extrapulmonary tuberculosis and focal complications of brucellosis .
Brucellosis remains one of the most widespread anthropozoonoses in the world , especially in the Mediterranean basin , the Middle East , India , Mexico and some countries of Central and South America [1] . Much evidence supports the conclusion that in countries without strong health systems , official data likely underestimate the true burden [2] . The high morbidity associated with brucellosis , together with its prolonged course and great tendency to produce relapses account for an important consumption of health care resources [3] , [4] . The global burden of tuberculosis ( TBC ) remains enormous [5] . Recent data in the WHO Global Tuberculosis Report 2012 confirm that TBC remains a major infectious killer today . In 2011 , there were an estimated 8 . 7 million new cases and 1 . 4 million people died from TBC [6] . Like TBC , brucellosis can cause focal complications in any organ or system . The larger studies place the rate of focal complications of brucellosis at around 25–35% of all cases [3] , [7] , [8] , similar to the rate of extrapulmonary complications in TBC , 15–40% [9] . Moreover , whilst in many countries there has been a reduction in the overall incidence of pulmonary tuberculosis , the number of extrapulmonary tuberculosis cases has increased in some industrialized countries [10]–[13] . When tuberculosis or brucellosis affect specific sites , e . g . , the CNS , or osteoarticular or genitourinary systems , the differential diagnosis between the two entities is virtually impossible based solely on clinical , haematological , biochemical or imaging studies . Furthermore , as both tuberculosis and brucellosis are granulomatous diseases , the pathological findings of focal complications of brucellosis and extrapulmonary tuberculosis can be very similar . Both Mycobacterium tuberculosis complex ( MTC ) and Brucella spp are slow-growing microorganisms . Classical methods for determining the presence of these microorganisms are time-consuming and labor-intensive . Hence , molecular methods , which offer speed , sensitivity and specificity , have been developed to address this problem . Multiplex real time PCR ( M RT-PCR ) is increasingly used in various fields of microbiology for the rapid differentiation of microbial species involved in specific syndromes [14]–[16] . Our group has shown that M RT-PCR is a useful strategy for the rapid differential diagnosis between extrapulmonary tuberculosis and brucellosis when they affect specific locations [17] . Later , we simplified the technique to make it more accessible to any clinical laboratory [18] . This study compared experimentally , in both monoplex and multiplex forms , the PCR combinations of three different targets for each microorganism , optimizing and simplifying the technique using SYBR Green , determining the sensitivity and reproducibility in a small sample of patients . The aim of the present study was to analyze comparatively the diagnostic yield of different strategies of M RT-PCR in a very representative sample of patients with focal complications of brucellosis or extrapulmonary tuberculosis and assessed the analytical specificity against a wide panel of microorganisms that included most of the non-tuberculous Mycobacteria related with human diseases , and the most important species and biovars of Brucella .
The study included 91 non-respiratory samples from 90 patients aged >14 years . Of the 90 patients , 30 had focal complications of brucellosis , 24 had extrapulmonary tuberculosis and 36 ( Control Group ) had various different infectious , autoimmune or neoplastic diseases in which the treating physician initially raised the possibility of extrapulmonary tuberculosis or brucellosis in the differential diagnosis . One patient with brucellosis provided two different samples from two simultaneous focal complications . The aims of the study were communicated to the participants and a written informed consent form was signed before the inclusion to the study . Whenever the subjects were minors the informed consent was given by the parents or legal guardians as appropriate in each case . The use of samples for research was approved by the Ethics Committees of Malaga University and Carlos Haya University Hospital , Malaga , Spain . The diagnosis of tuberculosis was established according to one of the following criteria: first , isolation of M . tuberculosis or second , the presence of caseating granulomas , with or without acid-fast bacilli in a patient with a compatible clinical picture and good therapeutic response to antituberculous treatment . The diagnosis of brucellosis was based on isolation of Brucella spp . in blood or any other body fluid or tissue sample or , second , the presence of a compatible clinical picture together with the demonstration of specific antibodies at significant titers or seroconversion . Significant titers were considered to be a standard agglutination test ( SAT ) ≥1/160 or immunocapture agglutination test ≥1/320 . The specificity of the M RT-PCR assays was assessed from a widely representative panel of the various biovarieties of Brucella spp and phylogenetically or serologically related microorganisms , species belonging to MTC and strains of non-tuberculous mycobacteria ( NTM ) from the collection and clinical isolates of the Microbiology Laboratory at Carlos Haya University Hospital ( HCH ) . The MTC strains were selected from the American Type Culture Collection ( ATCC ) and clinical samples from the Microbiology Laboratory at HCH . All the isolates of the clinical samples were later characterized at the Mycobacterium reference laboratory , at the Instituto de Salud Carlos III , Madrid , Spain . The different strains of NTM and Nocardia spp were supplied by the Spanish Type Culture Collection ( CECT ) or were clinical isolates from HCH . The strains of mycobacteria were cultured in Lowenstein-Jensen medium and incubated at 37°C for 2–4 weeks , in order to obtain sufficient bacterial growth for the later extraction of genomic DNA . The strains of Brucella spp were provided by the Department of Microbiology , of the Faculty of Medicine , University of Valladolid ( Spain ) , except for the vaccine strains B-19 and Rev 1 , which were supplied by the Andalusian Government Ministry of Agriculture and Fisheries . These strains were cultured in Brucella agar and incubated at 37°C in an atmosphere containing 5% CO2 for 48 hours . Genomic DNA from bacterial strains serologically or phylogenetically related with Brucella were provided by the CECT , except for the species Ochrobactrum intermedium , kindly provided by the Faculty of Medicine of the University of Navarra , Spain . All procedures were performed in a biosafety cabinet class II B3 . All samples destined for M RT-PCR were maintained at −20°C until processing . The amount or volume used varied depending on the type of sample . To monitor contamination , negative controls were included during each DNA extraction procedure . DNA was extracted using the Quiamp DNA Mini ( Qiagen , UK ) . Prior to DNA extraction , homogenized samples from the different tissues , CSF , synovial fluid , urine , purulent collections and strains were resuspended in 1 ml of molecular biology water , mixed and centrifuged at 15 . 000× g for 10 min . The supernatant was discarded and the pellet was resuspended with the volume of buffer outlined in the manufacturer's instructions . DNA pellets were resuspended in 50 µl molecular biology water and stored at 4°C until use . The concentration and purity of DNA were estimated by measuring the absorbance at 260 and 280 nm with a ND-1000 spectrophotometer ( Nanodrop ThermoFisher , USA ) . For detection of members of MTC , the primer sets IS6110f/IS6110r ( 5′ TCAAGGAGCACATCAGCC3′/5′TCACGGTTCAGGGTTAGC3′ ) and M1f/M3r ( 5′CGGCTAATCACGACGGCAC3′/5′CTCTTCCTCTCGTTGTGACCTGTT 3′ ) were used to amplify 82 and 164 bp fragments of IS6110 and senX3-regX3 , respectively . For Brucella , fragments of 152 and 142 bp of the bcsp31 gene and IS711 were amplified using primers bcsp31f/bcsp31r ( 5′ GCATTCTTCACATCCAGG 3′/5′ CACCGCATTCCATTATTCT 3′ ) and IS711f/IS711r ( 5′ TACAAGGAACGCCATCAGA 3′/5′ GCATTCAACGCAACCAGA ) [18] . The three real time reactions were monitored using a Light-Cycler 2 . 0 ( Roche Diagnostic , Indianapolis , IN ) with the LC FastStart DNA Master SYBR-Green I kit ( Roche Molecular Biochemicals , Mannheim , Germany ) . The M RT-PCRs for MTC and Brucella were performed as described previously [18] . Briefly , the mixture included 1× master mix , 3–3 . 5 mM MgCl2 , 0 . 5 µM primers , variable concentrations of DNA as template ( 150–250 ng depending on type of sample analyzed ) and nuclease free dH2O adjusted to a final volume of 20 µl . Each run included positive controls consisting of dilutions of Brucella spp and MTC DNA , and negative controls with all the elements of the reaction mixture except template DNA . The reactions were cycled 45 times , after an initial hold at 95°C for 10 min , between 95°C for 10 s , 60°C for 5 s , and 72°C for 6 s with programmed transitions of 20°C/s . The melting curves were acquired on the SYBR channel by heating momentarily at 95°C , cooling to 65°C and collecting fluorescence continuously at a ramping rate of 0 . 1°C/s until 95°C . To minimize experimental variability the Ct values , the threshold cycle where the fluorescence signal rises significantly above background in the exponential phase of the amplification , were determined by the second derivative maximum method . In order to avoid potential observer bias , the clinical and microbiological diagnoses of the patients were unknown to the technician who performed the M RT-PCR assay . The specificity of the primers was first tested in silico using the BLASTn program in order to prevent non-specific amplifications . The analytical specificity was then tested against the 59 microorganisms listed in Table 1 . To confirm the identities of the amplified fragments , some of the strains used as positive controls of Brucella spp and MTC and different clinical samples were sequenced . The ABI PRISM Big Dye Terminator Cycle sequencing reaction kit v . 3 . 0 ( Applied Biosystems , Madrid , Spain ) was used for the sequencing analysis , by capillary electrophoresis , in an ABI PRISM , model 3100 automated sequencer ( Applied Biosystems ) . Quantitative variables are represented as mean ± standard deviation and qualitative variables as percentages . Sensitivity , specificity , positive and negative predictive values , accuracy , likelihood ratios ( LR ) and 95% confidence intervals ( CI ) were calculated using the Twobytwo 1 . 0 analyzer program . Brucella: bcsp31 ( M20404 ) , IS711 ( AE017223 ) MTC: IS6110 ( BX842574 ) , senX3-regX3 ( BX842573 ) .
The three M RT-PCR strategies based on amplification of the target sequences senX3-regX3+ bcsp31 , senX3-regX3+ IS711 and IS6110+ IS711 correctly identified all the species and biovars of Brucella as well as all the species belonging to MTC . The target based on the IS711 sequence did not amplify any of the bacteria serologically or phylogenetically related with Brucella spp , and the amplicons of the genes of the bcsp31 protein gave a false positive result with Ochrobactrum anthropi and intermedium . Likewise , the amplicons of the intergenic region senX3-regX3 were negative in all NTM tested , but those of the IS6110 sequence amplified various NTM: M . fortuitum , M scrofulaceum , M simiae and M . intracellulare ( Table 1 ) . Figure 1 shows the similar melting temperatures ( Tm ) of two of these NTM ( as an example ) and the two strains of Ochrobactrum for bcsp31 compared with the control Tm for the two pathogens . Of the 91 clinical samples included in the study , 35 were vertebral or paravertebral tissue from patients with vertebral osteomyelitis , 21 were CSF from patients with subacute or chronic meningitis or meningoencephalitis , 13 were tissue or abscess aspirates from patients with liver or splenic abscess , 8 were urine samples from patients with orchiepididymitis , 7 were from synovial fluid from patients with subacute or chronic arthritis , and the remaining 7 were from different locations: two samples of purulent fluid from patients with neck abscesses , two bone biopsies from patients with osteomyelitis of the femur and sternum , respectively , two kidney biopsies from patients with chronic pyelonephritis , and a sample of seminal fluid from a patient with chronic orchiepididymitis . Table 2 summarizes the sample type and the final diagnosis of the study patients . Of the 30 patients with brucellosis , 25 ( 83 . 3% ) were primary infections and 5 ( 16 . 6% ) had had a previous episode of infection . Brucella melitensis was isolated in 17 ( 56 . 6% ) of the 30 patients with brucellosis; 13 ( 43 . 3% ) in blood culture , 8 ( 26 . 6% ) in non-blood samples ( three vertebral tissue and one each of the following: urine , CSF , hepatic tissue , synovial fluid and thyroid abscess ) and in 4 ( 13 . 3% ) in both blood and non-blood samples . In 12 of the other 13 patients ( 40% ) the diagnosis of brucellosis was based on clinical and serological criteria . One 43-year-old woman , who habitually consumed non-homogenized dairy products and who had osteomyelitis with thoracic segment involvement and whose biopsy showed non-caseating granulomas , constantly had negative cultures and absence of serological response , and was diagnosed with brucellosis based on her epidemiologic exposure and clear response to treatment with doxycycline plus streptomycin . Of the 24 patients with extrapulmonary tuberculosis , M . tuberculosis was isolated in 18 ( 75% ) and the other six ( 25% ) had necrotizing granulomas in their biopsies , with or without acid-fast bacilli . Only four ( 16 . 6% ) of the 24 cases had smear-positive samples . The three M RT-PCR strategies were positive in 49 ( 89 . 1% ) of the 55 samples from patients with tuberculosis or brucellosis; 28 ( 90 . 3% ) of the 31 focal complications of brucellosis and 21 ( 87 . 5% ) of the 24 extrapulmonary tuberculosis . M RT-PCR was negative in the 36 samples from the control group patients . Thus , the overall sensitivity of the M RT-PCR was 89 . 1% , ( 95% CI , 80 . 9–97 . 3 ) and the specificity 100% . The overall diagnostic yield of the M RT-PCR is shown in Table 3 . Of the six patients who had a false negative result with the M RT-PCR , two had received prolonged antimicrobial treatment before drawing the sample . The first of these was a 24-year-old woman with a kidney transplant and brucellar pyelonephritis in the transplanted organ , treated for two weeks before taking the renal biopsy with ciprofloxacin , meropenem and piperacillin-tazobactam . The second was a 33-year-old man with tuberculous vertebral osteomyelitis treated during the four months prior to taking the vertebral biopsy with rifampicin/isoniazid/pyrazinamide/ethambutol for the first two months and with rifampicin/isoniazid the second two months . In both cases the cultures were also negative . If these cases had been withdrawn from the analysis of efficacy , the sensitivity of the M RT-PCR for the overall sample would have risen to 92 . 5% . The other four false-negative results corresponded to two patients with brucellosis ( one brucellar orchiepididymitis with positive blood cultures and a negative urine culture and the other vertebral osteomyelitis with negative blood and vertebral tissue cultures ) and two patients with tuberculosis ( one meningitis and one vertebral osteomyelitis , both with positive cultures and negative microscopic study ) . Table 4 shows the results of the M RT-PCR according to the type of microorganism , culture result and sample type . The M RT-PCR was positive in the four cases of extrapulmonary tuberculosis with smear-positive samples and in 17 ( 85% ) of the 20 cases with smear-negative samples . The mean Ct values of the senx3-regx3+ bcsp31 , senx3-regx3+ IS711 e IS6110+ IS711 assays varied according to the type of sample , ranging from 31 . 03–34 . 85 , 26 . 99–33 . 00 and 29 . 95–34 . 74 cycles respectively for the samples from patients with extrapulmonary tuberculosis to 24 . 68–30 . 57 , 16 . 13–31 . 73 and 28 . 17–32 . 07 cycles for the samples from patients with focal complications of brucellosis . The amount and purity of total DNA ( microbial DNA and eukaryotic DNA ) differed significantly depending on the type of clinical sample studied ( Figure 2 ) , though this did not affect the percentage of positive results with the test , independently of the M RT-PCR strategy used ( Table 5 ) . Finally , the ranges of the differences between the Tm of the PCR products of the clinical samples and the control strains were 0 . 02–0 . 63°C; a figure we consider irrelevant , and which is very expressive of the specificity of the technique ( Table 6 ) .
Since it was demonstrated that PCR can simultaneously amplify multiple loci of one or more different genes , multiplex PCR has become firmly established as a general technique [19] . As procedures become cheaper and simpler , molecular technology is being increasingly used in rapid microbiological diagnosis . Because of its high sensitivity , molecular diagnosis has now become a very useful tool for the diagnosis of many viral , bacterial and fungal infections . Clinical microbiology is now directed more towards syndromic diagnosis , in which the most common causative agents of a particular clinical syndrome are all studied together at the same time in a single test . As M RT-PCR can do this ever more efficiently , it has experienced exponential development in recent years [14]–[16] , [20] , [21] . In many underdeveloped and developing countries , tuberculosis and brucellosis are still the most frequent causes of bacterial lymphocytic meningitis , granulomatous vertebral osteomyelitis , subacute arthritis and subacute orchiepididymitis . In these clinical scenarios , among others , M RT-PCR could be a useful tool for the rapid differential diagnosis between two pathogens whose isolation in culture is difficult and time consuming . Previous studies from our group have shown that of the different candidate genes , three combinations of amplicons of bcsp31 protein gene and the IS711 in the case of Brucella spp and the senX3-regX3 intergenic region and IS6110 for MTC permit a highly sensitive and reproducible co-amplification [18] . In this study we analyzed the diagnostic yield of the three possible combinations of the amplicons ( senX3-regX3+ bcsp31 , senX3-regX3+ IS711 and IS6110+ IS711 ) in a representative sample of patients with extrapulmonary tuberculosis and focal complications of brucellosis . The three primer combinations correctly identified all the species and biovarieties of Brucella and MTC , and there was no non-specificity with the strategy based on the sequence amplification of senX3-regX3+ IS711 . The target based on bcsp31 did , however , show a false positive result with Ochrobactrum spp . This cross-reaction , which has been described previously [22]–[23] , is not surprising if we consider that Ochrobactrum spp . is the closest known relative of the Brucella genus . Concerning MTC , the target senX3-regX3 showed no non-specificity with the panel of NTM , though the strategy based on IS6110 produced a cross-reaction with M . fortuitum , M . scrofulaceum , M . intracellulare and M . simiae . This lack of specificity has been previously described . Thus , a study analyzing the specificity of IS6110-based methods in nine laboratories from France demonstrated false-positive reactions with an average rate of 7% , most of them caused by NTM [24] . This explains why many authors request caution in designing and evaluating diagnostic PCR tests based on this element [25] . The overall sensitivity of our M RT-PCR method should be considered very good since it was 89 . 1%; 87 . 5% in extrapulmonary TB cases and 90 . 3% in cases of focal complications of brucellosis . These results are as good as or better than those with any of the monoplex PCR methods so far tried , sensitivities of which have ranged from 53–95% in clinical samples from patients with extrapulmonary tuberculosis [26]–[30] and from 92–94% for non-blood samples of focal complications of brucellosis [31] . The yield of molecular diagnostic techniques falls in patients with extrapulmonary tuberculosis with respiratory or nonrespiratory smear-negative specimens [26] , [32] . In our study , only 4 ( 16 . 6% ) of the 24 extrapulmonary tuberculosis cases were smear-positive , a percentage similar to that reported by other authors [26] , [30] . This very small number of samples makes it difficult to draw conclusions about the sensitivity of our M RT-PCR assay in patients with extrapulmonary tuberculosis with smear-negative samples . Nevertheless , the results of this study ( 85% sensitivity in smear-negative samples ) show the high sensitivity of the technique , even in paucibacillary specimens . This high sensitivity in smear-negative samples may be related with the fact that in our study most were aspirates from abscesses or tissue samples . Recently Moure et al , in a large study including 108 smear-negative extrapulmonary samples , found that the sensitivity of the Xpert was just 40 . 5% in sterile fluids versus 76 . 5% in abscess aspirates [33] . The diagnosis of brucellosis does not normally present problems in acute non-complicated forms . In these cases , all the serological tests commonly used have a high sensitivity . However , this is not the case in patients who have a more prolonged evolution , as occurs in most patients who have focal complications , particularly if they are patients who are professionally exposed or patients with recurrences of the disease . In both scenarios , serological studies have important limitations [34] . In addition , the sensitivity of the cultures , whether they are from peripheral blood or non-blood samples , does not usually surpass 50% in patients with focal forms of brucellosis . Other than our own studies , reports dealing with the usefulness of molecular techniques for the diagnosis of patients with focal complications of brucellosis are anecdotal , though they all show the superiority of these techniques as compared to cultures [35] , [36] . In clinical practice the volume of a sample sent to the laboratory for the diagnosis of patients with extrapulmonary tuberculosis or focal brucellosis can vary greatly , depending on the site of the complication and the form of obtaining the sample . In fine-needle aspiration biopsies this volume can be really small . In our study the amount of DNA extracted and its purity can be considered good in all types of samples , except for CSF , as mentioned by others [37] . Concerning the amount and purity of DNA , previous studies by our group [18] have shown the inhibitory effect that high concentrations of DNA have on the technique . Given these previous results , in this study we used DNA amounts no greater than 250 ng per reaction , both in tissue samples and in abscesses . The small variable volumes of CSF available in clinical practice together with the peculiar characteristics of subacute lymphocytic bacterial meningitis ;mild or moderate pleocytosis , and paucibacillary samples meant that the volume of DNA for each assay varied , ranging between 2 and 8 µl for a final volume of 20 µl in the PCR reaction . As is logical , the purity of the DNA differed widely depending on the type and location of the study sample , though this did not greatly affect the Ct or the Tm in comparison with what was seen in the collection strains of the two pathogens . From a qualitative point of view , neither the type of sample nor the amount or purity of the DNA influenced significantly the diagnostic yield of the M RT-PCR , independently of the strategy used , indicating the robustness of the three SYBR Green based M RT-PCR strategies . Though the comparative study of the three pairs of amplicons used showed no differences in the samples used , the M RT-PCR strategy based on the amplification of senX3-regX3+ IS711 seems to be the most suitable , as it avoids false positive results derived not only from cross-reactions of IS6110 with NTM but also from amplification of Ochrobactrum spp . , as this microorganism lacks IS711 [38] . The Ochrobactrum spp . comprises a group of very ubiquitous microorganisms . Although its ecology is not well known , it has been isolated from soil , water , multiple hospital material , and different clinical specimens and it may be part of the normal flora of the large intestine . Ochrobactrum spp . would seem to occupy a microbial niche similar to that of Pseudomonas aeruginosa , as most infections in humans have been in patients with catheters , other foreign bodies , or severely immunosuppressed persons [39] . Indeed , it is always important to exclude possible cross-reactions with potentially colonizing microorganisms . In addition to its high sensitivity , other important aspects of single-tube M RT-PCR make it especially attractive to clinical laboratories for use in samples from patients in whom extrapulmonary TBC or focal complications of brucellosis are suspected . First , M RT-PCR provides results within 4 hours , which is much less than the time required for conventional methods to rescue a fastidious microorganism such as M . tuberculosis or Brucella spp; second , the technique almost completely obviates the need for direct handling of the pathogen , thus drastically reducing the risk of infection of laboratory personnel; and third , the sample can either be processed immediately or easily stored at −20°C until processing . In conclusion , a SYBR Green single-tube M RT-PCR assay based on senX3-regX3+ IS711 coamplification allows a rapid and efficient identification of M . tuberculosis complex and Brucella spp in different clinical samples . Based upon our own experience with M RT-PCR and those of other authors , this new strategy is more specific than those previously reported , which , together with its high sensitivity , make it a very useful tool for the differential diagnosis between some forms of extrapulmonary tuberculosis and focal complications of brucellosis .
|
Both brucellosis and tuberculosis are systemic infections which may involve any organ . When they affect specific locations , extrapulmonary tuberculosis and brucellosis cause symptoms that are very difficult to differentiate clinically . Mycobacterium tuberculosis complex and Brucella spp are slow-growing microorganisms whose culture and isolation require several days to weeks . Methods based on the polymerase chain reaction ( PCR ) have proven more sensitive than conventional culture , for both extrapulmonary tuberculosis and focal complications of brucellosis . Multiplex real time PCR is a variant of PCR in which two or more target sequences can be simultaneously amplified in a single tube . We developed and evaluated the results of several multiplex real-time PCR strategies for the rapid differential diagnosis between extrapulmonary tuberculosis and focal complications of brucellosis . Multiplex real-time PCR targeting of SenX3-RegX3+IS711 sequences showed a sensitivity of 89 . 1% and a specificity of 100% when applied to 91 clinical specimens . These findings provide solid evidence suggesting that multiplex real-time PCR could be a useful tool to reduce the time required for the differential diagnosis between extrapulmonary tuberculosis and complicated brucellosis , thereby improving prognosis .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2013
|
Comparative Clinical Study of Different Multiplex Real Time PCR Strategies for the Simultaneous Differential Diagnosis between Extrapulmonary Tuberculosis and Focal Complications of Brucellosis
|
Targeting of Toxoplasma gondii by autophagy is an effective mechanism by which host cells kill the protozoan . Thus , the parasite must avoid autophagic targeting to survive . Here we show that the mammalian cytoplasmic molecule Focal Adhesion Kinase ( FAK ) becomes activated during invasion of host cells . Activated FAK appears to accompany the formation of the moving junction ( as assessed by expression the parasite protein RON4 ) . FAK activation was inhibited by approaches that impaired β1 and β3 integrin signaling . FAK caused activation of Src that in turn mediated Epidermal Growth Factor Receptor ( EGFR ) phosphorylation at the unique Y845 residue . Expression of Src-resistant Y845F EGFR mutant markedly inhibited ROP16-independent activation of STAT3 in host cells . Activation of FAK , Y845 EGFR or STAT3 prevented activation of PKR and eIF2α , key stimulators of autophagy . Genetic or pharmacologic inhibition of FAK , Src , EGFR phosphorylation at Y845 , or STAT3 caused accumulation of the autophagy protein LC3 and LAMP-1 around the parasite and parasite killing dependent on autophagy proteins ( ULK1 and Beclin 1 ) and lysosomal enzymes . Parasite killing was inhibited by expression of dominant negative PKR . Thus , T . gondii activates a FAK→Src→Y845-EGFR→STAT3 signaling axis within mammalian cells , thereby enabling the parasite to survive by avoiding autophagic targeting through a mechanism likely dependent on preventing activation of PKR and eIF2α .
Toxoplasma gondii is an obligate intracellular protozoan that can cause disease in humans , including retinochoroiditis and encephalitis . T . gondii actively invades host cells , a process powered by the parasite’s own motility that is dependent on the sequential secretion of proteins present in the apical organelles called micronemes and rhoptries [1–3] . Micronemal proteins ( MIC ) act as adhesins that interact with the host cell membrane and also function through their association with the parasite glideosome that powers motility [2] . A complex of rhoptry neck proteins ( RON ) are secreted into the host cell membrane anchoring the parasite to the cell being invaded [1–3] . This complex contains RON2 that associates with the host cell membrane , plus RON4 , RON5 and RON8 that are exposed to the host cell cytoplasm [1–3] . The complex forms a structure called moving or tight junction through which the parasite penetrates the host cell causing invagination of the host cell membrane [1–3] . Once invasion is completed , T . gondii dissociates from the host cell membrane and resides within a specialized niche called parasitophorous vacuole ( PV ) . T . gondii cannot withstand the lysosomal environment . The PV enables parasite survival since it is devoid of proteins required for fusion with endosomes and lysosomes [4] . However , in addition to the classical endosomal-lysosomal pathway , macroautophagy ( commonly referred as autophagy ) is an important constitutive mechanism that targets organelles and portions of cytoplasm to lysosomal degradation [5] . This indicates that T . gondii must avoid autophagic targeting as a survival mechanism within host cells . Autophagy is characterized by the recruitment of Atg proteins to the isolation membrane that encircles a portion of the cell leading to the formation of an autophagosome [5] . The process is driven by the activation of the Unc-51-like kinase 1/2 ( ULK1/2 ) complex and Beclin 1 –Phosphatidylinositol 3-kinase catalytic subunit type 3 ( PI3KC3 ) complex , and is inhibited by activation of the mechanistic target of rapamycin complex 1 ( mTORC1 ) [6–8] . We previously demonstrated that T . gondii induces autophosphorylation of epidermal growth factor receptor ( EGFR ) in host cells , a process mediated by MIC3 and MIC6 , parasite proteins with EGF-like domains [9] . EGFR autophosphorylation is followed by activation of Akt ( a molecule known to inhibit autophagy by activating mTORC1 [10] ) and inhibition of targeting of the PV by autophagosomes [9] . However , autophagy is regulated at various levels by an array of signaling molecules . The efficiency by which T . gondii avoids autophagic targeting raised the possibility that the parasite acts at more than one level to successfully impair autophagic killing . Herein , we report that during the process of invasion by T . gondii , the mammalian cytoplasmic molecule Focal Adhesion Kinase ( FAK ) became activated at a site that appears to associate with RON4 ( host cell-parasite junction ) . FAK activation led to activation of Src and its transactivation of EGFR that , instead of inducing Akt activation , triggered Y705 phosphorylation and activation of Signal Transducer and Activator of Transcription 3 ( STAT3 ) . In turn , STAT3 signaling prevented autophagic killing of the parasite likely by averting activation of Protein Kinase Double-stranded RNA-dependent ( PKR ) and its downstream effector Eukaryotic Initiation Factor 2α ( eIF2α ) . Blockade of the FAK→Src→Y845-EGFR→STAT3 signaling cascade led to parasite killing in host cells that were not activated by immune modulators .
Src constitutes a signaling node that regulates multiple cellular processes [11] . Src activity is regulated by phosphorylation of tyrosine residues . Phosphorylation at Y416 in the activation loop in the kinase domain locks the catalytic domain into a fully active conformation resulting in high kinase activity [12] . Incubation of human retinal pigment epithelial cells ( RPE ) with tachyzoites of the RH strain of T . gondii ( type I strain ) resulted in an increased phosphorylation of Src at Y416 ( Fig 1A ) . Similar results were observed in human lung epithelial cells ( A549 ) and mouse endothelial cells ( mHEVc ) ( Fig 1A ) . T . gondii enhanced Y416 Src phosphorylation not only in non-hematopoietic cells but also in a mouse microglia cell line ( BV-2 ) ( Fig 1B ) . In addition , increased Y416 phosphorylation was detected in cells infected with tachyzoites of the ME49 ( type II strain ) or VAND ( atypical ) strain ( Fig 1C ) . Subsequent experiments were conducted using a type I strain ( RH ) unless otherwise stated . Taken together , different strains of T . gondii induce rapid activation of Src in various hematopoietic and non-hematopoietic cells . T . gondii activates EGFR in mammalian cells [9] and EGFR signaling can activate Src [13] . However , T . gondii infection of CHO cells ( EGFR null ) still triggered Y416 Src phosphorylation , indicating that parasite-induced Src activation can occur independently of EGFR ( S1A Fig ) . T . gondii is reported to induce G protein-coupled receptor ( GPCR ) , TLR2 and TLR4 signaling [14 , 15] , molecules that signal through Src [16–18] . However , while treatment of A549 cells with pertussis toxin ( PTx , an inhibitor of GPCR signaling ) markedly impairs Akt phosphorylation induced by Lysophosphatidic acid ( LPA , a ligand for GCPR ) , PTx did not inhibit parasite-induced phosphorylation of Src Y416 ( S1B Fig ) . Similarly , knockdown of MyD88 ( adaptor protein that mediates TLR2/4 signaling ) in A549 cells did not inhibit parasite-induced phosphorylation of Src Y416 or affect the ability of tachyzoites to infect mammalian cells ( S1C Fig ) . The parasite kinases , rhoptry protein 16 ( ROP16 ) and ROP18 , are delivered into the host cells and induce tyrosine phosphorylation of STAT3/6 and threonine phosphorylation of Immunity Related GTPases ( IRG ) , respectively in infected mammalian cells [19–22] . To examine the potential role of these proteins in Src activation , A549 cells were infected with tachyzoites deficient in these proteins ( Δrop16 or Δrop18 ) or their WT controls . Infection with Δrop16 or Δrop18 tachyzoites did not diminish Src Y416 phosphorylation ( S1D Fig ) . Altogether , T . gondii can trigger phosphorylation of Src Y416 independently of EGFR , GPCR and TLR signaling as well as independently of ROP16 and ROP18 . FAK is a widely expressed cytoplasmic non-receptor tyrosine kinase that activates Src [23] . Y397 is the major autophosphorylation site of FAK that correlates with increased catalytic activity and creates a high-affinity binding site for the SH2 domain of Src [24–26] . We examined whether T . gondii causes FAK phosphorylation at Y397 and whether FAK mediates parasite-induced Src activation . T . gondii induced rapid phosphorylation of Y397 in FAK in RPE , A549 , mHEVc and CHO cells ( Fig 2A–2F ) . The experiments with CHO cells indicate that , similar to Src , T . gondii can activate FAK in the absence of EGFR ( Fig 2B ) . Next , we used tetracycline-induced conditional micronemal protein 8 knockout parasites ( MIC8KOi ) to determine whether host cell invasion is required for phosphorylation of Y397 . Incubation with aminotetracycline ( ATc ) led to the phenotype of MIC8 deficiency—parasites with markedly impaired ability to infect host cells [27] ( A549 cells; no ATC: 28 . 3 ± 4 . 3% and 30 . 3 ± 5% of infected cells at 2 and 18 hr respectively; ATc: 3 . 3 ± 0 . 5% and 3 . 6 ± 0 . 6% at 2 and 18 hr respectively ) . In contrast to MIC8-sufficient parasites , tachyzoites deficient in MIC8 failed to induce phosphorylation of FAK Y397 ( Fig 2C ) . FAK bridges extracellular stimuli with Src signaling . Thus , we examined if phosphorylation of FAK occurred at the site of parasite-host cell interaction . CHO cells were challenged with T . gondii-YFP followed by staining with Abs against anti-phospho-Y397 FAK and RON4 , a central component of the moving junction . This structure manifests initially as a cap around the apical tip of the tachyzoite and subsequently , as a ring around the invading parasite . During early stages of parasite invasion , enhanced activation of FAK ( i . e . , increased Y397 phosphorylation ) appeared to associate with RON4 around the apical pole of parasites although the overlap may not be complete ( Fig 2D , early phase ) . Increased phospho-Y397 FAK expression also co-localized with ring structures that expressed RON4 ( Fig 2D , mid phase ) . Finally , increased expression of phospho-Y397 FAK was also noted around parasites that had completed invasion ( Fig 2D , intracellular ) . Altogether , our findings indicate that T . gondii induces FAK activation during the process of invasion of host cells . A549 cells were treated with vehicle or FAK inhibitor ( PF-573228 ) , followed by challenge with T . gondii to test whether FAK mediates T . gondii-induced Src activation . The FAK inhibitor markedly impaired parasite-induced phosphorylation of Src Y416 ( Fig 2E ) . This occurred despite the fact that the FAK inhibitor did not impair the ability of tachyzoites to infect cells . Similarly , knockdown of FAK markedly reduced Y416 phosphorylation of Src in mHEVc cells challenged with T . gondii ( Fig 2F ) . Thus , infection with T . gondii causes Y397 phosphorylation of FAK that appears to associate with RON4 , a component of the moving junction . In turn , FAK mediates parasite-induced Src activation . Integrin clustering causes recruitment of FAK to the intracytoplasmic tail of integrins leading to FAK autophosphorylation at Y397 and FAK activation [25 , 28] . We took 3 approaches to examine the role of β integrins in T . gondii-induced FAK activation . First , we incubated mammalian cells with cRGDfV , a peptide that binds αvβ integrins inhibiting their interaction to ligands [29] . cRADfV peptide was used as control . As shown in Fig 3A , cRGDfV peptide markedly diminished Y397 phosphorylation of FAK in RPE cells challenged with T . gondii . Next , we examined the effects of knockdown of β1 integrin or of treatment with a neutralizing anti-αvβ3 Ab . We used MDA-MB-231 cells transduced with lentiviral vector encoding shRNA against β1 integrin ( proven to be deficient in β1 integrin expression [30] ) and MB-MB-231 cells transduced with a lentiviral vector encoding control shRNA . MDA-MB-231 cells deficient in β1 integrin exhibited reduced FAK Y397 phosphorylation after T . gondii infection ( Fig 3B ) . Incubation with a neutralizing anti-αvβ3 Ab decreased parasite-induced FAK Y397 phosphorylation both in cells sufficient and deficient in β1 integrin ( Fig 3B ) . None of the approaches to inhibit β integrin signaling impaired the initial percentage of infected cells ( see S2C and S2D Fig ) . Thus , FAK activation triggered by T . gondii infection appears to be dependent on β1 and β3 integrins . We examined whether T . gondii activates Src and FAK signaling in order to survive within mammalian cells . RPE , mHEVc and BV-2 cells were incubated with vehicle or PP2 , a Src kinase inhibitor , followed by challenge with T . gondii . While PP2 did not impair the initial percentage of infected cells , PP2 significantly reduced the percentage of infected cells at 24 h , the number of tachyzoites per 100 cells and the numbers of T . gondii-containing vacuoles per 100 cells at 24 h ( S2A Fig ) . Similar results were observed after knockdown of Src ( Fig 4A ) . The T . gondii-containing vacuoles that persisted after Src knockdown had similar numbers of parasites as those from control siRNA-expressing cells indicating that the effect of Src knockdown was to induce parasite killing rather than restrict parasite replication ( Fig 4A ) . Knockdown of FAK in mHEVc cells not only induced killing of RH T . gondii but also of the VAND strain of the parasite ( S2B Fig ) . Similarly , incubation of RPE with cRGDfV peptide , knockdown of β1 integrin in MDA-MB-231 cells or incubation of these cells with a neutralizing anti-αvβ3 Ab decreased the percentages of infected cells at 24 h and the numbers of parasites per 100 host cells ( S2C and S2D Fig ) . Taken together , these findings revealed that parasite-induced β integrin→FAK→Src signaling prevents killing of T . gondii within host cells . We examined whether parasite killing after blockade of Src and its upstream inducers is dependent on autophagy . The distribution of LC3 , a protein associated with autophagosome membrane , was examined in T . gondii-infected cells rendered deficient in Src . A549 cells transfected with either control siRNA or Src siRNA followed by challenge with T . gondii and immunofluorescence using an anti-LC3 Ab . Knockdown of Src resulted in a significant accumulation of LC3 ( ring ) around the parasite ( Fig 4B ) . Moreover , Src knockdown caused accumulation of the late endosomal/lysosomal marker LAMP-1 around T . gondii ( Fig 4C ) . At later times post-infection ( 12 h ) , tachyzoites encircled by LAMP-1 exhibited morphology that suggested they were undergoing degradation ( Fig 4C ) . To investigate whether T . gondii killing during blockade of Src is dependent on the autophagy machinery , we first examined the effects of knockdown of ULK1 and Beclin 1 . ULK1 and the Beclin 1-PI3KC3 complex promote autophagosome formation and maturation [5] . Moreover , ULK1 is key for the stimulation of canonical autophagy in mammalian cells [6–8] . Knockdown of ULK1 or Beclin 1 prevented induction of anti-T . gondii activity in mammalian cells incubated with PP2 or transfected with Src siRNA ( Fig 4D and 4E ) . We examined the effects of the lysosomal protease inhibitors leupeptin and pepstatin since autophagosomes deliver their cargo to lysosomes for degradation . Treatment with lysosomal inhibitors also ablated anti-T . gondii activity in Src deficient cells ( Fig 4F ) . Similarly , anti-T . gondii activity induced in cells transduced with FAK shRNA-encoding lentiviral vector was ablated by knockdown of ULK1 or Beclin 1 ( Fig 4G and 4H ) . Finally , lysosomal inhibitors prevented anti-T . gondii activity induced by cRGDfV peptide ( Fig 4I ) . Thus , engagement of the β integrin→FAK→Src pathway is critical to promote parasite survival since it prevented T . gondii killing dependent on the autophagy machinery and lysosomal protease activity . Src interacts with growth factor receptors including EGFR , a surface molecule engaged by T . gondii during host cell invasion [9] . The parasite adhesins MIC3 and MIC6 contain EGF-like domains and cause EGFR signaling that activates Akt and impairs autophagic targeting of the parasitophorous vacuole [9] . Src binds EGFR and transactivates this receptor by causing EGFR phosphorylation at a unique Y845 residue leading to signaling via alternate pathways downstream of EGFR [31–34] . We examined whether T . gondii induces EGFR phosphorylation at Y845 . A549 cells expressing either control siRNA or Src siRNA were challenged with T . gondii . T . gondii induced EGFR Y845 phosphorylation , an effect that was ablated by knockdown of Src ( Fig 5A ) . Next , we examined the relevance of parasite-induced Y845 EGFR phosphorylation . EGFR with dialanine substitution of L679 and L680 ( EGFR AA ) is defective in Src-induced Y845 phosphorylation [35] . NMuMG epithelial cells with stable expression of human WT EGFR or EGFR AA were infected with T . gondii . A significant reduction in the percentages of infected cells and the numbers of parasites per 100 cells at 24 h was detected in cells expressing EGFR AA ( Fig 5B ) . To further explore the importance of Y845 EGFR phosphorylation , A549 cells were transfected with a plasmid encoding either WT EGFR or a phenylalanine substitution of Y845 ( Y845F EGFR ) [31 , 32] followed by challenge with RH T . gondii . Y845F mutant EGFR does not signal through Y845 phosphorylation but retains full EGFR kinase activity [31 , 32] . Cells transfected with Y845F-EGFR plasmid exhibited a significant decrease in the percentages of infected cells , the numbers of tachyzoites and vacuoles per 100 cells without affecting the number of tachyzoite per vacuole ( Fig 5C ) . These findings indicate that preventing Y845 EGFR phosphorylation results in killing of T . gondii rather than reduction in intracellular replication of the parasite . Taken together , T . gondii-induced Src signaling causes EGFR transactivation that in turn prevents parasite killing . Direct ligation of EGFR by MIC3 and MIC6 causes activation of Akt that prevents autophagic targeting of the parasite [9] . However , inhibition of FAK or Src knockdown did not significantly diminish the rapid S473 Akt phosphorylation induced by T . gondii infection of host cells ( Fig 6A ) . Phosphorylation of the Y845 residue of EGFR activates signaling cascades including STAT3 [34] . Thus , we examined whether Src and EGFR Y845 mediate rapid STAT3 activation in T . gondii-infected cells . T . gondii is reported to induce Y705-STAT3 phosphorylation in macrophages , fibroblasts and 293T cells [20 , 36–38] . Fig 6B–6G shows that tachyzoites of the RH or ME49 strains of T . gondii caused rapid Y705 STAT3 phosphorylation in RPE , A549 , mHEVc , BV-2 and NMuMG cells . The parasite kinase ROP16 is injected into the host cell cytoplasm , rapidly migrates to the nucleus and causes STAT3 activation . However , while ROP16 is reported to induce Y705 STAT3 phosphorylation as early 1 . 5 h post infection in macrophages , ROP16 did not appear to mediate STAT3 phosphorylation at earlier time points after infection in these cells [38] . To confirm these findings , we infected A549 epithelial cells with WT parasites or Δrop16 T . gondii . A549 cells infected with both types of parasites exhibited similar levels of Y705 STAT3 phosphorylation at the early time points post infection ( Fig 6D ) . Lower STAT3 phosphorylation was noted at 1 h post-infection with Δrop16 T . gondii ( Fig 6D ) . Next , we examined the role of Src and EGFR transactivation in STAT3 activation induced by T . gondii infection . Epithelial cells expressing either Src siRNA or control siRNA were challenged with T . gondii . Knockdown of Src diminished the early Y705 STAT3 phosphorylation induced by the parasite ( Fig 6E ) . In addition , the presence of EGFR in CHO cells markedly enhanced T . gondii-induced early Y705 STAT3 phosphorylation ( S3A Fig ) . To examine whether EGFR Y845 phosphorylation promotes STAT3 activation triggered by T . gondii , human epithelial cells were transfected with a plasmid encoding WT EGFR or Y845F EGFR . Transfection with the Y845F EGFR plasmid markedly inhibited T . gondii-induced early Y705 STAT3 phosphorylation ( Fig 6F ) . Finally , NMuMG epithelial cells that express EGFR AA ( defective in Src-induced Y845 phosphorylation ) exhibited reduced Y705 STAT3 phosphorylation after challenge with WT or Δrop16 T . gondii ( Fig 6G ) . Taken together , T . gondii-induced Y845 EGFR signaling mediates ROP16-independent STAT3 activation . STAT3 can inhibit autophagy [39 , 40] . Thus , we examined whether blockade of STAT3 triggered autophagy-dependent killing of T . gondii . Human epithelial cells were transfected with STAT3 siRNA or control siRNA followed by incubation with T . gondii tachyzoites . Knockdown of STAT3 resulted in a significant reduction in the percentages of infected cells , the numbers of tachyzoites and parasite-containing vacuoles per 100 cells at 24 h ( Fig 7A ) . Knockdown of STAT3 resulted in a significant accumulation of LC3 and LAMP-1 around the parasite ( Fig 7B and 7C ) . Moreover , silencing of Beclin 1 prevented induction of anti-T . gondii activity in cells subjected to STAT3 knockdown ( Fig 7D ) . Next , we examined the role of phosphorylation of Y705 in parasite survival . Transfection of mouse endothelial cells with the non-phosphorylatable mutant Y705F STAT3 caused accumulation of LC3 around the parasite ( Fig 7E ) . Expression of Y705F STAT3 in endothelial ( Fig 7F ) or epithelial cells ( S4 Fig ) resulted in a reduction in the percentage of infected cells and the numbers of tachyzoites per 100 cells at 24 h . Parasite killing was dependent on lysosomal enzymes since it was markedly inhibited by leupeptin/pepstatin ( Fig 7F; S4 Fig ) . Thus , parasite-induced STAT3 signaling prevents T . gondii killing dependent on the autophagy machinery and lysosomal protease activity . S51 phosphorylation of eIF2α plays an important role in the stimulation of autophagy [41] . In addition , increased S51 phosphorylation of eIF2α accompanies autophagic killing of T . gondii induced by CD40 ligation [42] . Thus , we examined the effect FAK—Y845 EGFR—STAT3 signaling on activation of eIF2α . While infection with T . gondii did not significantly increase S51 phosphorylation of eIF2α in FAK-sufficient cells , knockdown of FAK increased eIF2α phosphorylation in cells infected with T . gondii ( Fig 8A ) . Similarly , expression of the EGFR AA mutant ( defective in Y845 phosphorylation ) or transfection with Y705F-STAT3 resulted in increased S51 eIF2α phosphorylation in T . gondii-infected cells ( Fig 8B and 8C ) . In addition , T . gondii induced marked eIF2α phosphorylation in EGFR-null CHO cells but not in EGFR+ CHO cells ( S3B Fig ) . Next , we examined PKR signaling since this molecule is a major activator of eIF2α and autophagy , PKR links CD40 to autophagic killing of T . gondii and PKR is targeted by STAT3 to inhibit autophagy [40–43] . Knockdown of FAK or expression of EGFR AA increased T451 phosphorylation of PKR in cells infected with T . gondii ( Fig 8D ) . Moreover , transfection with DN PKR impaired S51 eIF2α phosphorylation in T . gondii-infected cells that are deficient in FAK ( Fig 8E ) . Finally , we examined whether killing of T . gondii required PKR signaling . The accumulation of LC3 around the parasite observed after inhibition of Src was ablated by transfection of DN PKR ( Fig 8F ) . Moreover , expression of DN PKR also prevented parasite killing induced by inhibition of Src ( Fig 8G ) . Taken together , these results suggest that T . gondii avoids autophagic killing by inducing cell signaling in host cells that prevents PKR and eIF2α activation .
Intracellular survival of T . gondii requires that the parasite avoids being targeted by autophagy , a process that would otherwise lead to lysosomal killing of the parasite . Herein we identified a pathway by which T . gondii prevents autophagic targeting ( Fig 9 ) . We report that during the process of invasion of mammalian cells , T . gondii induced expression of phospho Y397 FAK that appeared to associate in the host cell with RON4 , a parasite protein expressed in the moving junction . In turn , FAK triggered Src activation and Src-mediated EGFR transactivation resulting in STAT3 signaling . FAK—Src—Y845 EGFR—STAT3 signaling induced by T . gondii was relevant because it prevented targeting of the parasite by the autophagy machinery . Inhibition of the components of this cascade in a broad range of resting host cells resulted in accumulation of LC3 and LAMP-1 around the PV and parasite killing that was dependent on ULK1 , Beclin 1 and lysosomal protease activity . Moreover , inhibition of parasite-induced cell signaling enhanced phosphorylation of PKR and eIF2α . These events were relevant since the killing of T . gondii was ablated by expression of DN PKR . Taken together , these findings identified molecular events set in motion during the process of active invasion of host cells that enable T . gondii to restrict activation of PKR and eIF2α , molecules that are key drivers of autophagy . FAK is a cytoplasmic molecule that responds to extracellular signals and regulates cell motility , survival and proliferation [44] . These effects are mediated by its function as a scaffold protein that interacts and phosphorylates other signaling molecules . Our studies indicate that invasion of mammalian cells by T . gondii caused activation of FAK and its interacting partner Src . EGFR was not required for activation of these signaling molecules . FAK can become activated by mechanical stimulation and integrin clustering [45] . The studies with the cRGDfV peptide , neutralizing anti-αvβ3 Ab and knockdown of β1 integrin suggest that FAK signaling induced by T . gondii infection is driven to a large extent by β integrins , presumably in the form of mechano-transduction-induced integrin clustering at the site of host cell penetration causing activation of mammalian cytoplasmic FAK . While host cell invasion is driven by the T . gondii glideosome , the host cell also contributes to this process by forming an F-actin ring underneath the moving junction and accumulating microtubules at this site that would anchor the moving junction to the host cell cytoskeleton during invasion [46–48] . The studies presented herein revealed that the process of invasion not only induces rearrangement of host cell cytoskeleton but triggers FAK-Src signaling that plays a critical role in promoting parasite survival . Src regulates various cellular processes through interactions with signaling proteins including EGFR . Indeed , binding of Src to EGFR can trigger EGFR signaling even in the absence of ligand binding [49] . While binding of EGF to EGFR induces autophosphorylation at tyrosine sites in the C-terminal non-catalytic domain of the receptor that act as docking sites for various signaling molecules , Src transactivation of EGFR is characterized by phosphorylation at a unique Y845 site located in the kinase domain of EGFR that leads to activation of alternate signaling molecules [31–34] . We reported that , through the effect of MIC3 and MIC6 ( adhesins with EGF-like domains ) , T . gondii induces EGFR autophosphorylation and Akt activation that prevents autophagic targeting of the PV [9] . In contrast , Src-dependent Y845 phosphorylation of EGFR was critical for parasite-induced STAT3 activation but played no significant role in T . gondii-induced early Akt signaling . These results differ from those reported in cells exposed to HIV Tat or cells infected with Salmonella typhimurium , where both HIV Tat-induced Src activation and bacterial-induced FAK activation signal through Akt to inhibit autophagy [39 , 50] . Our studies support the existence of 2 pathways by which T . gondii impairs autophagic targeting , FAK→Src→Y845-EGFR→STAT3 and MIC3/6→EGFR autophosphorylation→Akt . Our studies indicate that T . gondii induces rapid STAT3 Y705 phosphorylation dependent on FAK—Src and Y845 EGFR phosphorylation . Knockdown of STAT3 and expression of Y705F STAT3 revealed that T . gondii activates STAT3 to prevent autophagic killing . While phosphorylated STAT3 is a well-recognized transcriptional regulator , STAT3 can also signal in the cytoplasm . Phosphorylated STAT3 crosstalks with signaling molecules in the cytoplasm and has been proposed to modulate signaling at this level [51–53] . We report that in T . gondii-infected cells , expression of Y705F STAT3 or inhibition of upstream inducers of STAT3 Y705 phosphorylation resulted in increased phosphorylation of PKR and/or eIF2α . Moreover , the studies using DN PKR indicate that targeting by LC3 and parasite killing triggered by inhibition of host cell signaling was dependent on PKR . STAT3 can inhibit autophagy through constitutive binding to PKR that prevents PKR activation , an effect that does not require STAT3 phosphorylation at Y705 [40] . Our studies suggest an additional mechanism by which STAT3 Y705 phosphorylation induced by T . gondii impairs PKR-eIF2α signaling . STAT3 is also reported to inhibit autophagy through increased transcription of negative regulators of autophagy [54] . Moreover , S727 phosphorylation of STAT3 promotes its mitochondrial localization and is reported to increase autophagy [55] . Altogether , STAT3 appears to regulate autophagy through various mechanisms , the nature of which may be determined in a context-dependent manner . After being injected into the host cell during invasion , the T . gondii protein ROP16 migrates to the nucleus of the host cell and induces STAT3/6 activation that in turn impairs expression of IL-12 , a cytokine that confers host protection against T . gondii [37 , 38] . However , the effect of ROP16 on STAT3 phosphorylation is not noticeable until after 1–1 . 5 hr post-infection ( [38] and this study ) and the parasite induces STAT3 activation as early as 2 min post-infection [36] . The work herein identified FAK—Src—EGFR Y845 phosphorylation as a novel upstream inducer of STAT3 activation during T . gondii infection . It appears that this pathway rather than the ROP16 –STAT3 pathway is a major inhibitor of autophagic targeting of T . gondii . Taken together , T . gondii has evolved distinct mechanisms to activate STAT3 and through them sets in motion processes that impair the autophagic killing of the parasite and Th1 immunity . Our studies identified a molecular cascade that appears to be associated with the formation of the moving junction during parasite invasion that leads to activation of STAT3 preventing targeting of the parasite by the autophagy machinery . Moreover , these studies suggest that parasite-induced STAT3 signaling functions by preventing activation of PKR and eIF2α . The demonstration that immune ( CD40 ) -mediated induction of autophagic killing of T . gondii is dependent on activation of PKR [42] suggests that the intracellular survival of T . gondii may be affected by opposing effects of the parasite and cell mediated immunity on key components of autophagy signaling . Given that autophagy promotes in vivo protection against T . gondii [42 , 56] , pharmacologic approaches to prevent the parasite from activating host cell signaling that counter-regulates autophagy may result in novel treatment against toxoplasmosis .
Human retinal pigment epithelial ( RPE ) cells ( ARPE-19 ) , human lung epithelial cells ( A549 ) ( both from American Type Culture Collection , ATCC , Manassas , VA ) , mouse high endothelial venule cells ( mHEVc , gift from Joan Cook-Mills , Northwestern University ) , mouse microglia cell line ( BV-2 , gift from Kalipada Pahan , University of Nebraska ) , MDA-MB-231 human breast epithelial cells ( ATCC ) [30] and normal mouse mammary gland ( NMuMG ) epithelial cells ( ATCC ) [35] were cultured in either RPMI or DMEM plus 10% fetal bovine serum ( FBS; HyClone , Logan , UT ) . NMuMG with stable expression of human WT EGFR or mutant EGFR where 679 , 680-LL was converted to AA ( EGFR AA ) were previously described [35] . Chinese Hamster Ovary ( CHO; gift from Cathleen Carlin , Case Western Reserve University ) cells were cultured in MEM plus 10% FBS . Tachyzoites of the T . gondii strains RH ( Type I strain ) , RH-YFP , RH-RFP , RH ROP16 knockout ( Δrop16 [19] ) , RH ROP18 knockout ( Δrop18 ) , their corresponding WT controls , ( gifts from John Boothroyd , Stanford University , Stanford , CA ) , inducible conditional MIC8 knockout ( MIC8KOi ) ( gift from Markus Meissner , University of Glasgow ) , ME49 ( Type II strain ) or VAND ( atypical strain , BEI Resources , Manassas , VA ) were maintained in human foreskin fibroblasts ( ATCC ) . MIC8KOi parasites were cultured in HFF in the presence of anhydrotetracycline ( 1 μg/ml ) for 48 h to deplete MIC8 . T . gondii infection was synchronized using potassium buffer shift . Mammalian cells were incubated with the Src inhibitor PP2 ( 0 . 2 μM; Sigma-Aldrich , St . Louis , MO ) , FAK inhibitor PF-573228 ( 1 μM; Pfizer , Inc . , New York , NY; both 1 h prior to challenge with T . gondii ) , integrin blocking peptide cRGDfV or control peptide cRADfV ( 3 μM; Bachem , Torrance , CA; 1 h prior to challenge ) , neutralizing anti-αvβ3 mAb LM609 ( 15 μg/ml; EMD Millipore , Temecula , CA; 1 h prior to challenge ) , pertussis toxin ( PTx; 100 ng/ml; EMD Millipore; 4 h prior to challenge ) , lysosomal inhibitors leupeptin and pepstatin ( both 10 μM; EMD Millipore; 1 h after challenge with T . gondii ) . Parasite load was assessed as described [9] . In certain experiments cells were incubated with lysophosphatidic acid ( LPA; 10 μM; Sigma Chemical , St . Louis , MO ) . None of the reagents described above affected cell viability ( trypan blue exclusion ) . Cells were transfected with Src siRNA [57] , ULK1 siRNA ( Life Technologies ) , Beclin 1 siRNA [58] , STAT3 siRNA [59] , MyD88 siRNA [60] or control siRNA ( Dharmacon , Lafayette , CO ) as well as plasmids encoding WT EGFR , Y845F EGFR [32] , WT STAT3 , Y705 STAT3 ( gift from Jim Darnell; Addgene plasmids # 8707 and 8709 ) [61] , FLAG-tagged WT-PKR , dominant negative ( DN ) -PKR ( K296R ) or empty plasmid ( gifts from Bill Sugden , University of Wisconsin ) using Lipofectamine 2000 ( Invitrogen , Carlsbad , CA ) following manufacturer’s instructions . mHEVc cells were transduced with lentivirus vectors encoding FAK or control shRNA ( Open Biosystems , Lafayette , CO ) followed by selection with puromycin ( 10 μg/ml; Sigma-Aldrich ) . MDA-MB-231 cells transduced with lentiviral vectors encoding shRNA against β1 integrin or control shRNA were previously described [30] . Mammalian cells challenged with RFP T . gondii were fixed with 4% paraformaldehyde at 5 h and stained with rabbit anti-LC3 antibody ( MBL International , Woburn , MA ) plus Alexa 488-conjugated secondary antibody ( Invitrogen , Carlsbad , CA ) . Mammalian cells infected with YFP T . gondii were fixed at 8 h and stained with mouse anti-human LAMP-1 mAb ( Developmental Studies Hybridoma Bank; Iowa City , IA ) plus Alexa 555-conjugated secondary antibody ( Jackson ImmunoResearch Laboratories Inc . ) . Accumulation of LC3 or LAMP-1 around T . gondii was defined as the presence of a ring-like structure that surrounds the parasite [62 , 63] . At least 50 cells per well ( duplicate or triplicate wells per group per experiment ) were counted manually . CHO cells were challenged with YFP T . gondii and fixed at 3 min . Monolayers were stained with mouse anti-phospho-Y397 FAK antibody ( BD Biosciences ) plus Alexa 555-conjugated secondary antibody ( Jackson ImmunoResearch Laboratories Inc . ) and rabbit anti-RON4 antibody ( gift from John Boothroyd ) followed by incubation with goat anti-rabbit Alexa 647-conjugated secondary antibody ( Jackson ImmunoResearch Laboratories Inc . ) . Specificity of staining was determined by incubating monolayers with secondary antibody alone . Slides were analyzed using Leica DMI 6000 B automated microscope equipped for epifluorescence microscopy . Membranes were probed with antibodies against Src , phospho-Y416 Src ( all from Cell Signaling Technology , Danvers , MA ) , EGFR ( Santa Cruz Biotechnology , Santa Cruz , CA ) , phospho-Y845 EGFR ( Cell Signaling ) , FAK ( Santa Cruz Biotechnology ) , phospho-Y397 FAK ( BD Biosciences , San Jose , CA ) , STAT3 ( Cell Signaling ) , phospho-Y705 STAT3 ( Cell Signaling ) , eIF2α ( Cell Signaling ) , phospho-S51 eIF2α ( Cell Signaling ) , PKR ( Santa Cruz Biotechnology ) , phospho-T451 PKR ( Thermo Fisher Scientific , Waltham , MA ) , Beclin 1 ( BD Biosciences ) , ULK1 ( Sigma-Aldrich ) , MyD88 ( Cell Signaling ) or actin ( Santa Cruz Biotechnology ) followed by incubation with secondary antibodies ( Santa Cruz Biotechnology ) . Intensities of phosphorylated and total proteins were calculated using ImageJ ( NIH ) . Phospho-protein signal was normalized to total protein signal , before normalizing it relative to the control or 0 min time-point condition . Results from pooled experiments were analyzed for statistical significance using 2-tailed Student’s t test and ANOVA . Differences were considered statistically significant when P < 0 . 05 .
|
Toxoplasma gondii is a protozoan that resides within host cells . Avoiding lysosomal degradation including that mediated by autophagy is central to the ability of T . gondii to survive within these cells . We uncovered that during the process of active invasion of host cells , T . gondii activates in a broad range of mammalian cells a signaling cascade downstream of FAK-Src that prevents targeting of the intracellular parasite by autophagy enabling its survival . This pathway is different from the previously identified survival strategy dependent upon parasite micronemal proteins-mediated EGFR autophosphorylation and Akt activation . Importantly , preventing T . gondii-induced manipulation of host cell signaling led to parasite killing in host cells that were not activated by immune modulators . Pharmacologic approaches that impair parasite-induced activation of host cell signaling that counter-regulates autophagy may lead to novel treatment against toxoplasmosis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"phosphorylation",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"autophagic",
"cell",
"death",
"parasite",
"groups",
"viral",
"transmission",
"and",
"infection",
"toxoplasma",
"gondii",
"gene",
"regulation",
"cell",
"processes",
"microbiology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"apicomplexa",
"tachyzoites",
"integrins",
"protozoans",
"toxoplasma",
"cellular",
"structures",
"and",
"organelles",
"small",
"interfering",
"rnas",
"proteins",
"cell",
"adhesion",
"extracellular",
"matrix",
"gene",
"expression",
"biochemistry",
"rna",
"eukaryota",
"host",
"cells",
"cell",
"biology",
"post-translational",
"modification",
"nucleic",
"acids",
"virology",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna",
"organisms"
] |
2017
|
Toxoplasma gondii induces FAK-Src-STAT3 signaling during infection of host cells that prevents parasite targeting by autophagy
|
Adaptive immunity to Mycobacterium tuberculosis controls progressive bacterial growth and disease but does not eradicate infection . Among CD4+ T cells in the lungs of M . tuberculosis-infected mice , we observed that few produced IFN-γ without ex vivo restimulation . Therefore , we hypothesized that one mechanism whereby M . tuberculosis avoids elimination is by limiting activation of CD4+ effector T cells at the site of infection in the lungs . To test this hypothesis , we adoptively transferred Th1-polarized CD4+ effector T cells specific for M . tuberculosis Ag85B peptide 25 ( P25TCRTh1 cells ) , which trafficked to the lungs of infected mice and exhibited antigen-dependent IFN-γ production . During the early phase of infection , ∼10% of P25TCRTh1 cells produced IFN-γ in vivo; this declined to <1% as infection progressed to chronic phase . Bacterial downregulation of fbpB ( encoding Ag85B ) contributed to the decrease in effector T cell activation in the lungs , as a strain of M . tuberculosis engineered to express fbpB in the chronic phase stimulated P25TCRTh1 effector cells at higher frequencies in vivo , and this resulted in CD4+ T cell-dependent reduction of lung bacterial burdens and prolonged survival of mice . Administration of synthetic peptide 25 alone also increased activation of endogenous antigen-specific effector cells and reduced the bacterial burden in the lungs without apparent host toxicity . These results indicate that CD4+ effector T cells are activated at suboptimal frequencies in tuberculosis , and that increasing effector T cell activation in the lungs by providing one or more epitope peptides may be a successful strategy for TB therapy .
Even though its etiologic agent was discovered over 125 years ago , tuberculosis remains a global scourge , killing 1 . 7 million people in 2009 , at least ¾ of whom were immunocompetent [1] . Long-term persistence of Mycobacterium tuberculosis , which resides principally in phagocytic cells within the lungs , results in a chronic infection despite the presence of an apparently appropriate adaptive immune response . In mice infected with virulent M . tuberculosis , the early phase of infection proceeds with unchecked bacterial growth until day 17–21 post-infection , when adaptive immunity finally exerts control of bacterial growth in the lungs . Control of infection in both humans and mice critically depends on M . tuberculosis-specific CD4+ Th1 cell responses , which include production of IFN-γ [2] , [3]; however adaptive immune responses do not eradicate the infection . Several potential mechanisms may account for the failure of adaptive immune responses to eradicate the bacteria in tuberculosis . Generation of M . tuberculosis-specific CD4+ effector T cells is delayed compared with responses to other pathogens [2] , [4] . In addition , certain individuals , or strains of mice , may develop inappropriate ( e . g . , Th2 ) [5] , [6] or imbalanced effector phenotypes such as Th1/Th17 [7] in response to infection . However , even in humans or mice that develop Th1 responses , a failure of CD4+ effector T cells to recognize infected cells may preclude their optimal activation and limit induction of effector functions in the lungs . Prevention of effector T cell activation could result from impaired antigen presentation by lung APCs containing M . tuberculosis [8] , [9] , [10] or because the antigens that effector T cells recognize are not expressed or otherwise available in the lungs . Furthermore , host regulatory mechanisms that limit immune pathology , such as T regulatory cells [11] , production of inhibitory cytokines [12] , and , possibly , onset of T cell exhaustion [13] , [14] may inhibit the activity of effector T cells at the site of infection . Finally , even when CD4+ effector T cells are activated , the efficacy of these responses may be limited by the impaired ability of infected cells to respond to IFN-γ [15] , [16] , [17] , induce phagosome maturation [18] , [19] , or undergo apoptosis [9] , [20] , [21] , [22] . Understanding the contribution of each of these potential mechanisms limiting adaptive immunity to M . tuberculosis is an essential prerequisite for vaccine design and other immunologic approaches to tuberculosis prevention and therapy . Here , we report that CD4+ effector T cells are activated at submaximal and suboptimal frequencies in the lungs during M . tuberculosis infection , that this is due in part to bacterial modulation of antigen expression , and that increasing the availability of a single antigen results in improved immune control of M . tuberculosis .
We hypothesized that M . tuberculosis evades adaptive immunity by modulating the activation of CD4+ effector T cells at the site of infection in the lungs . Since in vitro studies have revealed evidence that M . tuberculosis modulates MHC class II antigen presentation [10] , [23] , [24] , [25] , [26] , we focused on in vivo activation of CD4+ T cells in the lungs . We reasoned that , if M . tuberculosis-infected cells do not present antigens efficiently to effector T cells in the lungs , then the frequency of activation of effector functions of CD4+ cells would also be low at the site of infection . To test this , we used direct intracellular cytokine staining of lung cells from infected mice for IFN-γ , without ex vivo restimulation . We found that that the frequency of IFN-γ expression by CD4+ T cells in the lungs varied with the time of infection ( Figure 1B ) . IFN-γ+ CD4+ cells were undetectable in the lungs at day 14 , increased in frequency beginning by day 21 to a peak at day 35 post-infection , and then markedly decreased afterward; no more than 7% of the bulk population of CD4+ T cells expressed IFN-γ at any time point after infection , and fewer than 1% expressed IFN-γ during the chronic phase . Other studies investigating IFN-γ production by CD8+ T cells in vivo have used treatment of mice with brefeldin A or inclusion of brefeldin A during cell isolation and staining [27] , [28] . However , we determined that these methods did not improve detection of intracellular IFN-γ by CD4+ T cells during M . tuberculosis infection ( Figure S1 ) . These data indicate that a small minority of polyclonal CD4+ T cells recruited to the lungs of M . tuberculosis-infected mice are activated to produce IFN-γ at a given time , and are consistent with defective antigen presentation , costimulation , and/or inhibition of effector T cell activation at the site of infection . Since the low frequency of CD4+ T cell expression of IFN-γ in the lungs of M . tuberculosis-infected mice could be due to the presence of effector cells that traffic to the lungs but are not specific for M . tuberculosis antigens , we performed the remainder of our studies using CD4+ TCR transgenic T cells that specifically recognize a well-characterized immunodominant M . tuberculosis antigen . To quantitate the frequency of activation of M . tuberculosis antigen-specific effector cells in the lungs , we prepared CD4+ Th1 effector cells ( P25TCRTh1 cells ) from transgenic mice with a TCR specific for peptide 25 ( amino acids 240–254 ) of Ag85B . When P25TCRTh1 cells were incubated with irradiated splenocytes in the absence of peptide 25 , <1 . 0% of the cells expressed IFN-γ as detected by intracellular staining and flow cytometry , whereas addition of peptide 25 in vitro induced IFN-γ expression in ∼90% of cells ( Figure 2A ) . This result demonstrated that the frequency of IFN-γ staining in P25TCRTh1 cells can specifically assay antigen dependent stimulation of P25TCRTh1 cells . Since Day 21 post-infection corresponds to an acute stage of infection when adaptive immune effector mechanisms have been initiated and reduce the rate of bacterial population growth in the lungs , and since it resembles the stage of LCMV infection in which a high frequency of antigen-specific CD8+ T cell responses are observed [28] , we chose this time point for initial characterization of P25TCRTh1 cell responses in vivo . We verified that adoptively transferred P25TCRTh1 cells traffic to the site of infection by examining sections of lungs from infected mice that had received CFP+ P25TCRTh1 cells . CFP+ cells were abundant in the lung parenchyma , and were concentrated in granulomas ( Figure 2B ) . Furthermore , we determined that >85% of the transferred cells were protected from labelling by an i . v . injection of PerCP-labeled anti-CD4 antibody , indicating that adoptively transferred P25TCRTh1 cells efficiently migrate out of the lung vasculature into the parenchyma of infected lungs ( Figure S2A ) . To determine the frequency of activation of antigen-specific CD4+ effector T cells in the lungs early in infection , we adoptively transferred P25TCRTh1 cells on day 18 and harvested them on day 21 after infection of wild-type mice with wild-type M . tuberculosis H37Rv . The frequency of IFN-γ+ P25TCRTh1 cells isolated from the lungs was unexpectedly low at Day 21 post-infection ( Figure 2C and 2D ) . Approximately 1–2% of the transferred P25TCRTh1 cells were stimulated to produce IFN-γ in vivo at that time point ( Figure 2C ) , and this percentage was similar to the frequency of total endogenous lung CD4+ T cells expressing IFN-γ on day 21 post-infection ( Figure 1B , 2D ) . Moreover , after intravenous injection of PerCP-labeled anti-CD4 antibody , the only IFN-γ+ P25TCRTh1 cells identified were PerCP negative ( Figure S2B ) , indicating that the responding cells were those that had migrated out of the vasculature into the lung parenchyma and were protected from staining by the in vivo injection of antibody . We verified that stimulation of P25TCRTh1 cells to express intracellular IFN-γ is due to recognition of Ag85B peptide 25 by transferring P25TCRTh1 cells into mice infected with an Ag85B-null strain of M . tuberculosis ( ΔAg85B ) , which is equivalent to wild-type H37Rv in virulence [2] . A lower mean percentage ( 0 . 74% ) of P25TCRTh1 cells isolated from ΔAg85B-infected mice expressed IFN-γ than those from H37Rv-infected mice ( Figure 2C and 2D ) . This indicates that in vivo IFN-γ production by P25TCRTh1 cells is antigen-dependent and not the consequence of inflammatory cytokines present at the site of infection . We also evaluated several alternative approaches to detecting effector T cell activation in the lungs . P25TCRTh1 cells expressed both CD25 and CD44 prior to adoptive transfer , which excluded their use in evaluating effector cell activation in vivo . Surface expression of CD69 was induced after adoptive transfer of P25TCRTh1 effector cells into H37Rv-infected mice; however , we found similar induction of CD69 in mice infected with ΔAg85B , indicating that it did not specifically reflect antigen-dependent effector cell activation . This result , together with evidence that CD69 can be induced by costimulation and by certain cytokines present at the site of M . tuberculosis infection [29] , [30] , [31] , indicates that expression of intracellular IFN-γ is the most accurate reporter of antigen specific Th1 effector cell activation in the lungs . Together , these results indicate that even though they traffic efficiently to the site of infection , Ag85B peptide 25-specific CD4+ effector cells are activated to execute their Th1 effector function at low frequency in the lungs of M . tuberculosis-infected mice . Although IFN-γ production by P25TCRTh1 cells at day 21 was dependent on Ag85B , the frequency of IFN-γ+ cells was surprisingly low in H37Rv infected mice . One possible explanation for the low frequency of activation of effector cells is that their cognate antigen is not available for recognition at the site of infection . To test this hypothesis , we provided antigen in vivo by injecting peptide 25 intravenously into mice that had been infected 21 days earlier . When P25TCRTh1 recipient , H37Rv-infected mice received peptide 25 six hours prior to lung cell harvest , the frequency of IFN-γ+ P25TCRTh1 cells increased to 20–50% ( Figure 2C and 2D ) . Similarly , peptide 25 injection stimulated a higher frequency of IFN-γ expression by endogenous CD4+ T cells from mice infected with H37Rv ( Figure 2C and 2D ) , consistent with prior evidence that peptide 25 of Ag85B is a dominant antigen in C57BL/6 mice infected with M . tuberculosis [32] , [33] . P25TCRTh1 cells transferred into ΔAg85B-infected recipients were also stimulated at a higher frequency after intravenous peptide 25 treatment , while endogenous CD4+ T cells from ΔAg85B-infected mice did not respond to peptide 25 with increased IFN-γ expression ( Figure 2D ) . The failure of endogenous CD4+ T cells from ΔAg85B-infected mice to respond to peptide 25 injection reflects the absence of Ag85B peptide 25-specific effector T cells generated in response to this infection . These results indicate that the frequency of IFN-γ+ P25TCRTh1 cells is an accurate and specific measure of CD4+ effector T cell stimulation in response to presentation of Ag85B peptide 25 in vivo . The observation that in vivo IFN-γ responses to peptide 25 injection depend on the presence of previously-generated ( endogenous or transferred ) peptide 25-specific effector T cells indicates that the responses are not due to a nonspecific effect of the epitope peptide on costimulation or responses of CD4+ T cells with specificity for other antigens . In addition , they demonstrate that if antigen is made available to them , adoptively transferred P25TCRTh1 cells can respond to antigen in the infected lungs , and they provide evidence against an exclusive role for T regulatory cells and/or suppressive cytokines in limiting the activation of CD4+ effector cells at the site of M . tuberculosis infection in the lungs . To further characterize the in vivo assay system , and to evaluate the possibility that low frequencies of P25TCRTh1 responses are attributable to either competition for antigen by endogenous CD4+ T cells and/or a dominant effect of T regulatory cells , we specifically ablated endogenous T cells from M . tuberculosis-infected CD4-DTR mice [34] prior to assaying P25TCRTh1 responses in vivo . Compared to untreated mice , DT treatment reduced the fraction of endogenous CD4+ T cells in the lung by an average of 48 . 9% , p = 0 . 0053 ( Figure S3A ) . However , this had no effect on the percentage of P25TCRTh1 cells activated to produce IFN-γ ( Figure S3B ) . These results strongly suggest that the low frequency of activation of P25TCRTh1 cells is caused neither by competition for peptide 25:MHC II complexes by endogenous CD4+ T cells , nor by the influence of T regulatory cells in the lungs . We therefore conclude that the response of adoptively transferred P25TCRTh1 cells is an accurate reflection of MHC II presentation of Ag85B peptide 25 by lung APCs during infection . Adaptive immunity restricts progressive growth of M . tuberculosis , but it does not eliminate the bacteria from the lungs , which results in chronic infection in mice and latent infection in humans . To determine whether suboptimal activation of M . tuberculosis-specific T cells contributes to the ability of the bacteria to persist , we first asked whether activation of P25TCRTh1 cells in the lungs changes as infection progresses to a chronic phase . To compare the frequency of effector T cell stimulation at various stages of infection , we transferred P25TCRTh1 cells into H37Rv-infected mice on day 11 , 18 , 25 , 32 , or 39 post-infection . Lung cells were harvested 72 hours after transfer ( day 14 , 21 , 28 , 35 , or 42 post-infection ) and analyzed by flow cytometry for intracellular IFN-γ without ex vivo restimulation . The proportion of P25TCRTh1 cells producing IFN-γ was highest ( ∼10% ) on day 14 ( Figure 3A and 3C ) . These results indicate that during the acute stage of infection , adoptively transferred P25TCRTh1 cells are stimulated in the lungs at a frequency comparable to that of TCR transgenic CD4+ effector cells at the site of injection of a protein antigen and adjuvant [35] . In contrast , expression of IFN-γ by endogenous ( CD45 . 2− ) CD4+ cells was rare ( <0 . 1% ) at that time point ( Figure 1B and 1C ) . The difference between transferred and endogenous cell responses on day 14 is consistent with our previous observation that initiation of adaptive immunity to M . tuberculosis is delayed until day 11–14 post-infection , and consequently , endogenous CD4+ effector T cells specific for M . tuberculosis antigens are first detected in the lungs on day 17 post-infection . [2] . The frequency of IFN-γ production by P25TCRTh1 cells progressively decreased from day 14 to day 42 post-infection , indicating a decrease in the efficiency of peptide 25-specific T cell stimulation as infection enters its chronic phase ( Figure 3A and 3C ) . These results with TCR transgenic CD4+ effector cells closely mimic the results observed with endogenous polyclonal CD4+ T cells after day 14 post-infection ( Figure 1B and 1C ) . Although Ag85B peptide 25-specific responses reached an earlier peak and decreased earlier than did those of endogenous polyclonal CD4+ T cell responses , the results with the two cell populations were similar , with endogenous CD4+ effector T cell responses also diminishing by day 42 post-infection . To determine whether activation of naïve Ag85B peptide 25-specific CD4+ T cells is also diminished in the later stages of M . tuberculosis infection , we assayed the response of adoptively transferred naïve P25 TCR-Tg T cells in the lung-draining mediastinal lymph nodes of H37Rv-infected mice at various time points post-infection . 7 days after transfer , we harvested lymph node cells and measured in vivo T cell proliferation by flow cytometry using a CFSE dilution assay . The rate of naïve P25TCR-tg T cells was highest upon transfer into mice on day 18 post-infection , while fewer cells exhibited CFSE dilution at days 24 and 48 post-infection ( Figure 3B ) . These results indicate that decreased stimulation of P25TCRTh1 effector cells is also accompanied by decreased generation of peptide 25 specific effector T cells from naive cells at later stages of infection . Since treatment of infected mice with exogenous peptide 25 enhanced T cell responses , indicating that adoptively-transferred P25TCRTh1 cells are capable of responding to antigen stimulation in the lungs , we hypothesized that availability and/or presentation of antigen is a limiting factor in the activation of CD4+ effector T cells at the site of M . tuberculosis infection . To test this hypothesis , we first investigated whether changes in the expression of the M . tuberculosis gene that encodes Ag85B influence the frequency of activation of P25TCRTh1 effector cells . We found that the frequency of in vivo activation of P25TCRTh1 cells mimicked the temporal pattern of expression of fbpB ( which encodes Ag85B ) by M . tuberculosis in vivo ( Figure 3C ) . This suggests that reduced expression of Ag85B contributes to the low frequency of activation of Ag85B-specific CD4+ effector cells in the lungs , thus resembling previously-reported observations with Salmonella FliC expression and FliC-specific CD4+ T cell responses [36] . To test the hypothesis that fbpB down-regulation contributes to the submaximal frequency of CD4+ effector cell activation and the limited efficacy of the Th1 response in vivo , we constructed a recombinant strain of M . tuberculosis to express fbpB at high levels during chronic infection . Using the ΔAg85B strain as a background , we introduced a wild-type fbpB allele under control of the hspX/acr/Rv2031c promoter to the M . tuberculosis chromosome via the pMV306 integrating vector . hspX is expressed at high levels during chronic phase infection in an expression pattern inverse to fbpB [37] , [38] . This strain ( hspXp:fbpB , termed “CPE85B” for chronic phase expressed Ag85B ) exhibited higher fbpB expression compared to H37Rv in the lungs of mice after aerosol infection ( Figure 4A ) . The expression of fbpB measured by RT-qPCR was approximately 10-fold higher ( normalized for the abundance of 16S rRNA ) at day 21 post-infection for CPE85B than for H37Rv . As the infection progressed to chronic phase ( day 28–42 post-infection ) , fbpB expression from the native promoter declined by approximately 100-fold while fbpB expression driven by the hpsX promoter remained at nearly constant , higher levels ( Figure 4A ) . Increased fbpB gene expression in the CPE85B strain was accompanied by markedly enhanced expression and secretion of Ag85B protein when the hspX promoter was induced in stationary liquid culture ( Figure 4B ) . To determine whether forced expression of fbpB in M . tuberculosis results in increased presentation of Ag85B peptide 25 to CD4+ T cells , we infected bone marrow-derived dendritic cells ( BMDC ) with either H37Rv or CPE85B and compared their ability to activate P25TCRTh1 cells in culture . At all APC∶T cell ratios examined , DCs infected with CPE85B induced significantly greater amounts of IFN-γ secretion from P25TCRTh1 cells than did DCs infected with H37Rv ( Figure 4C ) . To determine whether forced expression of fbpB can increase the frequency of P25TCRTh1 stimulation during H37Rv infection in vivo , we compared the frequency of P25TCRTh1 cell activation in the lungs of mice infected with either H37Rv or CPE85B . Compared to cells from H37Rv-infected recipients , P25TCRTh1 cells from CPE85B-infected mice produced IFN-γ with a 2-fold ( day 21 ) to 5-fold ( day 42 ) higher frequency ( Figure 4D and E ) . These findings indicate that forced expression of fbpB by M . tuberculosis increases the proportion of P25TCRTh1 cells that are activated to produce IFN-γ in the lungs . By suppressing fbpB expression after the initial stages of infection , wild-type M . tuberculosis can reduce the frequency of activation of Ag85B-specific effector T cells . Although expression of fbpB was maintained at high levels from day 14 to day 42 post-infection , P25TCRTh1 cell stimulation in CPE85B-infected mice was only two- to five-fold higher than in mice infected with H37Rv , and decreased as infection progressed to chronic stage , indicating that other mechanisms , such as inhibition of antigen presentation and/or induction of regulatory T cells , exist to limit the activation of CD4+ effector T cells in the lung . We reasoned that , if diminishing fbpB expression during chronic infection limits effector T cell activation and thereby enables M . tuberculosis to evade adaptive immunity , then constitutive expression of fbpB throughout infection should improve immune control of infection . To test this hypothesis , we infected mice with either H37Rv or CPE85B and quantitated M . tuberculosis CFUs in the lungs throughout the course of infection . The rates of bacterial growth for the two strains were indistinguishable prior to day 14 post infection ( Figure 4F ) , indicating that expression of fbpB by the hspX promoter does not attenuate M . tuberculosis in vivo during the innate immune stage of infection , prior to recruitment of CD4+ effector T cells to the lungs . Indeed , the in vivo generation time of the CPE85B strain ( 23 . 0 h ) was slightly shorter than that of H37Rv ( 26 . 4 h ) during days 1–14 of infection ( these are not significantly different by nonlinear curve fit and F test ) . However , at times corresponding to the adaptive immune phase of infection , the bacterial burden of the CPE85B strain in the lungs was approximately 10-fold lower than that of H37Rv ( Figure 4F ) . These results suggest that forced expression of fbpB partially overcomes the antigen deficit that limits the activation of CD4+ T cells in the lung during chronic infection and allows greater antimycobacterial efficacy of the adaptive immune response . The observation that CPE85B demonstrates a growth pattern indistinguishable from H37Rv during the first two to three weeks of infection , prior to onset of adaptive immunity , suggested that CPE85B was not inherently attenuated for growth in vivo . However , we considered the possibility that over-expression of fbpB could cause attenuation of M . tuberculosis as a result of gene dysregulation or toxicity of an overabundant Ag85B protein . Notably CPE85B demonstrated a similar growth pattern to H37Rv during in vitro shaking culture . Furthermore , under conditions of hypoxic stationary culture , when Ag85B protein is strongly expressed by CPE85B compared to H37Rv , the survival of the CPE85B strain is not impaired compared with that of wild-type bacteria ( Figure 4B ) . Taken together , these findings imply that impaired persistence of M . tuberculosis CPE85B in vivo is the consequence of increased antigen presentation and activation of CD4+ T cells , and not due to intrinsic attenuation of the CPE85B strain in vitro or in vivo . We reasoned that if the decreased lung bacterial burden of CPE85B compared with that of H37Rv is attributable to increased antigen presentation and recognition by CD4+ T cells , then the attenuated phenotype of CPE85B should be abrogated in mice lacking CD4+ T cells . Indeed , whereas wild type C57BL/6 mice infected with CPE85B survived significantly longer than those infected with H37Rv ( median survival >300 and 239 days , respectively; p = 0 . 0062 ) , MHCIIKO mice , which lack CD4+ T cells , exhibited indistinguishable susceptibility to infection with the CPE85B and H37Rv strains ( median survival 79 and 81 days , respectively; p = 0 . 425 ) , ( Figure 5A ) , clearly establishing that in vivo attenuation of the CPE85B strain depends on MHC II antigen presentation and CD4+ T cell responses . These results also indicate that increased antigen expression , accompanied by increased antigen-specific T cell activation , can enhance control of M . tuberculosis without detectable detrimental effects , since wild-type mice infected with the CPE85B strain survived longer than mice infected with H37Rv . Since MHC II-deficient mice are highly susceptible to M . tuberculosis infection , this could potentially mask any hypothetical CD4+ T cell-independent mechanisms of attenuation of the CPE85B strain . We reasoned that , if mechanisms other than increased CD4+ T cell recognition contribute to the lower burdens of CPE85B , then this strain would not recover and grow normally in the lungs when CD4+ T cells are depleted during the chronic phase of infection . We infected mice with H37Rv or CPE85B and allowed the infection to proceed for 28 days , when initial lung CFUs were measured for each group . As expected , bacterial CFUs for CPE85B were ∼3 fold lower than H37Rv at this time point ( Figure 5B ) . The remaining mice in each infection group were then treated with monoclonal antibody GK1 . 5 every 6 days until day 50 post-infection to deplete CD4+ T cells . After an initial lag , in which neither bacterial strain expanded , both CPE85B and H37Rv resumed growth in the lungs at indistinguishable rates ( Figure 5B ) . Taken together , these data provide strong evidence that improved control of the CPE85B strain is attributable to increased activation of Ag85B-specific CD4+ T cells , although we cannot exclude the possibility that other factors contribute to the lower lung burdens of CPE85B that appear after the development of adaptive immunity . Our observation that forced expression of fbpB increased the frequency of Ag85B peptide 25-specific CD4+ T cells and reduced the bacterial burden in the lungs ( Figure 4D , 4E , and 4F ) , together with our observation that injection of peptide 25 also increased activation of CD4+ effector T cells at the site of infection ( Figure 2B and 2C ) suggested that providing antigen by injection of peptide 25 might also result in improved immune control of infection . We first determined the duration of increased IFN-γ production by adoptively transferred P25TCRTh1 cells or endogenous CD4+ T cells after peptide 25 injection . The frequency of IFN-γ cells was highest in both 6 hours after treatment , and decreased to approximately 20% of maximal levels by 24 hours after peptide injection for both endogenous CD4+ and P25TCRTh1 cells ( Figure 6A ) . By 72 hours post-treatment , the frequency of IFN-γ+ cells returned to levels observed in the absence of peptide 25 injection , indicating that the activating effect of peptide 25 treatment is remarkably transient , entirely dissipating within 3 days of the treatment . Despite the transient nature of this effect , we found that treatment of M . tuberculosis H37Rv-infected mice with peptide 25 ( in the absence of adoptively transferred P25TCRTh1 cells ) every 2–3 days from day 28 to day 45 post-infection reduced lung bacterial burdens by 1 . 05±0 . 40×106 bacteria ( p = 0 . 018 ) compared with that in mice treated with OVA peptide , an unrelated MHC II epitope ( Figure 6B ) . Neither group of mice displayed any signs of toxicity , even after repeated peptide injections . These results indicate that during M . tuberculosis infection , CD4+ effector T cells are not stimulated at their maximum potential frequency at the site of infection in the lungs . Because effector T cell responses progressively decrease during chronic infection , and enhancing T cell responses with exogenous peptide antigen improves immune clearance of M . tuberculosis , we conclude that failure to optimally activate effector T cells at the site of infection is an important determinant of the limited efficacy of adaptive immunity in tuberculosis .
M . tuberculosis evades adaptive immunity to persist in the lungs , often for the lifetime of the host . Here , we have characterized one mechanism by which this impressive feat of immune evasion is accomplished in vivo . We found that , of the large number of CD4+ effector T cells recruited to the lungs of infected mice , few are stimulated to produce IFN-γ ( Figure 7A ) . While there are few precedents available for comparison , our findings are in stark contrast to those found in C57BL/6 mice infected with the Armstrong strain of LCMV [28] . In that context , which results in CD8+ T cell-dependent resolution of infection , >20% of virus-specific CD8+ T cells are activated to produce IFN-γ during the acute stage of infection when viral burdens and antigen availability are highest , and the frequency of in vivo-activated virus-specific CD8+ T cells does not decrease until the viral burden is reduced . We found that the initially low proportion of CD4+ T cells producing IFN-γ in the lungs of M . tuberculosis-infected mice diminishes further as infection progresses to chronic phase , even though the bacterial burden in the lungs remains high . Our studies using adoptively transferred Ag85B-specific P25TCRTh1 cells revealed that the decreasing responses of CD4+ effector cells are caused in part by decreasing expression of fbpB by M . tuberculosis . By reducing fbpB expression during chronic infection , M . tuberculosis restricts the availability of Ag85B , an immunodominant antigen , and thereby prevents infected APCs from optimally activating CD4+ effector T cells . Consistent with this model , we found that a recombinant strain of M . tuberculosis engineered to maintain the expression of fbpB at high levels during chronic infection ( CPE85B ) was attenuated during the chronic phase of infection in a strictly CD4+ T cell dependent manner , indicating that down-regulation of fbpB and limitation of antigen availability is important for evasion of adaptive immunity by M . tuberculosis . Treatment of infected mice with synthetic Ag85B peptide 25 also increased CD4+ effector T cell IFN-γ responses and significantly reduced the bacterial burden in the lungs . We conclude that suboptimal effector T cell activation enables M . tuberculosis to evade elimination by adaptive immunity during the chronic stage of infection , and that some of this suboptimal effector T cell activation is attributable to restricted antigen expression by the bacteria . In addition , other mechanisms that limit effector T cell activation , such as interference with the MHC class II antigen processing and presentation pathway and/or the action of regulatory T cells , likely contribute to the remarkable survival of M . tuberculosis in vivo . Infection with M . tuberculosis induces a robust T cell response involving CD4+ and CD8+ T cells and the effector cytokines IFN-γ and TNF [3] , which are all essential for control of infection [5] , [39] , yet adaptive immunity fails to eradicate M . tuberculosis . Mechanisms for the limited efficacy of the adaptive immune response in tuberculosis fall into two general ( not mutually exclusive ) categories: either the effector functions that T cells perform ( e . g . IFN-γ production ) are not effective because of failed responses by the infected cells targeted by effector T cells; or the T cells recruited to the site of infection do not optimally perform the effector functions required for immune clearance . Regarding the former , the ability of M . tuberculosis to resist and inhibit the TNF- and IFN-γ-induced microbicidal responses of the phagocytic cells it infects is one documented component of its immune evasion strategy in vivo [40] . However , our observation that only a small fraction of the CD4+ effector T cells in the lungs is activated to synthesize IFN-γ provides new support for the latter explanation . The potential causes of this mechanism include bacterial factors and host regulatory mechanisms that directly impair effector T cell function . As an example of a direct bacterial effect , mycobacterial cell wall glycolipids have been found to impair CD4+ T cell responses in vitro [41] . With regard to host regulatory mechanisms , during mouse infection , T regulatory cells limit the ability of adaptive immunity to restrict the bacterial population size in the lungs [11] , [42] . Interleukin-10 ( IL-10 ) , whether expressed by myeloid cells or T cells , provides an additional host regulatory mechanism that inhibits T cell effector functions in tuberculosis , as transgenic over-expression of IL-10 in infected mice impaired T cell responses and caused an increase in bacterial CFUs [12] , while deletion of IL-10 causes enhanced control of infection [43] , indicating that T cell-directed suppressive factors can limit the success of the adaptive immune response to M . tuberculosis . On the other hand , CD4+ effector T cells at the site of infection may not recognize or become activated optimally by APCs bearing M . tuberculosis-derived peptide:MHC II complexes , a process that is required for IFN-γ production in peripheral tissues [35] . Recent observations using live imaging revealed that a small fraction of Leishmania major-infected macrophages interact with Leishmania-specific CD4+ T cells in vivo [44] indicating that in certain infections , effector T cells may not recognize infected cells efficiently , and this may contribute to slow clearance or persistence of infection . Suboptimal stimulation of CD4+ T cells could occur via direct targeting and inhibition of MHC II antigen presentation pathways in infected APCs , or as a result of the limited availability of peptide T cell epitopes , a consequence of bacterial suppression of antigen encoding genes , or a combination of these mechanisms . In this study , we first determined that the frequency of endogenous polyclonal CD4+ T cells producing IFN-γ in the lungs was surprisingly low , and varied during the course of infection , with the highest responses during the acute stage and the lowest responses observed as infection reached the chronic stage . These reduced responses occur despite the presence of similar numbers of bacteria in the lungs during these stages of infection . To further understand the underlying mechanisms of the low frequency of effector T cell activation in the lungs , we quantitated CD4+ effector T cell responses to the peptide 25 epitope of M . tuberculosis Ag85B , a secreted protein targeted by a large number of M . tuberculosis-specific CD4+ T cells [45] . Ag85B is targeted by 5 of the 9 novel tuberculosis vaccine candidates currently in clinical trials [46] , thus understanding its behavior and responses to it in vivo has considerable importance for TB vaccine development . The reduced expression of fbpB we observed is consistent with regulation by the state of bacterial growth , though it may be indirectly triggered by the onset of Th1 immunity , since expression of fbpB is maintained in mice lacking IFN-γ [38] . Because Ag85B is a cell wall biosynthesis enzyme , down-regulation of fbpB has been interpreted as a consequence of transition by M . tuberculosis into a relatively stationary state . Alternatively , fbpB suppression during chronic infection may also be an evolved bacterial immune evasion mechanism that enables long-term persistence of M . tuberculosis by limiting T cell activation . In support of this , we found that forced expression of fbpB by the CPE85B strain during chronic infection resulted in a higher proportion of P25TCRTh1 cells producing IFN-γ than in H37Rv-infected mice . Other studies have suggested but not directly examined the possibility that over-expression of certain M . tuberculosis proteins ( including Hsp70 and ESAT-6 ) may cause attenuation of bacterial persistence by increased immune recognition [47] , [48] . Our finding that polyclonal CD4+ effector T cell responses diminish in chronic infection suggests that this may be a general phenomenon in tuberculosis . Importantly though , the higher frequency of P25TCRTh1 cell activation observed in CPE85B-infected mice diminished at a later time point as it did in H37Rv infection , implying that other mechanisms , especially impairment of MHC II antigen presentation by M . tuberculosis , exist to limit effector T cell activation during chronic infection in vivo . Several in vitro studies have found that M . tuberculosis subverts or impairs antigen presentation by the cells it infects , limiting the capability of infected APCs to activate antigen specific T cells [8] , [10] . Initial observations include the finding that M . bovis BCG survives in primary human macrophages that CD4+ T cells fail to recognize [26] and that M . tuberculosis-infected THP-1 cells express low amounts of surface MHC II [25] . Several mechanisms for inhibition of MHC II antigen presentation have been characterized using a spectrum of mycobacterial strains and cell components . Among these , impaired phagosome maturation , a well-characterized component of the ability of M . tuberculosis to survive in phagocytic cells [19] , has been found to limit activation of cathepsin D for efficient processing of mycobacterial antigens [24] , while inducing autophagy with rapamycin was recently found to improve the efficacy of BCG and other live mycobacterial vaccines , by enhancing presentation of mycobacterial antigens [49] . Impaired expression of MHC II by macrophages after IFN-γ treatment was also observed after in vitro infection or treatment of macrophages with certain mycobacterial cell components [23] , [50] , [51] , [52] , [53] . This effect may involve prolonged signals received through bacterial pattern recognition receptors ( PRRs ) including TLR2 , although we recently reported a TLR2 independent mechanism for impaired MHC II expression in response to IFN-γ [53] , [54] . These and other in vitro studies are consistent with our present results and lend support for the hypothesis that APCs do not efficiently stimulate CD4+ effector T cells in the lungs during M . tuberculosis infection in vivo . Attempts to verify and explore the significance of these in vitro findings with in vivo infection models have been limited thus far , until the present paper . One study of mouse infection with GFP-expressing M . bovis BCG found a modest decrease in surface expression of MHC II on some populations of lung APC that harbored intracellular bacteria when compared to those that did not contain bacteria [55] . In contrast , in a low dose aerosol infection of mice with GFP-expressing H37Rv , we did not detect a difference in surface MHC II expression between infected and non-infected APCs at various time points post-infection; we also found that M . tuberculosis-infected APCs isolated from the lungs expressed high levels of the costimulatory molecules CD80 and CD86 [54] . Nonetheless , there is evidence that the activation of M . tuberculosis-specific T cell responses is impaired during in vivo infection , indicating that M . tuberculosis may specifically impair presentation of its antigens without decreasing overall surface expression of MHC II . One recent study found that mice provided with CD4+ TCR-transgenic effector T cells specific for the M . tuberculosis antigen ESAT-6 prior to infection can restrict bacterial population size to a lower level but cannot prevent establishment of infection [56] . Despite the presence of this effector T cell population in the lungs from the onset of infection , control of bacterial growth was delayed until 7 days post-infection . Likewise , despite mounting apparently normal anti-M . tuberculosis CD4+ T cell responses , infected mice and humans treated with anti-mycobacterial drugs to eliminate primary infection remain susceptible to reinfection [33] , [57] . These studies indicate that susceptibility to persistent tuberculosis is more likely due to failure to activate antigen-specific effector T cells , rather than to insufficient development of antigen specific T cells in response to infection . We observed increased survival of wild type , but not CD4+ T cell-deficient mice infected with the CPE85B strain when compared to those infected with H37Rv , highlighting the importance of enhanced T cell stimulation to the long-term outcome of infection , and indicating that enhanced effector T cell activation , through increased antigen availability , can be accomplished without detrimental effects . Moreover , our finding that sustained expression of Ag85B during the adaptive immune phase of infection was associated with a 2- to 5-fold increase in antigen-specific CD4+ T cell activation , yet reduced the bacterial burdens approximately 10-fold implies that a massive increase in effector T cell activation is not necessary to significantly improve immune control of tuberculosis . Future efforts to develop tuberculosis therapies should therefore aim to bypass or overcome factors that limit effector T cell activation including direct T cell suppression , impaired antigen presentation , and bacterial gene regulatory mechanisms . For example , we found that the chronic phase antigen deficit resulting from bacterial suppression of fbpB could be overcome by systemic treatment of infected mice with synthetic peptide 25 , which strongly but transiently enhanced CD4+ T cell responses specific for this epitope and reduced the bacterial burden . This result implies that the endogenous CD4+ T cells generated in response to infection with M . tuberculosis and recruited to the infected lungs can be stimulated to perform their effector functions if they are provided antigen , resulting in improved bacterial clearance ( Figure 7B ) . The potential for anti-tuberculosis therapies that aim to enhance existing T effector cell responses in infected individuals with synthetically produced peptides encoding known T cell epitopes remains unexplored; however , given the steadily increasing prevalence of drug resistant M . tuberculosis , such immunotherapeutic approaches to tuberculosis are an attractive option . Although the consequences of increasing the activation of existing T cell responses have not been widely tested , in the context of certain highly monoclonal T cell responses , administration of epitope peptides has caused rapid mortality of infected or previously immunized mice [58] , [59] . However , despite these findings and concerns about possible immunopathology induced by hyperactivation of effector T cells in tuberculosis [60] , we observed no morbidity or mortality in infected mice repeatedly treated with peptide 25 , a result that encourages the continued exploration of this therapeutic strategy . Future studies should also aim to determine the host and bacterial regulatory mechanisms that account for chronic phase suppression of fbpB and whether genes encoding other immunodominant M . tuberculosis antigens behave similarly . Identification of the elements of this host-pathogen interaction may lead to the development of therapies that target antigen gene suppression and inhibition of antigen presentation and provide a novel strategy for overcoming bacterial persistence in vivo , leading to better outcomes in M . tuberculosis-infected individuals .
C57BL/6 , B6 . SJL-Ptprca Pepcb/BoyJ ( CD45 . 1+ ) , and MHCII KO mice for aerosol M . tuberculosis infection experiments were either bred in the New York University School of Medicine Skirball animal facility or purchased from Taconic Farms , Inc . P25TCR-Tg mice , whose CD4+ T cells express a transgenic T-cell antigen receptor that recognizes the complex of peptide 25 ( aa 240–254 ) of M . tuberculosis Ag85B and the mouse MHC II allele I-Ab were prepared on a C57BL/6 background , as previously described [2] , [61] . All animal experiments were done in accordance with procedures approved by the NYU School of Medicine Institutional Animal Care and Use Committee and in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health under the Assurance of Compliance Number A3435-01 . Wild type M . tuberculosis H37Rv was originally obtained from ATCC . Frozen stocks for aerosol infection and in vitro use were prepared and stored at −80°C . GFP-expressing H37Rv and Ag85B null ( ΔAg85B ) strains of M . tuberculosis were generated as previously described [2] , [62] . M . tuberculosis cultures were grown in 10 mL Middlebrook 7H9 liquid medium supplemented with 10% v/v albumin dextrose catalase enrichment and incubated under shaking conditions at 37°C . Mice at 8–12 weeks of age were infected with ∼100 CFU of M . tuberculosis via the aerosol route using an Inhalation Exposure Unit ( Glas-Col ) as previously described [62] . To verify inoculum size , 3–5 infected mice were euthanized 24 hours after infection and lungs were homogenized and plated on Middlebrook 7H11 medium supplemented with 10% v/v albumin dextrose catalase enrichment . To determine bacterial population size at time points post-infection , lungs were homogenized , diluted in PBS+Tween-80 ( 0 . 5% ) , and added to 7H11 plates . Plates were incubated at 37°C for 3 weeks and single colonies were counted . To determine M . tuberculosis survival in stationary culture , 7H9 medium was inoculated with H37Rv or CPE85B , grown in shaking conditions to saturation ( O . D . 600>1 . 0 ) , and initial CFUs were measured . Cultures were then placed in stationary incubator at 37°C for 17 days , and final CFUs were measured . C57BL/6 mice were infected with M . tuberculosis H37Rv and on day 25 post-infection received 1×106 CFP+ P25TCRTh1 cells via adoptive transfer . On 28 post-infection , lungs were perfused and frozen in OCT before 5 µm sectioning and fixation in cold acetone . Sections were stained with DAPI to label nuclei and analyzed on a Leica DMRB fluorescent microscope ( objective: Leica PL Fluotar 20×/0 . 50 ) equipped with a Spot RT digital camera . Separate images for DAPI and CFP fluorescence were acquired and merged using Spot software . P25 TCR-Tg CD4+ Th1 effector cells were generated in vitro as follows: naïve CD4+ T cells were magnetically isolated from lymph node cell suspensions of P25 TCR-Tg mice ( or for fluorescent microscopy , a P25TCR-Tg mouse expressing CFP under control of the ubiquitin promoter ) using CD4 ( L3T4 ) microbeads and an AutoMACS ( Miltenyi Biotech ) . P25TCR-Tg CD4+ T cells were co-cultured with irradiated C57BL/6 splenocytes in the presence of mouse IL-12p70 ( 10 ng/ml ) , mouse IL-2 ( 5 ng/ml ) , anti-IL-4 neutralizing antibody ( 50 ng/ml ) , and synthetic peptide 25 ( 0 . 5 µM ) . Cells were cultured at 37°C with 5% CO2 . On days 3 and 5 of culture , cells were split 1∶3 with fresh media containing IL-12p70 , IL-2 , and anti-IL-4 , but no peptide 25 . Cells were washed with PBS and counted on day 7 of culture before use for in vitro or in vivo assays . For in vitro restimulation , P25TCRTh1 cells were co-cultured with irradiated C57BL/6 splenocytes for 24 hours in RPMI-10 in the presence or absence of peptide 25 ( 0 . 5 µM ) or bone marrow derived dendritic cells infected with M . tuberculosis ( MOI: 0 . 1 ) . Cells were collected and analyzed by flow cytometry for intracellular IFN-γ , or culture supernatants were analyzed for IFN-γ by ELISA . For in vivo experiments , 1×106 P25TCRTh1 cells were injected via tail vein or retro-orbital sinus into recipient mice at various time points post-infection . Cells were routinely isolated from lungs of recipient mice 72 hours after adoptive transfer and analyzed by flow cytometry . 3×106 CFSE-labeled CD4+ T cells , harvested from the lymph nodes of P25TCR-Tg mice were adoptively transferred into infected recipients at various time points post-infection . 7 days after adoptive transfer , mediastinal lymph nodes were harvested from recipient mice and cells were analyzed for CFSE dilution by flow cytometry . The Ag85B null strain of M . tuberculosis ( ΔAg85B ) , previously created by our lab from wild-type H37Rv [2] , was used as a background strain for generating CPE85B . Both the hspX promoter sequence , consisting of 254 bp directly 5′ of the hspX start codon , as well as the fbpB open reading frame were amplified by PCR from H37Rv genomic DNA . Each of these fragments was ligated into the pMV306 integrating vector to create a recombinant construct , whose sequence was verified by Sanger sequencing performed by the NYU DNA sequencing facility . ΔAg85B was grown in 7H9 liquid media and transformed with this construct via electroporation . The reaction was plated on 7H11 plates containing 25 µg/ml kanamycin to select for bacteria incorporating the construct into the M . tuberculosis chromosome . Presence of the construct in kanamycin resistant colonies was verified by PCR . Expression and secretion of Ag85B by CPE85B was confirmed by SDS-PAGE and anti-Ag85B western blot of supernatants from 7H9 liquid medium after stationary culture . For stationary culture-induced expression of Ag85B by the CPE85B strain , 10 mL cultures were grown to late phase ( OD600∼1 . 0 ) in normal shaking conditions , then flasks were sealed and transferred to a stationary incubator for >1 week before supernatants were collected . To quantitate expression of M . tuberculosis genes during mouse infection , lungs of infected mice were rapidly placed into a solution of RNAlater ( Ambion ) and stored overnight at room temperature in accordance with manufacturer recommendations to allow permeation of the tissue . Thereafter , samples for RNA isolation were stored at −80°C . When comparing expression of genes at various time points , tissues were transferred to TRIzol ( Invitrogen ) and quickly homogenized using a Tissue Tearor homogenizer to disrupt mouse cells . Lung homogenates were centrifuged to pellet intact bacterial cells , and supernatants discarded . M . tuberculosis pellets were disrupted with zirconia/silica beads , RNA was extracted , and RT-qPCR was carried out as previously described [37] with fbpB copy number normalized to the constitutively expressed 16S rRNA and multiplied by a factor of 105 . The following RT-qPCR primers were used in this study . 16S rRNA: RT 5-ATTACGTGCTGGCAACATGA-3 , qPCR For 5-GCCGTAAACGGTGGGTACTA-3 , qPCR Rev 5-TGCATGTCAAACCCAGGTAA-3; hspx/acr/Rv2031c: RT 5-GAATGCCCTTGTCGTAGGTG-3 , qPCR For 5-AGATGAAAGAGGGGCGCTAC3 , qPCR Rev 5-TAATGTCGTCCTCGTCAGCA3; fbpB/Rv1886c: RT 5-TCCTGGAACTTCAGGTTGCT-3 , qPCR For 5-ACCCCCAGCAGTTCATCTAC-3 , qPCR Rev 5-TTCCCGCAATAAACCCATAG-3 . To isolate cells from infected tissues for flow cytometry , mice were euthanized with CO2 followed by cervical dislocation . Tissues were removed and mechanically disrupted by mincing in RPMI as previously described [62] or using a gentleMACS dissociator ( Miltenyi Biotec ) in the manufacturer-recommended HEPES buffer . Lung suspensions were incubated in Collagenase D and DNase at 37°C with 5% CO2 for 30 minutes and cells were isolated by forcing suspensions through a 70 µM cell strainer . RBCs were removed by ACK lysis and live cells counted by trypan blue exclusion . Cell suspensions were stained using the following fluorescently-labeled antibodies ( Biolegend , BD Pharmingen , or eBioscience ) : anti-CD3 PE , anti-CD4 ( L3T4 ) FITC , anti-CD45 . 2 PerCP , anti-CD45 . 1 Pacific Blue , anti-IFN-γ ( XMG1 . 2 ) APC , and rat IgG1 APC isotype control . Flow cytometry was performed using a FACSCalibur or LSR II ( BD Biosciences ) at the NYU Cancer Institute Flow Cytometry and Cell Sorting facility . Analysis of flow cytometry data was performed using FlowJo software . To detect intracellular IFN-γ produced by cells in vivo , a protocol was developed based on a previous study [28] . In contrast to this study , however , optimal detection of IFN-γ producing cells from the lungs of mice infected with M . tuberculosis did not require treatment of mice with i . v . brefeldin A or inclusion of brefeldin A in tissue processing buffers . Instead , after euthanasia , tissues were rapidly placed on ice and all cell isolation steps except collagenase/DNase digestion ( 37°C for 30 minutes ) and ACK lysis ( room temperature for 5 minutes ) were carried out quickly and on ice . Cells were stained for surface markers at 4°C for 30 minutes followed by permeabilization and fixation with Cytofix/Cytoperm ( BD Biosciences ) at 4°C for 20 minutes . Finally , fixed cells were stained with anti-IFN-γ or a rat IgG1 isotype control at 4°C for 30 minutes . Flow cytometry dot plot gates for IFN-γ+ cells were set based on comparison with isotype control and unpermeabilized cells stained for IFN-γ . Mice were treated with an intra-peritoneal dose of 500 µg of either monoclonal antibody GK1 . 5 , which depletes CD4+ T cells , or a rat IgG2b isotype control ( LTF-2 ) every 6 days from day 28 to Day 50 post-infection . Efficiency of CD4+ T cell depletion 6 days after GK1 . 5 treatment was determined to be >95% by flow cytometry of cell suspensions from lungs , spleen and blood . In mice treated with LTF-2 isotype control , no differences were observed in CD4+ T cell number or bacterial burden when compared to untreated mice . To determine the influence of endogenous CD4+ T cells on the response of adoptively transferred P25TCRTh1 cells in vivo , a system was developed to deplete endogenous CD4+ T cells selectively from infected mice . Mice expressing Cre recombinase under control of the CD4 promoter were crossed with those carrying an inducible Diphtheria Toxin Receptor ( iDTR ) allele , whose baseline expression is prevented by a stop codon flanked by loxp sites [34] . Progeny of this cross ( CD4-DTR ) carry CD4+ T cells that are sensitive to Diphtheria Toxin mediated ablation . CD4-DTR mice were infected with H37Rv and received daily intraperitoneal doses of DT ( 100 ng ) to ablate endogenous CD4+ T cells from day 21 to day 28 post-infection . The efficiency of CD4+ T cell ablation in the lungs was determined by flow cytometry to be 48 . 9% . P25TCRTh1 cells were adoptively transferred on day 25 post-infection and the frequency of IFN-γ production was assessed on day 28 post-infection . On day 25 post-infection , wild-type mice infected with M . tuberculosis H37Rv received P25TCRTh1 cells via adoptive transfer . On day 28 post-infection , mice were treated intravenously with 800 ng ( at 4 . 0 ng/µL ) PerCP-labeled anti-CD4 ( RM4-5 ) . Fifteen minutes later , mice were euthanized and total lung cells were stained with FITC-labeled anti-CD4 ( GK1 . 5 ) . Lung cells stained by anti-CD4-PerCP were considered to be CD4+ T cells residing in the intravascular compartment at the time of antibody injection . Cells staining positive for anti-CD4-FITC and negative for PerCP were considered to be CD4+ T cells residing in an extravascular or parenchymal lung compartment protected from labeling with intravenous antibody . IFN-γ production in vivo was assessed by intracellular staining of all cells with APC-labeled anti-IFN-γ as previously described . Mice were intravenously treated with 100 µg of Ag85B peptide 25 ( FQDAYNAAGGHNAVF ) or OVA peptide control ( ISQAVHAAHAEINEAGR ) in 100 µl sterile PBS via tail vein or retro-orbital sinus . Peptides were synthesized by EZBiolab or Peptides International to a purity of >95% . Data shown are representative of 2 or more experimental replicates . In all figures , error bars indicate mean ± SEM . To determine statistical significance when comparing experimental values from two groups of mice , one- or two-tailed student's t-tests were routinely used , each where appropriate . To compare the growth rate of H37Rv and CPE85B in vivo , a non-linear regression analysis ( curve fit ) with F-test was used to determine whether a single curve could account for both data sets . In mouse survival experiments , Logrank test was used to evaluate statistical significance when comparing survival of one mouse strain after infection with either of the two bacterial strains . * = p<0 . 05; ** = p<0 . 005; n . s = not significant .
|
Mycobacterium tuberculosis causes persistent infection even in human or animal hosts that develop antigen-specific CD4+ and CD8+ T cell responses . To understand this phenomenon , we tested the hypothesis that the CD4+ effector T cells that are generated in response to M . tuberculosis infection fail to encounter their antigens at the site of infection in the lungs . Using mice infected with M . tuberculosis , and an assay of in vivo antigen-dependent activation of CD4+ T cells , we found that both polyclonal CD4+ and T cell receptor transgenic CD4+ T cells specific for antigen 85B peptide 25 are activated at low frequencies in the lungs . We found that this is due in part to downregulation of antigen gene expression by M . tuberculosis , as forced expression of the antigen gene resulted in higher frequency activation of CD4+ T cells , as well as CD4+ T cell-dependent reduction in bacterial burdens and prolonged survival of infected mice . We also found that administration of antigen 85B peptide 25 , which is recognized by a high proportion of M . tuberculosis-specific CD4+ T cells , reduced the bacterial burden in the lungs , indicating that stimulation of existing antigen-specific CD4+ T cells may be a promising approach to therapy of TB .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"immune",
"cells",
"immune",
"activation",
"antigen-presenting",
"cells",
"immunity",
"to",
"infections",
"immunology",
"microbiology",
"bacterial",
"diseases",
"adaptive",
"immunity",
"immune",
"defense",
"immunomodulation",
"immunotherapy",
"infectious",
"diseases",
"major",
"histocompatibility",
"complex",
"tuberculosis",
"t",
"cells",
"biology",
"immune",
"response",
"immune",
"system",
"antigen",
"processing",
"and",
"recognition",
"immunity"
] |
2011
|
Suboptimal Activation of Antigen-Specific CD4+
Effector Cells Enables Persistence of M. tuberculosis In
Vivo
|
The Philippines has a population of approximately 103 million people , of which 6 . 7 million live in schistosomiasis-endemic areas with 1 . 8 million people being at risk of infection with Schistosoma japonicum . Although the country-wide prevalence of schistosomiasis japonica in the Philippines is relatively low , the prevalence of schistosomiasis can be high , approaching 65% in some endemic areas . Of the currently available microscopy-based diagnostic techniques for detecting schistosome infections in the Philippines and elsewhere , most exhibit varying diagnostic performances , with the Kato-Katz ( KK ) method having particularly poor sensitivity for detecting low intensity infections . This suggests that the actual prevalence of schistosomiasis japonica may be much higher than previous reports have indicated . Six barangay ( villages ) were selected to determine the prevalence of S . japonicum in humans in the municipality of Palapag , Northern Samar . Fecal samples were collected from 560 humans and examined by the KK method and a validated real-time PCR ( qPCR ) assay . A high S . japonicum prevalence ( 90 . 2% ) was revealed using qPCR whereas the KK method indicated a lower prevalence ( 22 . 9% ) . The geometric mean eggs per gram ( GMEPG ) determined by the qPCR was 36 . 5 and 11 . 5 by the KK . These results , particularly those obtained by the qPCR , indicate that the prevalence of schistosomiasis in this region of the Philippines is much higher than historically reported . Despite being more expensive , qPCR can complement the KK procedure , particularly for surveillance and monitoring of areas where extensive schistosomiasis control has led to low prevalence and intensity infections and where schistosomiasis elimination is on the horizon , as for example in southern China .
Schistosoma japonicum is the causative agent of intestinal schistosomiasis in the Philippines , China and parts of Indonesia . In the Philippines , 10 out of 16 regions have reported cases of schistosomiasis , with an estimated 6 . 7 million people living in endemic areas [1] . Of these , 1 . 8 million , nearly 2% of the total Philippine population ( 105 million ) [2] , are considered to be directly exposed to infection through farming , fishing and other essential occupational and domestic activities involving regular water contact [3 , 4] . The spread of schistosomiasis japonica requires the presence of the intermediate aquatic snail host , Oncomelania hupensis quadrasi , appropriate mammalian definitive hosts and specific environmental conditions to support transmission . Schistosomiasis is a focal disease and in some Filipino communities the prevalence of infection has been reported to be as high as 65% [5–9] . Common diagnostic techniques , such as the microscopy-based Kato-Katz ( KK ) method , have been shown to have low sensitivity for detecting schistosome infections [10–15] . The low sensitivity of KK means that low intensity infections are missed and individuals harboring schistosomiasis may not be treated , thereby contributing to ongoing transmission that may result in re-emergence in areas thought to be clear of schistosomiasis [16] . In China , for example , after decades of mass drug administration ( MDA ) with praziquantel ( PZQ ) and other integrated control measures , the intensity of S . japonicum infection is now quite low in many endemic locations , and therefore the parasite is likely to be missed using KK . It has been shown in China that , after successful control , with significant reduction in prevalence and intensity , if control efforts lapse , there can be re-emergence in transmission and infection [16] . This may be due in part to the KK procedure missing low intensity infections and a more sensitive approach , such as the use of molecular diagnostics , may be required to determine whether , for a particular area , elimination has been achieved . Molecular methods are playing an increasingly important role in schistosome diagnosis and identification due to their increased performance when compared with traditional copro-parasitological techniques [13 , 17–19] . A number of molecular techniques have been used for the detection of schistosome infections including conventional PCR ( cPCR ) , real-time PCR ( qPCR ) and loop-mediated isothermal amplification ( LAMP ) [20–25] . We have previously reported a pilot study examining 50 humans for the presence of S . japonicum infections in Western Samar , the Philippines , comparing the KK , cPCR and qPCR methods which showed 30 . 8% , 75 . 0% and 92 . 3% prevalence , respectively [26] . Here we substantially expand the scope of the earlier findings , to further assess the prevalence of S . japonicum in humans in Samar Province; and evaluate on a larger scale the diagnostic sensitivity of qPCR compared to KK . We highlight the need for more sensitive diagnostic procedures , particularly in areas where the intensity of infection is low , and discuss the possible application of qPCR in current control efforts
Informed written consent was received from all human participants in the study and ethical approval was provided by the Ethics Committee of the Research Institute of Tropical Medicine ( RITM ) , Manila , and the Queensland Institute of Medical Research ( QIMR ) Human Research Ethics Committee ( Approval Number: H0309-058 ( P524 ) ) . Meetings were held in each barangay prior to the start of the study to inform the population about the research , and to answer any questions on the procedures to be carried out . Local medical staff then visited each household over the subsequent week with a consent form ( signed by all participants or by a parent or legal guardian for minors before the start of the study ) for each participant as well as household and individual questionnaire forms , and to further explain the study to individuals . Individual questionnaires ( comprising information on their demographics; age , gender and barangay ) were completed by each household member , or by the head of the household in the case of minors . Individual identification numbers ( IDN ) were assigned from the individual questionnaires . Study participants were aged between 4 and 82 years of age . Participants who were positive by KK for S . japonicum were treated with PZQ using the World Health Organisation ( WHO ) recommended clinical dosage ( 40 mg/Kg ) . Local and government health officials were informed of those subjects positive for S . japonicum by qPCR so that follow up treatment could be provided . This study is part of a larger , UBS Optimus Foundation ( Switzerland ) funded project , where stool samples were sought from all residents ≥ 4 years across the six barangays ( Fig . 1 ) . Inclusion into this study was based on having a completed consent form and submitting at least one stool sample for analysis . A cross-sectional survey was carried out in six barangays in the municipality of Palapag , Northern Samar Province , the Philippines in 2011 ( Fig . 2 ) , to determine the human S . japonicum prevalence and infection intensity using both microscopic ( KK ) and molecular ( qPCR ) diagnostic techniques . Diagnostic performances of the two techniques were also assessed through sensitivity and specificity calculations . A barangay , the smallest administrative division in the Philippines , is the native Filipino term for a village . The study was undertaken in six barangays: Napo , Capacujan , Matambag , Mabaras , Magsaysay and Manajao , all located in the municipality of Palapag ( Fig . 2 ) . These barangays have been subjected in the past five years to annual mass treatment programs as part of the national schistosomiasis control program in the Philippines . None of the barangays selected had been treated for schistosomiasis before the onset of the study in 2011 . Sample collection . Individual samples were obtained by handing out labelled ( IDN , name , date ) stool cups which were collected over seven days from study participants from each barangay . Participants submitted a stool sample between 5–10 g . In total two stool samples were sought from each individual on different days . The KK procedure ( ideally performed on 2 stools , 3 slides per stool ) was performed for the diagnosis of S . japonicum . A subset of 100 stool samples from each barangay ( total = 600 ) was selected following KK examination for the qPCR based prevalence assessment presented here . Of the 600 samples satisfying inclusion criteria , 40 samples were excluded due to low DNA quality and quantity after DNA extraction ( Fig . 1 ) . Only 339 ( 60 . 5% ) participants of the selected sub-cohort submitted two stool samples and had a complete slide set of two stools , three KK slides per stool examined . Comparison of diagnostic performance between qPCR and KK was undertaken using samples obtained from these 339 individuals such that the diagnostic sensitivity for KK was standardized across the sub-cohort . Kato-Katz ( KK ) method . The KK was performed on all collected human stool samples . Briefly , individual stool samples were pressed through a thin gauze , the non-retained material was then used to fill a standard volume template ( representing 41 . 7 mg of fecal material ) , sitting on top of a glass slide , on to which the fecal sample was transferred through the template . Cellophane soaked in glycerin-malachite green was then placed over the sample on the glass slide and pressed against a firm surface to spread the stool evenly . The slide was then viewed under a microscope after at least 30 minutes had passed . Three slides were prepared from each stool sample and the slides were read independently by a team of microscopists . Individual microscopists were blinded to the results of the other microscopists . KK was performed by microscopists either from RITM ( Research Institute of Tropical Medicine ) or from local schistosomiasis control units , and had previous training and experience . DNA extraction . For the selected samples , 1–3 g of the remaining stool was then stored in 80% ( v/v ) ethanol at 4oC , for subsequent DNA isolation and molecular analysis at QIMRB , Australia . Genomic DNA was isolated from 200 mg of the stool samples using a QIAamp mini stool kit ( QIAGEN ) following the manufacturer’s protocol . DNA concentrations were determined using a NanoDrop 2000 ( Thermo Scientific ) and all DNA samples were diluted to 50 ng/μl for subsequent analysis . Real-time PCR . Full details of the qPCR assay , utilizing primers which amplify a fragment of the NADH dehydrogenase I ( nad1 ) mitochondrial gene , are available elsewhere [24 , 26 , 27] . Briefly , reaction mixtures of 24 μl were prepared containing 12 . 2 μl SYBR Green ( Invitrogen ) , 5 pM of each primer , 7·8 μl of H2O and 2 μl of DNA ( 50 ng/μl ) template . The PCR cycling conditions were as follows: 2 minutes initialization at 50°C , 10 minutes denaturation at 95°C , followed by 45 cycles of 15 sec denaturation at 95°C , 60 sec annealing at 60°C , 90 sec extension at 72°C and a final dissociation phase at 60–95°C . The PCR was performed using a real time thermocycler ( Corbett RotorGene 6000 ) . Melt curve analysis was performed after each qPCR . The results were quantified as eggs per gram ( EPG ) in the range 1–100 EPG using qPCR cycle threshold ( Ct ) scores [26] . Negative and positive controls were run in each qPCR . Negative controls used distilled H2O instead of template while the positive control was DNA extracted from eggs purified from the livers of infected mice . Seeding and dilution experiments were performed to determine equivalent egg numbers ( qEPG ) relating to Ct scores . Eggs were purified from the livers of mice experimentally infected with S . japonicum [28] . Negative control human and laboratory mice stool samples were seeded with a known number of eggs and DNA extracted from these seeded samples . A standard curve using the dilutions and the results of the seeding experiments was then created to determine a range of Ct scores which corresponded to a known number of eggs . This curve was then used to quantify the qEPG in the collected human stool samples from Palapag . Based on the results of the seeding and dilution experiments a Ct of 23 . 0 was determined as the cut off for a positive result . Validation of the qPCR by microscopy . A random selection of 20 human stool samples that were: qPCR-positive and KK-negative; qPCR- and KK-positive; or qPCR- and KK-negative were re-examined by microscopy after processing , using a modification of the formalin-ethyl acetate sedimentation-digestion ( FEA-SD ) technique developed recently for the precise quantification of S . japonicum eggs in bovine stool [26 , 29] . The remaining volume of stool not used for DNA extraction was used for the microscopic examination . The amount remaining varied between samples . Between 0 . 6–2 . 5 g of faeces ( depending on how much of the faecal sample remained ) were sieved through a 250 μm ( pore opening size ) nylon mesh directly onto a 38 μm nylon mesh , using water to wash the sample through the mesh and sieving until the water running out of the 38 μm mesh was clear . The remaining sediment on the 38 μm mesh was then washed into a 15 ml tube and centrifuged at 2000 rpm for 5 minutes to settle the contents . The supernatant was removed , 10% ( v/v ) formalin was added to a volume of 5 ml , followed by a further 5ml of 10% ( w/v ) potassium hydroxide solution . The tube was then vortexed thoroughly and left on a shaker at 37°C overnight . The tube was then vortexed again , the sample was centrifuged at 500 g for 10 min , the supernatant was removed and the pellet washed once with water . The water was removed and the final pellet resuspended in 1–2 . 5 ml of water , depending on the size of the pellet and the precise volume recorded . The suspension was shaken thoroughly; two 200 μl samples were spread on glass microscope slides and examined microscopically using an inverted microscope . Statistical analyses . Microsoft Excel ( Microsoft , Silicon Valley , LA , 2010 ) and SAS ( SAS Institute , Cary , NC , version 9 . 3 ) software were used for data analyses . A sample was considered positive if at least one S . japonicum egg on any KK slide; or if a positive Ct score was seen ( Ct score greater than 22 . 00 was considered negative , less than 21 . 99 was considered as positive ) by qPCR . Egg counts from the KK and qEPG from the qPCR were transformed to eggs per gram and geometric mean eggs per gram ( GMEPG ) in infected stool samples calculated by using the log-transformed egg counts . Confidence limits were calculated using standard formulae based on the binomial distribution ( prevalence ) and the lognormal distribution ( infection intensity ) . Relative diagnostic sensitivity and specificity for the KK were calculated using a subset of samples for which there was a full complement of slides ( two stools , three slides per stool N = 339 ) . Prevalence for KK was calculated from the full 560 individuals , including those with only one stool ( three slides ) and those with two stool ( 6 slides ) . Relative sensitivity and specificity of the KK compared to qPCR was calculated using the qPCR as the reference standard . The qPCR was used as the reference standard as we predict this to be a better diagnostic method . Microscope readers for the KK were blinded to the results of other readers . The qPCR was performed independently from the KK and the results compared only after the qPCR had been completed .
A subset of 3810 individuals ( Fig . 1 ) were selected for study in a larger UBS Optimus Foundation project . Of these , 69 had missing demographic data of either age or gender and were excluded . A further 331 individuals did not submit any stool samples and were also excluded ( N = 3410 ) . Of those remaining , 2005 individuals submitted two stool samples and 1405 submitted one stool sample ( Fig . 1 ) . KK was performed on each stool provided ( 3 slides per stool ) . From the remaining individuals , 600 were selected for qPCR analysis and DNA was extracted using the Qiagen kit . Of these , 40 were excluded due to low DNA quality and quantity so that the qPCR was undertaken on 560 subjects ( Fig . 1 ) . Only 339 ( 60 . 5% ) participants of the selected sub-cohort submitted two stool samples and had a complete slide set of two stools , three KK slides per stool examined . Prevalence was determined on all samples from 560 individuals . Comparison of diagnostic performance between qPCR and KK was undertaken using samples obtained from these 339 individuals such that the diagnostic sensitivity for KK was standardized across the sub-cohort The overall human S . japonicum prevalence across the six barangays in our study area in Palapag , Northern Samar , the Philippines , was considerably higher when determined by qPCR ( 90 . 2%; 95% CI 87 . 7–92 . 7 ) than by the KK method ( 22 . 9%; 95% CI 19 . 4–26 . 4 ) ( Table 1 ) . Overall infection intensity ( GMEPG ) was also higher when determined by qPCR ( 36 . 6; 95% CI 32 . 0–41 . 8 ) than by the KK ( 11 . 5; 95% CI 9 . 4–13 . 9 ) . Males had a higher prevalence than females , although similar infection intensities were seen using both diagnostic techniques ( Table 1 ) . Barangay prevalence ranged from 84 . 2% ( 76 . 7–91 . 7 ) to 96 . 9% ( 93 . 3–100 ) determined by qPCR; and 18 . 6% ( 10 . 2–27 . 0 ) to 29 . 2% ( 19 . 9–38 . 4 ) by KK . Manajao had the highest prevalence of all the barangays under investigation both by qPCR ( 96 . 9%; 95%CI 93 . 3–100 ) and the KK ( 29 . 2%; 95% CI 19 . 9–38 . 4 ) . The highest infection intensity determined by qPCR was in Napo ( 85 . 3; 95% CI 63 . 6–114 . 3 ) ; although Capacujan had the highest intensity of infection when KK was used ( 16 . 5; 95% CI 9 . 4–29 . 1 ) . The sensitivity of the KK , using qPCR as the reference standard , was 26 . 1% ( 95% CI 21 . 3–31 . 4% ) . Specificity was calculated as 82 . 8% ( 95% CI 64 . 2–94 . 2% ) . A cross tabulation comparing KK and qPCR shows that 7 KK positive samples were negative by qPCR ( Table 2 ) . A total of 48 samples were negative by both techniques and 384 samples were negative by KK but positive by qPCR ( Table 2 ) . The range of qPCR Ct scores was compared to four categories of intensity of infection as determined by EPG ( negative , low , medium and high ) ( Table 3 ) . A total of 9 . 82% ( N = 55 ) of samples were negative ( Ct 22 . 00–37 . 21 ) , 61 . 61% ( N = 345 , Ct 15 . 00–21 . 99 ) classified as low EPG ( N = 123 , <100 EPG ) , 21 . 96% ( N = 37 , Ct 12 . 02–14 . 99 ) medium EPG ( 100–400 epg ) and 6 . 61% ( Ct 7 . 78–11 . 93 ) as high EPG ( >400 EPG ) ( Table 3 ) . Twenty samples were randomly selected for further microscopic examination using a modification of the FEA-SD technique [26 , 29] . After processing using this procedure , S . japonicum eggs were found in 16 samples by microscopy ( Table 4 ) . Two of the microscopy-negative samples were also negative by qPCR and the KK , while one was positive by qPCR but negative by the KK and the fourth was positive by both techniques ( Table 4 ) .
In this study we found a high prevalence of schistosomiasis japonica in humans in the municipality of Palapag in Northern Samar province , the Philippines—higher than previously reported for this region . The high prevalence of S . japonicum mirrored the results of a small pilot study we undertook in September 2010 in Western Samar [26] . The prevalence ( 90 . 2% ) and infection intensity ( 36 . 5 qGMEPG ) values determined by the qPCR assay in this study were considerably higher than those obtained using the KK method ( 22 . 9%; 11 . 5 GMEPG ) . The sensitivity of the KK technique was estimated as 26 . 1% and the specificity as 82 . 8% . It is noteworthy that the KK sensitivity calculated in this study was from individuals submitting two stool samples , with three slides per stool only . The disparity between the two methods used in the current study clearly highlights the need for the development of a more sensitive technique for detecting schistosome infections in humans . The majority ( N = 345 , 61 . 61% ) of qPCR positive samples were low intensity infections . The differences in GMEPG determined by the KK and qPCR , are likely due , in part , to an overestimation of the qPCR GMEPG values . The qPCR is based on an approximation of the quantity of DNA present in one egg and how that corresponds to a Ct score . However the GMEPG obtained using the KK is also an estimate as it assumes a uniform distribution of eggs within the stool and does not take into account any egg clumping which may occur [30] . The qPCR is the more sensitive technique , especially for determining prevalence . However , there is clearly a lack of concordance with intensity of infection calculations when compared with the KK ( Table 4 ) . However samples positive by KK were only rarely negative by qPCR ( Table 2 ) . Both techniques have limitations and the methodology used to determine intensity of infection by qPCR needs to be included in future studies . The large difference in prevalence determined using the qPCR and the KK may be explained by the fact the qPCR may be detecting DNA from adult worms or schistosomula , as well as eggs , from past or pre-patent infections . False positives due to eggs from past infections are unlikely as eggs have been shown to hatch in tissues after PZQ treatment [31] . Eggs that do not hatch due to PZQ treatment are caught in granulomas and do not hatch due to the host immune response [5] . How DNA from adult worms , living in blood vessels , would end up in the feces cannot , at present , be explained . DNA is unlikely to have originated from schistosomula as lung stage larvae would have to enter the alveolar space , be coughed up and swallowed , thereby reaching the gastrointestinal tract . If DNA from other lifecycle stages is being detected , it could account for the much higher prevalence obtained using qPCR compared with the KK . However , similar results were obtained in our pilot study in Samar for both humans and animals [26] , and in an earlier study in Leyte by Wu et al . [27] . In humans there was a large difference in prevalence between the KK and qPCR which was mirrored in bovine infections . However , a comparison of the FEA-SD technique [29] with the qPCR indicated the S . japonicum prevalence in bovines determined by the two techniques was similar . This further highlights the requirement for a more sensitive microscopic technique for diagnosis of schistosomes in human fecal samples . The qPCR is specific for S . japonicum DNA but to confirm and validate the qPCR results generated during this study , a random selection of fecal samples previously analysed by the qPCR were chosen and subjected to a modified version of the diagnostic FEA-SD technique , a method previously developed for the analysis of bovine fecal material [29] . We observed , by microscopy , S . japonicum eggs in all but two of the samples that were positive by qPCR; one of these samples was positive by the KK method . For these two samples , the total weight of stool processed was only 1 . 2 g and 0 . 8 g , respectively , which may explain why no eggs were observed after processing by the modified FEA-SD technique . Morbidity control was implemented in China as part of the World Bank Loan Project ( WBLP ) ( 1992–2001 ) [32] . Recently , re-emergence of schistosomiasis japonica has been recorded in some areas where transmission control had been reported , indicating that low intensity infections can still contribute to schistosomiasis spread [16 , 33 , 34] . Additionally , it has been shown that S . japonicum-induced morbidity may be unrelated to intensity of infection , so that even low intensity infections can result in severe morbidity outcomes [35] . This has particular relevance when considering the KK method , which is well recognized to lack sensitivity , particularly when infection intensity is low [13 , 36 , 37] . By missing low intensity infections the burden of disease in a community may be considerably underestimated . Transmission of S . japonicum is complex due to the large number of animal hosts that can act as reservoirs of infection [38] . Water buffaloes are considered the major reservoir host for schistosomiasis japonica in China [30 , 39–41] , and this may also be the case in the Philippines where we have shown high prevalence in carabao using the same qPCR assay utilized here [26 , 42] . The use of this qPCR-based diagnostic is therefore not limited to detecting S . japonicum infections in humans and may be extremely useful as a research tool when considering transmission control of zoonotic schistosomiasis . However , qPCR has some limitations when considered as a routine diagnostic method , including the high cost of reagents and required equipment , the necessity for trained personal , and the fact it cannot be performed in the field . Indicators for successful schistosomiasis control outlined by WHO are reduced prevalence and intensity of infection [43] . Prevalence and intensity of infection data are then used to assign categories for quantification of those suffering severe disease—assuming that a high intensity infection correlates with higher morbidity . The category of intensity infection is then used to decide on the control measures required and monitoring the control program implemented . Currently if less than 20% of a population has a high intensity of infection ( >400 EPG ) the area is classified as category I and screening followed by treatment is recommended . However , as previously stated S . japonicum morbidity can be unrelated to intensity of infection and if the KK is missing many light infections then the current guidelines for control may need to be reconsidered , particularly if the KK will continue to be used . The benefits of the KK , low cost and ease of execution , mean that it will continue to be a popular choice for control programs in the future . Low compliance in MDA programs for schistosomiasis is an added complication to control in the Philippines . One study in Western Samar used barangay leaders to help with advocating and informing the study population about an upcoming MDA program [44] . Even with this leadership endorsement , less than 50% of individuals in the study area made themselves available for treatment . Compliance during MDA programs can be even lower through a lack of community involvement . In addition , individual case finding is rarely done , even when clear clinical symptoms of schistosomiasis are present . The focal nature of the disease , the low sensitivity of commonly used diagnostic procedures , low compliance during MDA programs and the lack of case finding make it difficult to determine the correct prevalence of schistosomiasis in the Philippines and to monitor control efforts .
The results presented here clearly indicate that the prevalence of S . japonicum in Palapag , Northern Samar is much higher than has previously been reported , a situation that may be reflected in other endemic areas in the Philippines . Through diagnostic surveillance , molecular tools such as qPCR can provide improved assessment of the effectiveness and impact of integrated schistosomiasis control strategies . For China , which is nearing schistosomiasis elimination [20 , 45] , improved diagnostic techniques such as real-time qPCR and conventional PCR ( cPCR ) , will be required to monitor low intensity infections . Current guidelines for schistosomiasis control that are based on prevalence and intensity measurements obtained by the KK may also need to be revised as improved sensitive diagnostic techniques , including the qPCR , become more extensively used . The WHO recommends mass treatment , particularly of school aged children , as a way to prevent morbidity in later life [46] . However with zoonotic transmission occurring , simply treating human cases will not result in elimination of S . japonicum . Integrated control , utilizing mass treatment in combination with other strategies to break the life cycle of the disease , has been suggested as a more cost-effective way to achieve elimination [47] . The advent of more sensitive diagnostics , such as qPCR , may assist in achieving elimination in the long term .
|
Schistosomiasis is caused by infection with trematode blood flukes of the genus Schistosoma . Schistosoma japonicum is the causative agent of schistosomiasis in the Philippines , China and parts of Indonesia . In the Philippines , 6 . 7 million people live in endemic areas and 1 . 8 million are at risk of infection whereas concerted control efforts over the past 50 years in China have reduced the number of infected individuals considerably . Currently used microscopic techniques for diagnosis , notably the Kato-Katz ( KK ) technique , lack sensitivity in areas with low intensity schistosome infections . We have used a molecular diagnostic approach ( qPCR ) , to assess the prevalence of S . japonicum in humans from six barangays in Northern Samar , the Philippines . The qPCR performed considerably better than the KK as a diagnostic procedure and could be an important tool in the future for surveillance and monitoring of areas where extensive schistosomiasis control has led to low prevalence and intensity infections and where schistosomiasis elimination is possible .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion",
"Conclusion"
] |
[] |
2015
|
Real-time PCR Demonstrates High Prevalence of Schistosoma japonicum in the Philippines: Implications for Surveillance and Control
|
Opportunistic infections caused by Pseudomonas aeruginosa can be acute or chronic . While acute infections often spread rapidly and can cause tissue damage and sepsis with high mortality rates , chronic infections can persist for weeks , months , or years in the face of intensive clinical intervention . Remarkably , this diverse infectious capability is not accompanied by extensive variation in genomic content , suggesting that the genetic capacity to be an acute or a chronic pathogen is present in most P . aeruginosa strains . To investigate the genetic requirements for acute and chronic pathogenesis in P . aeruginosa infections , we combined high-throughput sequencing-mediated transcriptome profiling ( RNA-seq ) and genome-wide insertion mutant fitness profiling ( Tn-seq ) to characterize gene expression and fitness determinants in murine models of burn and non-diabetic chronic wound infection . Generally we discovered that expression of a gene in vivo is not correlated with its importance for fitness , with the exception of metabolic genes . By combining metabolic models generated from in vivo gene expression data with mutant fitness profiles , we determined the nutritional requirements for colonization and persistence in these infections . Specifically , we found that long-chain fatty acids represent a major carbon source in both chronic and acute wounds , and P . aeruginosa must biosynthesize purines , several amino acids , and most cofactors during infection . In addition , we determined that P . aeruginosa requires chemotactic flagellar motility for fitness and virulence in acute burn wound infections , but not in non-diabetic chronic wound infections . Our results provide novel insight into the genetic requirements for acute and chronic P . aeruginosa wound infections and demonstrate the power of using both gene expression and fitness profiling for probing bacterial virulence .
Infections caused by opportunistic bacterial pathogens are a primary cause of morbidity and mortality in both the developed and developing world . These infections are often characterized by robust growth of the pathogen in the infection site and increasingly high resistance to antibiotic treatment . The opportunistic pathogen Pseudomonas aeruginosa is responsible for a wide range of infections in immunocompromised hosts [1] . Among the most significant of these infections are those localized to soft tissues , including chronic and burn wounds [1] , [2] . Chronic wounds are defined as wounds that have “failed to proceed through an orderly and timely process to produce anatomic and functional integrity , or proceeded through the repair process without establishing a sustained anatomic and functional result” [3] . Chronic wounds include pressure ulcers ( bed sores ) , diabetic ulcers , venous ulcers , and arterial ulcers , and affect approximately 5–7 million people per year in the US at a cost of $10–20 billion per year [4] . Infections in burn wounds also carry a heavy medical and economic burden not only in the developed world , but also in the developing world , where 70% of burns affect children , and mortality in patients with burns covering >40% total body surface area approaches 100% [5] , [6] . Interestingly , burn infections caused by P . aeruginosa often deteriorate rapidly and lead to systemic spread and death within days or weeks , yet P . aeruginosa chronic wound infections persist for much longer with little associated mortality [5] . As this difference in infection trajectory is not thought to be a result of colonization with specific P . aeruginosa strains , the mechanisms underlying this difference remain a mystery . One possibility is that the type of injury impacts key features of the host environment , such as the immune response , and that this dictates disease progression [7] . A second , non-mutually exclusive possibility is that P . aeruginosa physiology and gene expression is different in chronic and acute wounds . In this work , we set out to address this second possibility using genomic methods . Phenotypes thought to be associated with acute or chronic P . aeruginosa infections have been extensively studied in vitro , and much is known about the genes responsible for these phenotypes and their regulation . For example , the Gac/Rsm and cyclic-di-GMP signaling networks both control expression of “acute” virulence determinants ( e . g . , type III secretion ) and “chronic” virulence determinants ( e . g . , exopolysaccharides ) [8] , [9] . Yet the genetics and physiology of acute and chronic infections have not been directly compared in vivo using open-ended methods such as those enabled by recent advances in high-throughput genomics . Wounds represent an excellent system to study chronic and acute infections since both occur in soft tissues . Furthermore , mouse models of these infections recapitulate key features of infections in humans , such as rapid sepsis and mortality in acute infections , and prolonged healing times in both diabetic and non-diabetic chronic infections [10] , [11] . In this study , we chose to study acute and chronic wound infections in mice using two complementary genomic technologies: RNA sequencing ( RNA-seq ) and transposon-junction sequencing ( Tn-seq ) . Sequencing RNA-derived cDNA with high-throughput methods has proven to be a highly sensitive and comprehensive method for profiling bacterial transcriptomes [12] . Transcriptome-based approaches essentially use the infecting bacterium as a “biosensor” , using differential gene expression as a measure of the molecular and physiological cues sensed by the organism . Performing RNA-seq on RNA isolated from infected tissue from both model hosts and human patients has yielded genome-scale insights into differential gene expression by numerous bacteria during infection [13]–[15] . The requirement of individual genes for growth in an environment can be assessed genome-wide using Tn-seq . This method involves the quantitative sequencing of genomic DNA adjacent to a transposon insertion site to measure the abundance of an insertion mutant in a complex library containing tens or hundreds of thousands of individual mutants [16]–[18] . By subjecting this library to growth in a particular condition ( such as infection of a model host ) and subsequently profiling the abundance of each mutant by high-throughput sequencing , mutations that affect fitness in that condition can be identified upon comparison with an appropriate control condition . This approach has proven successful in identifying determinants of antibiotic resistance , carbon and energy utilization , and in vivo fitness [19]–[21] . However , despite the fundamentally distinct insights that can be gained by examining genome-wide gene expression and mutant fitness , few studies have performed these analyses on the same set of conditions , and none have done so during infection . The goal of this work was to compare P . aeruginosa metabolism and virulence during models of acute and chronic infection . To this end , we performed RNA-seq and Tn-seq in P . aeruginosa grown in vitro and in two non-diabetic murine wound infection models , an acute infection model resulting in high levels of mortality and a non-lethal chronic infection model . Our results reveal that gene expression and mutant fitness are not correlated for most genes with the exception of metabolic genes , where differential expression is more predictive of a gene's role in fitness . By comparing gene expression and mutant fitness in vivo to growth in a defined medium , we reconstructed metabolism of P . aeruginosa during wound colonization and identified several metabolic pathways , including long-chain fatty acid catabolism , that are required for colonization and persistence . Additionally , we discovered that the ability to chemotax is required for P . aeruginosa fitness in burn but not chronic wound infections . These findings identify key features that are required for P . aeruginosa fitness in wound infections and demonstrate the utility of simultaneous gene expression and knockout fitness profiling for the study of bacterial metabolism , virulence , and physiology during infection .
P . aeruginosa causes both acute and chronic wound infections , and we hypothesize that both bacterial and host factors mediate the outcome of wound infections . Here , we investigated P . aeruginosa wound infections from the perspective of the infecting bacterium to uncover similarities and differences between bacterial physiology in acute and chronic infections . Investigation of genetic requirements for P . aeruginosa colonization and persistence during infection requires animal models that encapsulate many of the key characteristics of human infections . In this study , two non-diabetic murine models of wound infection were used , one acute and one chronic . In the acute model , a dorsal full-thickness ( third degree ) burn is induced by scalding and infected subcutaneously with 102–106 P . aeruginosa . This infection is highly virulent , rapidly causing sepsis that leads to ∼100% mortality within 48 hours [11] . The chronic model involves infection of a surgically created full-thickness dorsal excision wound with 105 P . aeruginosa that is covered by an adhesive dressing . This prevents contractile healing and ensures that these wounds heal by deposition of granulation tissue , much like human chronic wounds [22] . This infection can persist for weeks and is highly resistant to antibiotic treatment , and underlying conditions such as diabetes can extend the persistence time of these infections [10] , [23] . Importantly for our purpose , these two infections can be initiated at approximately the same infecting dose with the same strain of P . aeruginosa . To examine the physiology of P . aeruginosa during growth in these two model infections , we initially used RNA-seq ( Table S1 ) . The rationale for these experiments was that when compared to an appropriate control , transcriptomic methods such as RNA-seq can provide a genome-wide view of differential gene expression during infection , essentially using the infecting bacterium as a “biosensor” to report signals and cues sensed by the bacterium in vivo . As P . aeruginosa exhibits robust growth and persistence in these two infection models , much like in clinical infections , we were particularly interested in its primary metabolism during infection . Therefore , to interpret our RNA-seq results with a focus on central metabolism , we chose to compare the transcriptome of P . aeruginosa grown in vivo to growth in a defined minimal medium , specifically , growth to mid-logarithmic phase in a MOPS-buffered medium containing succinate as the sole carbon source ( MOPS-succinate ) . This allowed comparison of metabolic gene expression in an unknown environment ( in vivo ) to an environment in which metabolism is largely understood ( defined medium ) . We chose to profile the acute infection 40 hours post inoculation and the chronic infection 4 days post inoculation because these two timepoints represent midpoints of the trajectory of these respective infections . We reasoned that P . aeruginosa would have sufficient time to adapt to the infection environment and the host would have sufficient time to mount any immune response it was capable of raising at these timepoints . While no single timepoint can capture the dynamics of gene expression throughout the course of an infection , the timepoints chosen reflect a similar degree of progression in both wounds . We found that P . aeruginosa differentially regulates 14% and 19% of its genome during growth in murine burn and chronic wounds , respectively , as compared to MOPS-succinate ( P<0 . 01 , negative binomial test , fold change ≥4 ) ( Table S2 ) . The transcriptional responses of P . aeruginosa during growth in these two wound types as compared to MOPS-succinate are highly correlated ( Spearman rank correlation coefficient = 0 . 840 ) , suggesting that the cues sensed by P . aeruginosa in acute and chronic wound infections are largely similar ( Figure 1A ) . Notably , 7 . 3% of the genome is commonly up- or down-regulated in both wound infections , which is a significant overlap ( P<4 . 72×10−110 , Fisher's exact test ) . The P . aeruginosa genome encodes numerous virulence factors , and our data provides a genome-wide perspective on the expression of these virulence genes ( Table S3 ) . We saw that many genes in the PA3160-PA3141 cluster , which encodes genes required for lipopolysaccharide O antigen biosynthesis [24] , were down-regulated in vivo , and to a greater extent in chronic wounds . This suggests either that P . aeruginosa may alter its outer surface during infection , or that O antigen biosynthesis is regulated as a consequence of more static growth in vivo . We also saw that genes responsible for the biosynthesis of the siderophores pyochelin and pyoverdine were greatly up-regulated in vivo . Iron is known to be a limited resource in numerous infections , and our results suggest that iron acquisition is important in P . aeruginosa soft tissue infections as well [25] . Many type II and type III secretion system genes were up-regulated in both acute and chronic wounds as well , indicating that P . aeruginosa may be modulating host cellular physiology and extracellular environment through these well-characterized secretion systems [26] , [27] . Finally , we saw down-regulation of many genes in the psl cluster , which is responsible for synthesis of the Psl exopolysaccharide . In strain PAO1 , Psl is the primary exopolysaccharide involved in biofilm formation on abiotic surfaces [28] . Thus , P . aeruginosa differentially regulates much of its virulence repertoire upon wound infection , further underscoring the multifaceted nature of its virulence . To determine what general features of P . aeruginosa physiology are altered in vivo , we performed COG enrichment analyses of genes differentially expressed in wounds as compared to MOPS-succinate ( Figure 1B ) . As expected , genes involved in transport of inorganic ions , such as those encoding predicted ferric and ferrous iron transport systems , are enriched in the set of genes up-regulated in vivo . We also noted that amino acid biosynthetic genes are significantly enriched in the set of down-regulated genes in both wound types as compared to MOPS-succinate , suggesting that many amino acids are available in both chronic and acute wounds . Finally , the most extensive regulation in vivo was seen in COG category C , which includes genes involved in energy production and conversion , suggesting that the primary metabolism of P . aeruginosa is extensively remodeled during infection relative to growth in minimal media . Our transcriptomic results suggest that bacterial gene expression is extensively regulated during infection . Yet it is unclear whether those genes that are differentially regulated play a role in in vivo fitness . To address this question , we chose to complement our in vivo transcriptomic studies with Tn-seq to identify the genetic determinants of bacterial fitness in acute and chronic wound infections . Briefly , a library of ∼100 , 000 P . aeruginosa transposon mutants [20] was grown in MOPS-succinate and in both acute and chronic wound models , and mutant abundance was profiled by Tn-seq either 24 hours or 3 days post inoculation for the acute and chronic infections , respectively ( Table S4 ) . As the abundance of a particular mutant in the library will be influenced by its relative fitness throughout the history of the library , these timepoints are sufficient to query genes required for both initial colonization and subsequent growth in these infections . We found that 11% and 16% of the genome contributes to fitness in murine burn and chronic wounds as compared to growth in MOPS-succinate ( P<0 . 05 , negative binomial test , fold change ≥4 ) , respectively , and that 3% of the genome contributes to fitness in both wounds , which is a significant overlap ( P<1 . 66×10−25 , Fisher's exact test ) ( Table S5 ) . We first examined the fitness contribution of known virulence factors ( Table S6 ) . We saw that the flagellum is required only in burn wounds , confirming previous studies and the validity of our Tn-seq approach [29] . Many genes in the type III and the type VI secretion systems contribute to fitness in chronic wounds , further suggesting that inter-cellular delivery of effector proteins may be important in these wounds . Interestingly , despite their down-regulation , the psl exopolysaccharide genes contribute to fitness in both acute and chronic wounds . Finally , some genes involved in producing type IV pili , another motility system [30] , appear to be required in both acute and chronic wounds . Taken together , our results emphasize that virulence in P . aeruginosa is multifactorial , involving the coordinated action of motility , biofilm formation , and secretion systems . Since transcriptomics has been used in the past to identify bacterial genes potentially important for in vivo fitness [14] , [15] , [31] , we hypothesized that genes identified as important for fitness using Tn-seq would display increased expression in vivo . If this hypothesis is true , one would expect that a correlation coefficient ( which expresses correlation between two variables on a scale from −1 , or perfectly anticorrelated , to 1 , or perfectly correlated ) would be closer to −1 . However , we found that mutant fitness and differential expression are uncorrelated , suggesting that in this case RNA-seq is not a good predictor of genes important for fitness in wounds ( Figures 2A and S1 , Table 1 ) . One should keep in mind that Tn-seq is a competitive infection since most strains are wild-type for a given genetic locus , and RNA-seq may be more predictive if individual mutants are examined . Although it is clear that Tn-seq and RNA-seq results are not correlated when all genes are examined , we hypothesized that particular subsets of genes may show a stronger correlation . If this is the case , identifying these subsets of genes would have the potential to guide hypotheses regarding genes important for fitness in bacteria with poor genetic tools , or in natural microbial populations such as those associated with primary human samples , where methods like Tn-seq are not feasible . One possible subset of genes that may be more predictive are those that are most highly differentially regulated . To test this , genes were ranked from high to low fold-change expression and correlated with fitness scores for ever-increasing subsets of genes along that ranking . No significant improvement in correlation was observed , indicating that the magnitude of differential in vivo expression is not more predictive of fitness ( Figure S2 ) . We found that the same is true for genes that contributed strongly to fitness , as ranking from low to high mutant fitness also does not enhance the correlation . As Tn-seq measures the fitness of single mutants , we hypothesized that genetic redundancy might mask a role of some genes in fitness , and that limiting our analysis to genes without predicted redundancy might improve expression-fitness correlation . To test this , Enzyme Commission ( EC ) numbers , which describe the enzymatic function of a gene product , were used to determine which genes lack functional paralogs elsewhere in the genome . However , limiting our analysis to those differentially expressed genes with a unique EC number did not substantially alter expression-fitness correlation ( Table 1 ) . It should be noted that this approach does not address more complex manifestations of redundancy , such as robustness in functional gene interaction networks [32] , which may contribute to the lack of correlation between our Tn-seq and RNA-seq results . We next examined whether the predictive power of gene expression for mutant fitness is better for certain functional classes of genes . Therefore , we examined expression-fitness correlation by COG category . We saw that differential expression and mutant fitness are more negatively correlated in both wound models for several COG categories ( Figure 2B ) . One of these COG categories is amino acid metabolism and transport , suggesting that , relative to growth in MOPS-succinate , P . aeruginosa down-regulates amino acid biosynthetic genes in vivo to avoid the fitness cost associated with expressing them when they are not needed ( Figure 2B ) . We also saw that differential expression and mutant abundance are more negatively correlated for genes in the energy production and conversion , lipid metabolism , and inorganic ion transport and metabolism COG categories . For genes in these categories , up-regulation is more predictive of a fitness defect of mutants lacking those genes ( Figure 2B ) , suggesting that changes in metabolic gene expression are adaptive , conferring a fitness benefit on the organism . These results underscore the importance of scavenging available nutrients and limited-availability ions ( such as amino acids and iron ) while up-regulating key central metabolic pathways during infection . Our analysis of the correlation between differential expression and conditional mutant fitness by COG category ( Figure 2B ) indicates that expression is a better predictor of fitness contribution for genes involved in primary metabolism . Therefore , to characterize the primary metabolism of P . aeruginosa during wound infection , we projected our transcriptome profiling results onto the Kyoto Encyclopedia of Genes and Genomes ( KEGG ) PATHWAYS database ( Figure 3 ) . As mentioned previously , our choice of a defined medium ( MOPS-succinate ) as a control condition provided a reference point from which to understand bacterial physiology and metabolism in the unknown nutritional environment of the infected wound . Our metabolic reconstruction suggests that genes encoding decarboxylating steps of the TCA cycle ( isocitrate dehydrogenase and α-ketoglutarate dehydrogenase ) are down-regulated , and that the gene encoding the entry point to the glyoxylate shunt ( isocitrate lyase ) is up-regulated . The glyoxylate shunt is a variation on the TCA cycle not present in mammals that allows bacteria , including P . aeruginosa , to grow on reduced carbon sources such as fatty acids by bypassing TCA cycle reactions that would result in the loss of carbon [33] . This serves to replenish TCA cycle intermediates utilized in biosynthesis , which is known generally as anaplerosis . We also observed up-regulation of ppc , which encodes a second non-mammalian anaplerotic enzyme , PEP carboxylase , in wound infections , though it is much more highly expressed in chronic wounds than in acute wounds ( Table S2 ) . Finally , expression of a number of genes associated with oxygen-limited environments is also up-regulated , including those encoding high-affinity terminal oxidases and the denitrification pathway [34] , suggesting that at least some bacterial cells in these infections sense decreased oxygen tension . This is consistent with frequent observations of ischemia at wound sites in the clinic , and suggests that oxygen limitation affects the physiology of both the infecting organism and host tissue at wound sites [35] . The transcriptomic results suggest that P . aeruginosa differentially regulates a large portion of its genome in soft tissue infections , and that this reflects a response to differential availability of key metabolic factors such as carbon and energy sources , biosynthetic endproducts , and terminal electron acceptors . In the remainder of this manuscript , we will examine each of these in detail . Infected host tissue is a complex nutritional environment for a bacterium , with many potential metabolites available for bacterial catabolism . Manipulation of key metabolites during infection has therapeutic potential in much the same way as arginine-auxotrophic cancers can be treated by depletion of available L-arginine [36]; however , the nutrients utilized by bacteria in wound infections are not known . Comparing the expression of primary metabolic genes in vivo to growth in defined minimal media led us to hypothesize that fatty acids are a primary carbon source available to P . aeruginosa in vivo ( Figure 3 ) . Examination of our Tn-seq data in detail ( Table S4 ) revealed that the faoAB ( or fadBA5 ) genes , which are required for robust growth on long-chain ( C12 or greater ) fatty acids [37] , contribute to P . aeruginosa fitness in vivo ( Figures 4A and S3A ) . This was confirmed by single mutant infections: both an faoA transposon mutant and an faoA deletion mutant are attenuated in both acute and chronic wounds , indicating that long-chain fatty acids are likely an important energy source in wounds ( Figure 4BC ) . The faoAB genes have also been shown to contribute to resistance to tobramycin , so they could potentially contribute to resistance to an unspecified chemical stress in vivo as well [20] . However , our in vivo gene expression data suggests that growth in wound infections involves pathways active during growth on reduced carbon sources such as long-chain fatty acids ( Figure 3 ) . We did not observe an in vivo fitness defect for genes annotated as homologs of the Escherichia coli long-chain fatty acid outer membrane transporter gene fadL ( fadL1 , fadL2 , or fadL3 ) in P . aeruginosa; however , these genes are not thought to be required for long-chain fatty acid transport in P . aeruginosa [38] . In addition to identifying primary carbon and energy sources during wound infections , our results also allow identification of biosynthetic end products that are available to P . aeruginosa in wound infections . We reasoned that biosynthetic pathways required in minimal media but dispensable in vivo would likely be responsible for the synthesis of metabolites available to P . aeruginosa in vivo . To identify biosynthetic genes , we used the manually curated PseudoCyc annotation [39] . Metabolites for which 33% or more of the biosynthetic genes contribute more to fitness in MOPS-succinate than in both acute and chronic wounds ( P<0 . 05 , negative binomial test , fold change ≥2 ) were deemed “available” , and include many amino acids , the electron carriers FAD and NAD , and the B vitamin thiamine ( Figure 5A and Table S7 ) . Metabolites for which 95% or more of the biosynthetic genes have a similar effect on fitness in MOPS-succinate and in both acute and chronic wounds were deemed “not available” , and include the amino acids glutamate , tyrosine , phenylalanine , aspartate , and asparagine , purines , many other vitamins and cofactors including the folate precursor p-aminobenzoate ( PABA ) and several other B group vitamins . The remaining metabolites that do not match either of the above sets of criteria were deemed “potentially available” . To confirm the validity of this approach , we constructed two in-frame , unmarked deletion mutants lacking the ability to biosynthesize metabolites predicted to be unavailable in wounds: one lacking pabC ( which requires PABA for growth in a minimal medium ) and one lacking purF ( which requires purines for growth in a minimal medium ) ( Figure S3B ) . These two mutants were completely attenuated for virulence in the burn wound ( Figure 5B ) . However , in-frame , unmarked deletion mutants unable to grow without histidine or without isoleucine , leucine , and valine , all of which are predicted to be available in wounds , were significantly more virulent than the pabC or purF mutants . Thus , by comparison with minimal media , we demonstrate that genome-wide bacterial mutant fitness can be used to comprehensively profile bioavailable metabolites in a complex , undefined environment . Bacterial-specific pathways responsible for biosynthesis of any of the unavailable metabolites identified may represent promising targets for therapeutic intervention in wounds . While our focus on metabolism revealed numerous similarities in chronic and acute infections , the genomic techniques employed here also provided new insight into how these infection types differ ( Table S8 ) . As a motile bacterium , P . aeruginosa possesses the ability to detect and move toward nutrients ( including long-chain fatty acids [38] ) , a process referred to as chemotaxis . Examination of our Tn-seq results revealed that several genes with putative roles in chemotaxis , including cheA , cheB , cheR1 , and a homolog of cheW ( PA3349 ) are required in burn , but not chronic wounds ( Figure 6A ) . In addition nearly every annotated flagellar gene is required for fitness in burn wounds [29] , but is dispensable in chronic wounds ( Figure 6B ) . To further confirm the role of chemotaxis in acute wound infections , single-strain infections with a cheR1 transposon insertion mutant and an in-frame , unmarked cheR1 deletion mutant ( Figure S3C ) were performed . While both the cheR1 insertion and deletion mutants have virulence defects in burn wounds ( Figure 6C ) , the cheR1 insertion mutant is as fit or more fit than wild-type P . aeruginosa in chronic wounds ( Figure 6D ) . These results suggest that the ability to chemotax along a spatio-chemical gradient by utilizing flagellar motility is a key feature of acute but not chronic wound P . aeruginosa infections .
The opportunistic pathogen P . aeruginosa is remarkably versatile , able to thrive in a wide range of environments and cause infections in diverse tissue types . These infections can have wide-ranging timescales , from mere days in acute infections such as those in burn wounds or in the cornea to the decades-long pulmonary infections associated with cystic fibrosis [1] . Remarkably , P . aeruginosa is able to achieve this breadth of infectious capability with highly conserved genomic content [40] . This suggests that the capacity to be an acute or a chronic pathogen is innate to the organism , and is determined largely by the context in which the infection is found . Using two complementary genomic techniques , we have investigated the physiology and fitness of P . aeruginosa in two soft tissue infections , one acute and one chronic , and shown similarities and differences between them . Interestingly , with the exception of chemotactic motility , there do not seem to be many infection type-specific genetic pathways required for fitness in one infection versus the other , suggesting that components of host physiology , likely the immune system , dictate the fate of soft tissue infections . While our data include numerous implications for the role of characterized virulence systems in wound infections , we have chosen to focus mainly on metabolic genes in this study . The reason for this is three-fold: ( 1 ) We found that , when compared to growth in a defined medium , metabolic gene expression could be easily interpreted and correlated well with mutant fitness . This approach allowed us to fully profile catabolism , anabolism , and respiration for an infecting bacterium solely from transcriptomic data , which may prove useful in the study of bacterial physiology in conditions in which mutant fitness experiments are not feasible ( such as human infections ) . ( 2 ) Bacterial metabolism during infection is poorly understood . The primary carbon and energy sources utilized by bacteria during infection are only known for a few instances [41] . Our data suggest that by comparing Tn-seq data obtained after growth in vivo to that obtained in a defined minimal medium can make great inroads towards a greater understanding of metabolism during infection . ( 3 ) Modulation of the host metabolic environment has shown promise in treating other diseases characterized by fast-growing and invasive cells such as cancer [36] , and has immediate therapeutic potential for treatment of infections . Our data suggest that interfering with long-chain fatty acid catabolism ( Figure 4 ) or transport , or biosynthesis of several key metabolites ( Figure 5 ) in P . aeruginosa wound infections may impair bacterial fitness in vivo . As in other Gram-negative bacteria , the core chemotaxis system of P . aeruginosa transduces signals from methyl-accepting chemotaxis proteins ( MCPs ) , each of which is thought to respond to a distinct signal , to the flagellar motor , ultimately resulting in chemotaxis along a gradient [42] . We found by Tn-seq that the major aerotaxis receptor gene aer is a fitness determinant in burn wounds , suggesting a role for aerotaxis in burn wounds ( Tables S5 and S8 ) . However , the P . aeruginosa genome encodes at least two aerotaxis MCPs [43] , and mutants lacking either aer alone or aer and aer2 together exhibited full virulence in single-strain burn wound infections ( data not shown ) , suggesting that multiple MCPs can contribute to chemotaxis in acute wounds . Thus , genetic redundancy may mask the role of other genes in fitness in studies of single mutant strains such as ours . This weakness may be exacerbated in bacteria with large genomes that contain more paralogs , like P . aeruginosa [44] . In addition to the function of paralogs , genetic redundancy may also result from robustness in genetic interaction networks , which is more difficult to predict a priori [32] . The relative lack of chronic wound-specific single mutant phenotypes ( Figure 6B ) may be partially attributable to this redundancy . Therefore , systematic approaches to examine the phenotypes of double and triple mutants are needed to uncover the basis for and importance of polygenic traits in bacteria . A second weakness inherent to pooled selection approaches such as Tn-seq is cross-complementation , in which the lack of a particular gene product in one strain is complemented by the production of that product by a neighboring strain . This is often thought of in the context of “public goods” , which are often equated with secreted products [45] . However , we noted that several genes presumed to be involved in the production of secreted products , such as the siderophores pyochelin and pyoverdine , are required of an individual strain in co-infection ( Table S6 ) . This suggests that some secreted products may confer a benefit on the secreting cell without fully transferring those benefits to neighboring cells . The extent to which cross-complementation affects phenotypes of other mutants lacking certain products thought to be cytoplasmic is unclear as well . For example , are metabolic precursors shared between strains , and does that affect our ability to identify essential anabolic pathways by Tn-seq ( Figure 5 ) ? Further study on the exact nature and molecular basis of public goods is required , and can help inform our understanding of community interactions both within and between species during infection . By investigating the correlation between gene regulation and knockout fitness , we showed that , generally , P . aeruginosa gene regulation in wound infections is nonadaptive . As an opportunistic pathogen whose evolutionary trajectory is not thought to be shaped by its fitness in mammalian infections , it is not surprising that regulation of factors that contribute to fitness in wounds is not necessarily tied to signals or cues present in the infection environment . In longer-lasting infections where P . aeruginosa can evolve to be more fit in its environment , such as in the cystic fibrosis lung , adaptive changes in global gene expression have been observed over time [46] . This suggests that P . aeruginosa gene regulation can be better “tuned” by evolution to express in vivo fitness determinants . It would be interesting to explore expression-fitness correlation in “professional” pathogens , as it may be improved in organisms that are more adapted for growth in the human host .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal Care and Use Committee of Texas Tech University Health Sciences Center ( Protocol Numbers 07044 and 96020 ) . P . aeruginosa PAO1 transposon insertion mutants , including the ∼100 , 000 transposon mutant library and individual transposon mutants in faoA ( strain PW6048 ) and cheR1 ( strain PW6640 ) , were obtained from Colin Manoil ( University of Washington ) [20] , [47] . Transposon insertions were confirmed by PCR . For single-strain infection experiments , the parental PAO1 strain was used . The PAO1 strain used in RNA-seq experiments was obtained from Dennis Ohman ( Virginia Commonwealth University ) . Growth in vitro was performed either in LB Miller broth , in morpholine-propanesulfonic acid ( MOPS ) -buffered minimal medium [48] with 20 mM succinate ( hereafter referred to as MOPS-succinate ) , or in MOPS-buffered minimal medium with 0 . 2% oleic acid and 1% Brij-58 [37] ( hereafter referred to as MOPS-oleate ) shaking at 250 rpm at 37°C . In vitro cultures for Tn-seq analysis were grown as follows: frozen aliquots of the PAO1 transposon insertion library were washed twice with 1 mL MOPS-buffered media base , inoculated into 10 mL media at 106 CFU/mL and grown for approximately 9 generations ( to ∼109 CFU/mL ) . In vitro cultures for RNA-seq analysis were grown overnight in the test media , diluted to an OD600 of 0 . 01 , and grown to mid-logarithmic phase ( OD600 = ∼0 . 5 ) before harvesting as described below . Deletion constructs contain two 600–800 basepair ( bp ) fragments flanking the gene of interest in which the coding sequence of the gene of interest was replaced by the sequence 5′-GCGGCCGCC-3′ ( preserving the native start and stop codons ) . This insert was cloned into plasmid pEXG2 [49] on a SacI-KpnI fragment , and these deletion alleles were introduced to strain PAO1 by allelic exchange to generate strains PAO1 ΔpurF , PAO1 ΔpabC , PAO1 ΔilvD , PAO1 ΔhisE , PAO1 ΔfaoA , and PAO1 ΔcheR1 as described [50] . Complementation plasmids were constructed by amplifying the coding sequence of the gene of interest ( including the native start and stop codons ) by PCR with Phusion Hot Start II DNA Polymerase ( Thermo Scientific , Waltham , MA ) . The forward primer in these reactions had a 5′ tail of 5′-GCTATGACCATGATTACGAATTCNNNNNNNNTACAT-3′ , and the reverse primer in these reactions had a 5′ tail of 5′-CATGCCTGCAGGTCGACTCTAGA-3′ . PCR products and the plasmid pUCP18 ( linearized by triple digestion with SacI , BamHI , and KpnI ) were gel purified , and the PCR product was introduced to the plasmid backbone by Gibson assembly as described [51] . This generated a library of >300 plasmids in E . coli strain DH5a for each plasmid , and these plasmid libraries were prepared from E . coli . Then , these plasmid libraries were used to transform the appropriate mutant PAO1 derivative to be complemented by electroporation , and complementing plasmids were isolated as follows: ( 1 ) For plasmids pPurF and pPabC , the pPurF and pPabC candidate plasmid libraries were transformed into PAO1 ΔpurF and PAO1 ΔpabC , respectively , and complemented electroporants were isolated by plating on solid MOPS-succinate agar and restreaking colonies on solid MOPS-succinate agar for isolation . ( 2 ) For plasmid pFaoA , the pFaoA candidate plasmid library was transformed into PAO1 ΔfaoA , and complemented electroporants were isolated by plating on solid MOPS-oleate agar and restreaking colonies on solid MOPS-oleate agar for isolation . ( 3 ) For plasmid pCheR1 , the pCheR1 candidate plasmid library was transformed into PAO1 ΔcheR1 , and complemented electroporants were isolated by spotting the transformation mix on semisolid LB media containing 0 . 3% agar supplemented with 150 µg/mL carbenicillin and restreaking a region that had swam out from the initial spot for isolation on solid LB agar supplemented with 150 µg/mL carbenicillin . At least three individual complemented strains were tested as shown in Figure S3 for each deletion , with a representative strain shown for each . For the burn wound infections shown in Figure 5B , mid-logarithmic phase cells of the indicated PAO1 mutants were starved for 2 hours in MOPS buffer before inoculation as described below . Murine burn wound infections were performed with adult female Swiss Webster mice essentially as described [11] , with the following modifications . For Tn-seq experiments , 106 CFU of the PAO1 transposon mutant library was used as an inoculum , and wound tissue was harvested 24 hours post infection and stored in RNAlater ( Qiagen ) at room temperature for 24–48 hours , and subsequently at −20°C . For RNA-seq experiments , 105 CFU of wild-type PAO1 was used as an inoculum , and wound tissue was harvested 40 hours post infection as described above . For single strain infections , 102–103 CFU of the indicated strain was used as an inoculum , and animals were monitored for mortality daily for up to 7 days . Each experiment was performed at least twice with at least 5 animals per experimental group , and the average time of death for all animals is reported here . Murine chronic wound infections were performed with non-diabetic adult female Swiss Webster mice essentially as described [23] , with the following modifications . For Tn-seq experiments , 105 CFU of the PAO1 transposon mutant library was used as an inoculum , and wound tissue was harvested 3 days post infection and stored in RNAlater as described above . For RNA-seq experiments , 105 CFU of wild-type PAO1 was used as an inoculum , and wound tissue was harvested 4 days post infection as described above . For single strain infections , 105 CFU of the indicated strain was used as an inoculum , wound tissue was harvested 4 days post infection , and CFU/g tissue was determined by plating . Each experiment was performed at least twice with at least 5 animals per experimental group . To prepare DNA for Tn-seq analysis , ∼100 mg sections of wound tissue or cell pellets were resuspended in 1 mL 1× Buffer A [52] +0 . 1% SDS , homogenized in a Mini-Beadbeater ( Biospec ) in 2 mL vials preloaded with Lysing Matrix B ( MP Biomedicals ) 3–5 times for 1 minute each , resting on ice in between each pulse . Proteinase K was then added to 1 mg/mL , and samples were incubated overnight at 50°C . Samples were then homogenized once more as above , separate sections from the same wound were pooled , and samples were extracted with an equal volume of 25∶24∶1 phenol∶chloroform∶isoamyl alcohol pH 8 . 0 . DNA was ethanol precipitated from the aqueous phase , and was resuspended in 200–500 µL water after extensive pellet washing with 75% ethanol . Tn-seq sequencing libraries were prepared by a modified version of INSeq [16] . DNA was sheared to approximately 500 bp either in a S220 Focused-ultrasonicator ( Covaris ) , a Hydroshear Sonicator ( Digilab ) , or a Q880R Sonicator ( Qsonica ) . 500 ng ( in vitro samples ) to 1 µg ( murine wound samples ) of DNA was used as template in two linear PCR reactions using the 5′ biotinylated oligonucleotide primer Kbio-T8OE-Out2 ( 5′-ATAAGAATGCGGCCGCGGGATGGAAAACGGGAAAGGTTCCGTCCAGGACGCTACTTGTG-3′ ) and performed with KOD Hot Start DNA Polymerase ( EMD Biosciences ) with the following program: 95°C , 5′; 99× ( 95°C , 30″; 68°C , 1′ ) ; 68°C , 10′ . Kbio-T8OE-Out2 is specific to the “OE” end of transposon T8 [47] , with two key features: ( 1 ) A NotI site is contained towards the 5′ end of the primer for NotI cleavage-mediated elution ( see below ) , and ( 2 ) the primer sequence ends 12 bp from the end of transposon T8 , leaving that additional 12 bp sequence for additional sequence quality control . Biotinylated linear PCR products were bound to Streptavidin-coupled Dynabeads ( Invitrogen ) and a second strand was synthesized as described [52] , except that the oligonucleotide used to prime second strand synthesis had the sequence 5′-NSNSNSNSNS-3′ . Double-stranded DNA was eluted from the Dynabeads by digesting with NotI-HF ( New England Biolabs ) , and this DNA was prepared for Illumina sequencing with the NEBNext DNA Library Prep Master Mix Set For Illumina ( New England Biolabs ) according to the manufacturer's protocol . Libraries were sequenced at the Genome Sequencing and Analysis Facility at the University of Texas at Austin on a HiSeq 2000 ( Illumina ) on a 2×100 paired end run . All sequences are deposited with the National Center for Biotechnology Information Sequence Read Archive under Accession Number SRP033652 . To prepare RNA for RNA-seq sequencing libraries , cell pellets or ∼100 mg sections of wound tissue were homogenized 2–4 times in a Mini-Beadbeater in 1 mL RNA Bee ( Tel-Test ) in 2 mL vials with Lysing Matrix B , and aqueous phases of extractions from different sections of the same wound were pooled before continuing with the extraction . RNA was then prepared according to the RNA Bee manufacturer's protocol . DNA contamination was then removed by DNAse digestion as described [31] . rRNA integrity was then verified by agarose gel electrophoresis . Starting with 5 µg of total RNA , bacterial rRNA was depleted from all samples with the Ribo-Zero rRNA Removal Kit ( Bacteria ) ( Epicentre ) , and then mammalian rRNA was depleted from all samples with the Ribo-Zero Gold Kit ( Human/Mouse/Rat ) ( Epicentre ) according to the manufacturer's protocol . Remaining RNA was fragmented with the NEBNext Magnesium RNA Fragmentation Module ( New England Biolabs ) according to the manufacturer's protocol with a 5′ incubation time . Illumina sequencing libraries were then prepared with the NEBNext Multiplex Small RNA Library Prep Set for Illumina ( New England Biolabs ) according to the manufacturer's protocol . Finished sequencing libraries were size selected on a polyacrylamide gel for fragments ∼140–300 bp . Libraries were sequenced at the Genome Sequencing and Analysis Facility at the University of Texas at Austin on a HiSeq 2000 ( Illumina ) on either a 1×100 single end or a 2×100 paired end run . All sequences are deposited with the National Center for Biotechnology Information Sequence Read Archive under Accession Number SRP033652 . RNA-seq reads were analyzed and differential gene expression was determined with the R package DESeq [53] largely as described [31] with the following modifications: the P . aeruginosa PAO1 genome ( GenBank accession no . AE004091 . 2 ) was used for read alignment , and COGs used for enrichment analyses were obtained from the Pseudomonas Genome Database [54] . P values given for differentially expressed genes are adjusted for multiple testing . Enrichment of differentially regulated genes in a given COG category was determined by comparing the prevalence of up- or down-regulated genes assigned to a specific COG category to the prevalence of genes in the entire genome assigned to that COG category using Fisher's exact test . Tn-seq reads were parsed , mapped , and tallied , and differential mutant abundance was determined using a custom Unix , Perl , and R pipeline ( available at http://github . com/khturner/Tn-seq ) . First , reads containing the 12-bp transposon T8 end sequence 5′-TATAAGAGTCAG-3′ were identified ( allowing for 1 mismatch or indel ) using fqgrep ( http://github . com/indraniel/fqgrep ) , and sequence up to and including the transposon end sequence were removed with the modified Perl script called “trimmer” . The remaining sequence was then mapped to the P . aeruginosa PAO1 genome ( GenBank accession no . AE004091 . 2 ) using bowtie version 2 . 10 [55] , and individual insertion sites and the number of reads originating from them were tallied with the Unix script “TnSeq . sh” . The data analysis method , contained in the Unix script “TnSeqAnalysis . sh” and the R script “TnSeqDESeq . R” , was inspired largely by the ESSENTIALS software package [56] , and is described in detail below . After removing the 50 most abundant insertion sites from analysis to correct for amplification bias , insertion location vs . number of reads data was smoothed using locally weighted LOESS smoothing using a smoothing parameter ( α ) of 1 to correct for genomic position-dependent effects on apparent mutant abundance . Then , this data was normalized using DESeq [53] with default parameters . For gene knockout abundance analysis , a modified annotation was generated with the 3′ 10% of every gene removed ( to ignore insertions that may not abolish gene function ) . Then , the smoothed , normalized number of transposon-derived reads per gene and the number of insertions mapping to each gene was tallied using this modified annotation in R . The number of transposon-derived reads was incremented by one for each gene to avoid dividing by zero when comparing to a condition with no mutants detected . Finally , differential mutant abundance was calculated using a negative binomial test with DESeq , artificially setting normalization factors to 1 ( because the data was normalized per insertion ) .
|
Soft tissue infections , such as those in burns , bed sores , and diabetic ulcers , are a significant healthcare and economic burden in the developed and developing world . The opportunistic pathogen P . aeruginosa can cause both acute and chronic infections , and the trajectory of these two types of infections is vastly different . We used high-throughput sequencing to profile P . aeruginosa genome-wide gene expression and mutant fitness during mouse model acute and non-diabetic chronic wound infections . Using these data , we show that wounds are nutrient-rich growth environments in which long-chain fatty acids are a primary source of carbon and energy . We also show that the ability to travel along spatio-chemical gradients by chemotaxis is critical for bacterial fitness and virulence in acute but not chronic infections . Our results demonstrate the utility of simultaneous mutant fitness and gene expression profiling to discover critical functions in complex growth environments .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"bacteriology",
"gram",
"negative",
"bacteria",
"genome",
"expression",
"analysis",
"microbial",
"metabolism",
"microbiology",
"animal",
"models",
"model",
"organisms",
"genome",
"analysis",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"microbial",
"physiology",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"mouse",
"models",
"systems",
"biology",
"bacterial",
"physiology",
"transcriptome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"computational",
"biology"
] |
2014
|
Requirements for Pseudomonas aeruginosa Acute Burn and Chronic Surgical Wound Infection
|
Trypanosoma cruzi , the causative agent of Chagas disease , has high affinity for lipoproteins and adipose tissue . Infection results in myocarditis , fat loss and alterations in lipid homeostasis . This study was aimed at analyzing the effect of high fat diet ( HFD ) on regulating acute T . cruzi infection-induced myocarditis and to evaluate the effect of HFD on lipid metabolism in adipose tissue and heart during acute T . cruzi infection . CD1 mice were infected with T . cruzi ( Brazil strain ) and fed either a regular control diet ( RD ) or HFD for 35 days following infection . Serum lipid profile , tissue cholesterol levels , blood parasitemia , and tissue parasite load were analyzed to evaluate the effect of diet on infection . MicroPET and MRI analysis were performed to examine the morphological and functional status of the heart during acute infection . qPCR and immunoblot analysis were carried out to analyze the effect of diet on the genes involved in the host lipid metabolism during infection . Oil red O staining of the adipose tissue demonstrated reduced lipolysis in HFD compared to RD fed mice . HFD reduced mortality , parasitemia and cardiac parasite load , but increased parasite load in adipocytes . HFD decreased lipolysis during acute infection . Both qPCR and protein analysis demonstrated alterations in lipid metabolic pathways in adipose tissue and heart in RD fed mice , which were further modulated by HFD . Both microPET and MRI analyses demonstrated changes in infected RD murine hearts which were ameliorated by HFD . These studies indicate that Chagasic cardiomyopathy is associated with a cardiac lipidpathy and that both cardiac lipotoxicity and adipose tissue play a role in the pathogenesis of Chagas disease . HFD protected mice from T . cruzi infection-induced myocardial damage most likely due to the effects of HFD on both adipogenesis and T . cruzi infection-induced cardiac lipidopathy .
Chagas disease , caused by the parasite Trypanosoma cruzi , is classified by WHO as a neglected tropical disease and is a major cause of morbidity and mortality in Latin America [1] . Globalization has led to the increased recognition of this infection among immigrants from Latin America in non-endemic countries [2] . It has been estimated that 18 to 20 million people have Chagas disease [2] . Symptoms of infection are varied , but include heart disease and megasyndromes of the gastrointestinal tract . Chagas disease has acute , indeterminate and chronic phases . Acute systemic infection is often asymptomatic , but , in those that are symptomatic the disease is characterized by myocarditis and/or meningoencephalitis [1]–[3] . T . cruzi infection causes an intense systemic pro-inflammatory response in many organs including the heart . Following infection the majority of patients develop an asymptomatic latent infection termed the indeterminate ( or latent ) stage of infection . As many as thirty percent of infected individuals may progress to chronic disease characterized by cardiomyopathy and/or mega syndromes . Myocardial dysfunction is associated with extensive remodeling caused by the initial infection and ensuing fibrosis [4] . The low density lipoprotein receptor ( LDLr ) is involved in LDL internalization and regulation of cholesterol homeostasis [5] . We have demonstrated that T . cruzi utilizes LDLr to invade host cells and that LDLr likely plays an important role in the pathogenesis of Chagas disease [6] . T . cruzi has high affinity for LDL and HDL and the rate of invasion increases in the presence of lipoproteins [7] . T . cruzi primarily targets lipid rich adipose tissue as their reservoir and causes lipolysis during acute infection [8] , [9] . Altered serum triglyceride and cholesterol levels are associated with acute infection [6] . The role of host lipids in the pathogenesis of Chagas disease is understudied . Diet plays a major role in the regulation of systemic and whole body lipid levels including adipogenesis and lipogenesis [10] . Recent changes in diet , underlying the well-recognized obesity epidemic , in regions endemic for Chagas Disease are likely to have significant effects on the interaction of this parasite with its human host . Herein , we report , for the first time , the effect of diet on myocardial inflammation and damage seen during acute T . cruzi infection . We also provide data on the role of parasite associated LDL/HDL in the regulation of systemic lipid homeostasis in white adipose tissue ( WAT ) and in the heart .
All animal experimental protocols were approved by the Institutional Animal Care and Use Committees ( IACUC ) of Albert Einstein College of Medicine ( No . 20130202 ) which is adhered to the National Research Council guidelines ( Guide for the Care and Use of Laboratory Animals: Eight Edition , Washington , DC: The National Academies Press , 2011 ) . The Brazil strain of T . cruzi was maintained by passage in C3H/Hej mice ( Jackson Laboratories , Bar Harbor , ME ) . Male CD-1 mice ( Jackson Laboratories ) were infected intraperitoneally ( i . p . ) at 8–10 weeks of age with 5×104 trypomastigotes of the Brazil strain [8] . Mice were maintained on a 12-hour light/dark cycle . Mice , starting at the day of infection , were randomly divided into two groups ( n = 15 per group ) and fed on either high fat diet ( HFD; 60% fat ) or Regular diet ( RD , 10% fat ) ( D12492 or D12450 Research Diets , Inc . , New Brunswick , NJ ) . Uninfected mice were fed on either HFD ( n = 15 ) or RD ( n = 15 ) and used as respective controls in all the experiments . For each replication of this experiment the same numbers of mice were used in all groups . Plasma samples were obtained from 75 µl of blood collected from the orbital venous sinus ( using isoflurane anesthesia ) at 10 , 15 , 20 , 25 and 30 days post infection ( dpi ) . Parasitemia was evaluated by counting in a Neubauer hemocytometer as described previously [8] . Thirty five days after infection the mice were euthanized and heart and epididymal white adipose tissues ( WAT ) were harvested for analysis . At this time-point there was no peripheral parasitemia and mice appeared normal . Colorimetric assays were performed using colorimetric assay kit for non-esterified fatty acid ( NEFA from Cell Biolabs , Inc . CA ) , triglyceride ( TG from Cayman Chemicals , MI ) , and low density lipoprotein ( LDL ) , high density lipoprotein ( HDL ) and total cholesterol ( TC ) ( Enzychrom ( E2HL-100 ) , Bioassay system , CA ) in serum samples6 . Cholesterol levels were quantified in the hearts and WATs of mice at d35pi using a colorimetric assay kit and samples were prepared and assayed following manufacturer's protocol ( Total cholesterol colorimetric assay kit , Cell Biolabs Inc . , CA ) . The lipid content of frozen WAT was quantified using Oil Red O staining . Frozen sections stained with Oil Red O were washed 3 times in isopropyl alcohol for 3 minutes each . Bound Oil Red O was eluted by incubating in isopropyl alcohol ( 5 mL ) for 16 hours; the eluted lipid content stained with Oil Red O was then measured at 540 nm in a spectrophotometer ( Shimadzu uv-1201 ) as previously reported [9] . Tissue lysates were prepared as previously described [6] . An aliquot of each sample ( 40 µg protein ) was subjected to a 4–15% gradient SDS-PAGE and the proteins transferred to nitrocellulose filters for immunoblot analysis . LDLr specific rabbit monoclonal antibody ( 1∶1000 dilution , AB52818 Abcam , Cambridge , MA ) , lipoprotein lipase ( LPL ) specific mouse monoclonal antibody ( 1∶1000 dilution , AB21356 , Abcam ) , or ABCA1 specific rabbit monoclonal antibody ( 1∶1000 dilution , AB18180 , Abcam ) were used as primary antisera . Horseradish peroxidase-conjugated goat anti-mouse immunoglobulin ( 1∶2000 dilution , Amersham Biosciences , Piscataway , NJ ) or horseradish peroxidase- conjugated goat anti-rabbit immunoglobulin ( 1∶5000 dilution , Amersham Biosciences ) were used to detect specific protein bands ( explained in Figure Legends ) using a chemiluminescence system [6] . GDI ( 1∶10000 dilution , 71-0300 , and rabbit polyclonal , Invitrogen , CA ) and a secondary antibody horseradish peroxidase conjugated goat anti-rabbit ( 1∶2000 dilution , Amersham Biosciences ) were used to normalize protein loading . Rabbit Adiponectin antibody was produced in the laboratory of Dr Scherer as described previously [8] . Heart and white adipose tissue were collected from mice on 35 days post-infection and stored at −80°C . A quantitative real-time polymerase chain reaction ( qPCR ) was used to quantify parasite load employing PCR SYBR Green Master Mix ( Roche Applied Science , CT ) containing MgCl2 employing an iQ5 LightCycler ( Bio-Rad ) . Isolation of DNA , preparation of standard curves for host and epimastigote DNA , and qPCR analysis was performed as previously published [8] . Host 18srRNA gene was used for normalization [18S forward: 5′-AGGGTTCGATTCCCGGAGAGG-3′ , reverse , 5′-CAACTTTAATATACGCTATTGG-3′] . An RT2 Profiler ( SA Biosciences , Valencia , CA ) custom designed PCR array for mouse genes involved in LDL internalization , cholesterol metabolism , fatty acid and triglyceride metabolism , glucose metabolism and inflammatory signaling was used to analyze gene expression . Data analysis was performed normalized to the expression of 18sRNA using the ΔΔCT method according to the manufacturer's protocol ( SABiosciences ) and statistical analysis was performed as suggested [9] . Freshly isolated tissues were fixed with phosphate-buffered formalin overnight and then embedded in paraffin wax ( n = 5 ) . Hematoxylin and eosin ( H&E ) staining was performed and the images were captured as previously published [8] . Four to six sections of each heart were scored blindly . For each myocardial sample , histologic evidence of myocarditis and inflammation was classified in terms of degree of degenerating cardiac muscle fibres , inflammation , fibrosis and adipocyte presence and was graded on a five point scale ranging from 0 to 4+ . A zero score indicated lowest or negligible changes and 4 the most damaged state . IFA was performed on the frozen sections using anti-LDL and the images were captured as previously published [6] . The fluorescent intensities of the images were quantified using NIH-Image J program for four to six images of each heart . All mice were imaged after 3 hours of fasting . Mice were administered 300–400 uCi ( 12–15 MBq ) in 0 . 1 mL normal saline , [18F] fluoro-2-deoxyglucose ( FDG ) , via tail vein and imaging was started at 1 hour after injection . This period permits the tracer to be delivered throughout the body and trapped by the glycolytic pathway . Prior to FDG administration the mice were anesthetized with 1 . 5% isoflurane-oxygen mixture , which continued throughout the imaging portion of the procedure . After MicroPET imaging , the animals were housed in the imaging facility for ten half lives ( 18F has a half-life of 110 minutes ) until they could be safely moved back to the Animal Institute for continued housing . The mice were imaged by an Inveon Multimodality scanner ( Siemens , Knoxville , TN ) using its PET module . PET imaging is performed using the PET gantry , which provides a 12 . 7 cm axial and 10 cm transaxial active field of view . The PET scanner has no septa , and acquisitions are performed in the 3D list mode . A reconstructed full-width-half-max resolution of <1 . 4 mm is achievable in the center of the axial field of view . List mode acquisition of data is performed to permit dynamic re-framing for kinetic evaluation of the radiotracer uptake , where indicated . After each acquisition , data were sorted into 3D sinograms , and images were reconstructed using a two dimensional ( 2D ) -Ordered Subset Expectation Maximization algorithm . Data were corrected for decay , dead time counting losses , random coincidences and the measured nonuniformity of detector response ( i . e . normalized ) , but not for attenuation or scatter . Analysis was performed by using the Inveon Research Workplace 4 . 1 software ( Siemens ) . All imaging studies were inspected visually in a rotating 3D projection display to identify interpretability and image artifacts . Regions of interest ( ROI ) were manually defined . Successive scrolling through 2D slices ( each 1 . 2 mm thick in the axial images ) permitted measurement of radioactivity within defined volumes . Corrected counts per cc within this volume divided by the counts per gram of total body mass of injected radioactivity determined the SUV . SUVmax , the maximum value of SUV within the heart was determined . The SUVmax is the maximum value of the percent-age injected dose per gram of cardiac tissue multiplied by the body weight of each animal . The SUVmax has been validated in numerous animal and human models as a reproducible and robust measure of radioactivity in longitudinal studies . Cardiac gated MRI was performed on uninfected and infected mice at 26 dpi were imaged using a 9 . 4 T Varian Direct Drive animal magnetic resonance imaging and spectroscopic system ( Agilent Technologies , Inc . Santa Clara , CA ) as previously published [10] . Briefly , anesthesia was induced with 2% isoflurane in air , mice were positioned supine inside an MR compatible holder and positioned within a 35-mm ID quadrature 1H volume coil ( Molecules2Man Imaging Co . , Cleveland , OH ) . Body temperature was maintained at 34∼35°C using warm air with feedback from a body surface thermocouple . A respiratory sensor balloon was taped onto the abdomen . Cardiac ( ECG electrodes inserted subcutaneously in front left paw and rear right paw ) and respiratory signal ( from sensor balloon taped to the abdomen ) were continuously monitored and used for MR gating/triggering by an SA Monitoring and Gating System ( Small Animal Instruments , Inc . , Stony Brook , NY ) . Ten to fourteen 1-mm-thick slices without gap was acquired in short-axis orientation covering the entire heart using an ECG-triggered and respiratory gated multi-frame tagged cine sequence . The imaging parameters used were field of view ( FOV ) of 40×40 mm2 , matrix size of 256×256 , TE of 2 . 6 ms , TR of 5 . 5 ms , flip angle of 25° , number of averages of 2 . The number of frames was twelve to eighteen . Data were transferred to a PC and analyzed using MATLAB-based software . Left ventricle ( LV ) and right ventricle ( RV ) dimensions in millimeters were determined from the images representing end-diastole . The left ventricular wall is the average of the anterior , posterior , lateral , and septal walls . The right ventricular internal dimension is the widest point of the right ventricular cavity . Immunoblot , immunofluorescence and quantification of parasite load studies were performed at least three times and representative data are presented in the figures . Lipid profile analysis and gene arrays were done in duplicates . Data were pooled and statistical analysis was performed using a Student's t-test ( Microsoft Excel ) as appropriate and significance of difference was determined as p values between <0 . 05 and <0 . 005 . Cxcl16 NM_023158 . 6 , Stab1 NM_138672 , Vldlr NM_013703 , Lrp6 NM_008514 , Ldlr NM_010700 , Scarf1NM_001004157 , Apoa1NM_009692 , Apob NM_009693 , Apoe NM_009696 , Acaa1aNM_130864 , Acad9 NM_172678 , Acad10NM_028037 , Acox1NM_015729 , Fabp1NM_017399 , Acsbg1NM_053178 , Lipe NM_010719 , Npc1NM_008720 , Lcat NM_008490 , Abca1NM_013454 , Abcg1NM_009593 , Cyp39a1NM_018887 , Cyp7a1NM_007824 , Hmgcr NM_008255 , Insig1NM_153526 , Lep NM_008493 , Ppara NM_011144 , Pparg NM_011146 , Adig NM_145635 , Adipoq NM_009605
To investigate the impact of diet on the course of acute T . cruzi infection we initially studied CD-1 mice that were placed on either a high fat diet ( HFD ( 20 kcal% protein , 20 kcal% carbohydrate and 60 kcal% fat ) ) or regular diet ( RD ( 20 kcal% protein , 70 kcal% carbohydrate and 10 kcal% fat ) ) at the time of infection ( i . e . on the first day of infection ) . HFD fat content is composed of saturated ( 81 . 5 g ) , monounsaturated ( 91 . 5 g ) and polyunsaturated ( 81 . 5 g ) fat and RD fat content is composed of saturated ( 9 . 9 g ) , monounsaturated ( 13 g ) and polyunsaturated ( 20 . 7 g ) fat . No significant differences in the body weight were observed between the groups at the start of infection . As time progressed , the infected RD fed mice displayed fat loss ( 60% ) , as quantified by oil red O staining [9] compared to uninfected mice and this was associated with edema on 35 days post infection ( dpi ) as previously described [8] . Analysis suggested that infected RD fed mice gained weight due to edema and the infected HFD fed mice mainly due to fat . In contrast , infected HFD fed mice displayed no signs of edema and had only a 20% fat loss compared to uninfected HFD fed mice . No significant difference was observed in the food take between RD-fed or HFD-fed mice during infection . Peak parasitemia was observed between 15 and 28 dpi in all groups . There was a twofold decrease in parasitemia in HFD fed mice compared to RD fed mice ( Fig . 1a ) . HFD fed mice had a significantly increased survival rate ( 85% ) compared to RD fed mice ( 40% ) during acute infection ( Fig . 1b ) . In addition , a higher parasite load was seen in the myocardium of RD fed mice as determined by qPCR ( Fig . 1c ) . However increased parasite loads were detected in white adipose tissue ( WAT ) of HFD compared to RD fed mice ( Fig . 1c ) . We next investigated the effect of T . cruzi infection on the rate of mortality and parasitemia in mice that were placed on a HFD or RD for 30 days before infection ( i . e . diet pre-fed mice ) and were then continued on the respective diets for 35 dpi . HFD pre-fed mice had a 92% survival compared with a 40% survival of the RD pre fed mice due to acute infection . These data demonstrate that HFD has a protective effect on the mortality seen during acute T . cruzi infection in this murine model . Similar to what was seen in mice started on HFD and RD at the time of infection , we also observed a significant decrease in body weight with HFD pre fed mice ( 86% ) and an increase in body weight with RD pre-fed mice ( 112% ) at 35 dpi compared to their respective uninfected mice ( 100% ) . This appeared to be due to edema which was seen in infected RD pre-fed mice , but which did not occur in infected HFD pre-fed mice . However , uninfected HFD mice tended toward obesity ( 25% greater body weight ) compared to uninfected RD fed mice at d35pi . There were no significant histological differences in the hearts obtained from HFD and RD fed uninfected mice ( Fig . 2a ) . In contrast , the hearts obtained from infected RD fed mice displayed an intense inflammatory reaction most prominent at the right ventricle and left ventricle junction associated with vasculitis and fibrosis . Hearts obtained from infected mice fed a HFD had fewer parasites and a reduction in inflammation and fibrosis ( Fig . 2a ) . Histological scoring of hearts ranged from 0 to 4+ in each of the categories of degenerating cardiac fibers , inflammation , fibrosis and presence of adipocytes . Infected RD animals had higher scores than infected HD fed animals ( figure 2b ) . Summing all scores ( maximum cumulative score 16 ) using this method , the hearts from infected RD fed scored 13 and the hearts from HFD fed infected mice 5 . 2 . Lipolysis and adipolysis are observed during acute T . cruzi infection [8] , [9] . Adipogenesis is regulated by adipokines such as adiponectin , leptin , PPAR-γ , and TNF-α [11] , [12] . These adipokines contribute to the regulation of fatty acid oxidation . The protein encoded by the gene Adig ( adipogenin ) has been implicated in adipocyte differentiation and fat accumulation . Examination of the cellular morphology of WAT of infected RD- and HFD-fed mice demonstrated that WAT from the uninfected HFD-fed mice displayed enlarged ( 280±25 µm ) and lipid enriched adipocytes compared to WAT of uninfected RD-fed mice ( 170±60 µm ( Fig . 2b , top ) . WAT from both the infected RD- and HFD-fed mice demonstrated enlarged adipocytes ( 2 . 0–2 . 5 fold enlarged ) surrounded by adipocytes with smaller lipid droplets . The average size of inflamed adipocytes of HFD fed mice was 500±150 µm and RD fed mice 400±125 µm during infection . HFD-fed mice had lipid droplets that were significantly smaller compared to RD-fed mice ( 70±20 µm and 150±30 µm respectively , ) . There was an increase in the number of dead cells in the WAT of RD-fed mice ( Fig . 2c ) . The number of dead cells was 680±72 and 206±40 in RD fed and HFD fed infected WAT respectively surrounding inflamed adipocytes ( for n = 200 adipocytes ) . The fat loss in WAT was determined using Oil red O staining [9] . RD-fed infected mice displayed 60% fat loss and HFD-fed infected mice 20% compared to their respective uninfected control mice ( p<0 . 05 ) . qPCR demonstrated a significant decrease in the mRNA levels of the genes involved in adipogenesis such as adiponectin , ( −11 . 5 fold ) , Peroxisome proliferator activated receptor gamma ( PPAR-γ , −6 . 0 ) , Adipogenin ( Adig , −3 . 5 fold ) and leptin ( −55 fold ) in WAT of RD-fed mice compared to HFD-fed mice during acute ( d35pi ) infection ( Table 1 ) . There was an upregulation of adipogenic genes , except for PPAR-γ , in the heart of infected HFD-fed mice compared with RD-fed mice . These data suggest that HFD increases adipogenesis during infection and that this alteration affects the course of acute T . cruzi infection . It is likely that adipogenesis or lipolysis directly affects serum lipid homeostasis . T . cruzi has a high affinity for LDL and HDL , and the rate of invasion increases in presence of lipoproteins [7] . We have reported that there is a decrease in serum TG and TC during acute infection in mice fed with a standard chow diet ( 17% fat ) [6] . To analyze whether the HFD would improve the serum lipid levels during infection , we measured TG , FFA and TC levels in HFD fed mice and compared it with RD fed mice at different time points of infection . Serum levels of TG , FFA and TC were significantly decreased in T . cruzi infected mice irrespective of the type of the diet fed ( Fig . 3 ) . Serum TG content of uninfected RD fed mice was higher ( 38% ) than uninfected HFD fed mice; however the TC was significantly higher ( 25% ) in the uninfected HFD fed mice compared to uninfected RD fed mice . Such alterations in the serum lipid level could potentially regulate myocardial lipid levels and the associated inflammation . The rate of T . cruzi invasion has been shown to depend on serum cholesterol levels [7] . Invasion results in elevated intracellular/tissue cholesterol levels ( Fig . 3d ) which could affect triglyceride and cholesterol metabolism ( Table 2 ) . Analysis demonstrated a significant increase in the cholesterol levels in the hearts ( 260% ) and the WATs ( 394% ) of the infected RD fed mice compared to uninfected mice ( 100% ) . Though infection increased cholesterol levels in the HFD-fed infected mice , this analysis demonstrated a significant decrease in their level in both the hearts ( 202% ) and WAT ( 211% ) compared to uninfected RD-fed control mice . IFA of the myocardium using anti-LDL displayed decreased accumulation of LDL ( 30% ) in infected HFD-fed mice compared to infected RD-fed mice on d35 pi ( Fig . 4 ) . qPCR analysis demonstrated increased mRNA levels of the inflammatory markers such as TNF-α ( 5 fold ) and IFN-γ ( 40 fold ) in the WAT of HFD-fed mice at d35pi compared to their respective control groups . WAT of RD-fed mice had mostly necrosed cells and showed 2 fold increase in TNF-α and 9 . 0 fold increase in IFN-γ which was significantly different than HFD-fed mice . The myocardium of the infected RD-fed mice demonstrated a significant increase in TNF-α ( 92 fold compared to uninfected ) whereas , infected HFD-fed mice showed only 28 fold increase in TNF-α . This data supports our histological observations ( Fig . 2 ) . The mRNA level of IFN-γ was higher in the myocardium of both the infected HFD and RD-fed mice ( 185 and 195 fold respectively , compared to their respective controls ) . There was no statistical difference between the mRNA level of IFN-γ seen in infected HFD and RD fed mice . Inflammatory cells use glucose as a primary source of metabolic energy , and thus increased uptake of glucose and high rates of glycolysis are characteristics of inflammatory cell infiltration into tissues . Using microPET technology , the metabolic state of the myocardium was assessed by determining the regional uptake of the glucose analogue , 18F-FDG . The mean value of the myocardial SUVmax was used to compare the microPET data between the uninfected age-matched controls and infected groups fed on either HFD or RD 30 dpi . Tracer uptake was significantly higher in the myocardium of the RD-fed ( 14 . 5±2 . 2 ) infected mice compared with infected HFD-fed mice ( 5 . 5±1 . 03 ) and compared to uninfected control mice ( Fig . 5 a ) . We believe this increased tracer uptake indicates increased inflammation and myocarditis in the RD-fed infected mice [13] . MRI examination of the heart during both diastole and systole revealed a significant decrease in the left ventricle internal diameter ( LVID ) and an increase in the wall thickening of the ventricles during acute infection compared with uninfected mice ( Fig . 5b ) ( Table 3 ) . The alterations in heart morphology were less pronounced in the infected mice on the HFD compared to those on the RD . The right ventricle internal diameter ( RVID ) was increased in infected mice and no significant difference was observed between RD- and HFD- fed mice ( Table 3 ) . During the acute stage of infection the percent fractional shortening was significantly increased in the infected mice fed RD and likely reflects hypertrophic cardiomyopathy and possibly hypokinetic motion or dyskinesis of the heart wall [14] . Our results demonstrate an alteration in the whole body and systemic lipid homeostasis during acute infection . We further investigated the role of diet on the expression levels of scavenger receptors , lipoproteins , and the proteins involved in lipid metabolism both by qPCR and Immunoblot analysis . T . cruzi utilizes the host LDLreceptor to invade cells and invasion upregulates LDLr levels [6] , [7] . Analysis of the mRNA levels of other LDL receptors ( scavenger receptors ) , such as stabilin 1 ( Stab 1 ) , and scavenger receptor class F member 1 ( Scarf 1 ) in the heart and WAT of both the HFD and RD fed infected mice demonstrated a significant increase compared to their respective uninfected control mice ( Table 2 ) . The expression of the oxidized LDL receptor and the chemokine ( C-X-C motif ) ligand 16 ( Cxcl16 ) was significantly upregulated in WAT of infected mice ( Table 2 ) . The fold increase in the expression of Cxcl16 and Stab1 in WAT was 10-fold and 60-fold respectively , in RD-fed infected mice compared with HFD-fed infected mice . Infected heart tissue displayed a significant increase in LDLr and Scarf1 mRNA levels in both RD- and HFD-fed mice , however , a significant increase in the expression of Stab1 was observed only in RD-fed mice ( 19 fold increase ) ( Table 2 ) . Apolipoproteins play a major role in lipid metabolism and cholesterol homeostasis . We measured the mRNA levels of apolipoproteins such as ApoA , ApoB and ApoE in the heart and WATs of mice fed a HFD and RD during infection . Analysis of the mRNA levels demonstrated a significant difference in the expression levels of these genes in WAT and heart in HFD and RD fed mice ( Table 2 ) . RD-fed mice down regulated Apo A1 ( −165 fold ) and Apo B ( −54 fold ) expression in WAT and had increased expression of ApoA1 and ApoB in the heart ( 2 and 23-fold respectively ) compared to their respective uninfected mice . In these RD-fed mice Apo E was up regulated both in WAT and heart tissue . Infected HFD mice had increased Apo A1 ( 5-fold ) and Apo B1 ( 7-fold ) in the WAT and increased levels of Apo A1 , Apo B1 and Apo E in heart tissue . The expression levels of Apo A1 , Apo B1 and Apo E were significantly lower than that seen in RD-fed mice . It is likely that changes in apolioprotein levels affect systemic cholesterol homeostasis in these RD- and HFD-fed mice . Fatty acid ( FA ) transport and β-oxidation are important in the functioning of WAT and the heart . The genes responsible for FA transport , TG metabolism and β-oxidation such as acetyl-Coenzyme A acyltransferase 1A ( Acaa1a ) , mitochondrial acyl-CoA dehydrogenase family members ( Acad 9 &10 ) , acyl-CoA oxidase ( Acox1 ) , glycerol-3-phosphate dehydrogenase ( Gpd1 ) , FA acid binding protein ( Fabp1 ) and acyl-CoA synthetase bubblegum family member 1 ( Acsbg1 ) were analyzed ( Table 2 ) . WAT from the infected RD-fed mice demonstrated a significant fold decrease in the expression of these genes; however infected HFD-mice showed no significant change in the majority of these genes ( the only exceptions being Acad9 ( 2 fold ) and Acsbg1 ( 3 fold ) where there was increased expression compared to uninfected control mice ) . In the hearts of both RD and HFD fed infected mice the levels of these genes were significantly increased especially in HFD compared to control mice ( Table 2 ) . The Niemann–Pick disease type C1 ( Npc1 ) gene encodes a large protein that resides in the limiting membrane of endosomes and lysosomes and mediates intracellular cholesterol trafficking via binding of cholesterol to its N-terminal domain [15] . Analysis of the mRNA levels demonstrated an elevated expression of Npc1 mRNA ( 3-fold ) in the WAT of both the RD- and HFD-fed infected mice . No significant difference , however , was seen in Npc1 in the hearts of these mice . The gene lecithin-cholesterol acyltransferase ( LCat ) encodes the extracellular cholesterol esterifying enzyme Lcat , which is required for cholesterol esterification [14] is down regulated in infected RD-fed mice ( 7 fold ) compared to the infected HFD-fed mice ( Table 2 ) . The esterification of cholesterol is required for cholesterol transport [16] . We investigated the expression of genes involved in cholesterol efflux like ATP binding cassette ( Abc ) transporters such as Abca1 and Abcg1 in WAT and heart [17] . With cholesterol as their substrate , these proteins function as cholesterol efflux pumps in the cellular lipid removal pathway . Abca1 is significantly increased in both the WAT and heart of infected mice ( Table 2 ) . Abcg1 is mainly associated with macrophage cholesterol efflux . Abcg1 is significantly upregulated ( 1270 fold ) in the WAT of infected RD fed mice compared with infected HFD fed mice which suggests that increased macrophage infiltration is associated with RD-fed mice . We have published that there is an increased influx of macrophages in WAT during acute infection [8] , [9] . Abcg1 is significantly higher in the hearts of both HFD- and RD-fed infected mice ( Table 2 ) . Intracellular cholesterol is mainly converted to bile acids in liver . Cytochrome p450 monooxygenases like Cyp39a1 and Cyp7a1 were up regulated ( 3- and 5-fold respectively ) in WAT of HFD-fed infected mice compared with infected RD-fed mice ( Table 2 ) . Cyp39a1 and Cyp7a1are endoplasmic reticulum proteins involved in the conversion of cholesterol to bile acids [18] , [19] . Cyp7a1 is the rate limiting enzyme in the primary pathway of bile acid synthesis [19] . Even though adipocytes are not a classical bile acid synthesizing tissue , it has been shown that farnesoid x-receptor ( FXR ) a nuclear receptor which is involved in bile acid synthesis is expressed in adipose tissue during metabolic dysfunction [20] . The mRNA levels of HMG co A reductase ( Hmgcr ) , a rate limiting enzyme in the cholesterol biosynthesis is upregulated in both the WAT and the hearts of both RD-and HFD-fed mice during acute infection ( Table 2 ) . Immunoblot analysis demonstrated an upregulation of LDLr in infected mice . LDLr levels in the hearts of infected RD-fed mice were significantly higher compared to that of infected HFD-mice ( Fig . 6 ) . In WAT of infected mice the reverse was seen ( Fig . 7 ) . We also analyzed the expression of LOX1 ( oxidized LDLr ) in heart tissue and found results similar to that seen with LDLr in RD and HFD fed infected mice ( Fig . 6 ) . Hearts from infected mice displayed increased macrophage infiltration ( probed with anti-F4/80 ) , lipoprotein lipase activity ( anti-LPL ) and cholesterol efflux levels ( ant-Abca1 ) ( Fig . 6 ) . However , the expression levels of F4/80 ( −2 fold ) , LPL ( −2 . 5 fold ) and Abca1 ( −0 . 7 fold ) were lower in the infected HFD mice compared with the infected RD mice . The enzyme involved in the rate limiting step of cholesterol biosynthesis HMGCR is significantly increased ( 400-fold ) in the hearts of infected mice . In the hearts of uninfected HFD-fed mice there was an increase in HMGCR compared to uninfected RD- fed mice . This suggests that HFD induces HMGCR but that infection further increases the expression of HMGCR in heart tissue ( Fig . 6 ) . Interestingly , infected heart tissue had increased amount of adiponectin multimers , but decreased monomers ( Fig . 8 ) , whereas we observed no multimers in infected WAT . Also the amount of adiponectin monomers were decreased in infected WAT ( Fig . 7 and Fig . 8 ) . Uninfected HFD fed mice heart tissue had higher adiponectin levels compared to uninfected RD fed mice hearts which was similar to what was seen in WAT from these mice ( Fig . 7 and Fig . 8 ) .
T . cruzi has a high affinity for lipoproteins/cholesterol [7] and depends on host lipid molecules for invasion and survival [6] , [7] , [21] . Adipose tissue is the largest endocrine organ in the body and is a rich source of lipids and is involved in energy homeostasis . Previously we reported that adipose tissue is an early target of T . cruzi infection and serves as a reservoir for parasites [8] , [9] . T . cruzi infection-induced lipolysis is a hallmark of acute infection [8] , [9] and lipolysis is known to alter lipid homeostasis . Diet plays a major role in adipogenesis and in lipid homeostasis . In the present study , we systematically analyzed the impact of high fat ( HFD , 60% fat ) and regular diet ( RD , 10% fat ) on an acute model of Chagas disease and demonstrated a link between diet , adipogenesis and myocarditis . HFD increased adipogenesity and reduced lipolysis which affected peripheral parasitemia and parasite load in the heart . Even though a significant difference in the parasite load of heart was demonstrated between the RD- and HFD- fed infected mice , the adipose tissue of HFD fed mice had a significantly higher number of parasites . We believe the increased fat tissue in HFD mice resulted in a sequestration of parasites ( i . e . a sponging effect ) that may have led to a reduction in the parasite load in the heart . Consistent with this hypothesis , HFD-fed mice displayed an increased survival rate ( 95% ) with diminished myocardial damage during acute infection compared with RD-fed mice . HFD induced a modest obese condition ( as seen with uninfected mice ) may bring metabolic changes during infection . Overall , this increase in survival with HFD supports the “so-called” obesity paradox hypothesis [21] . The various techniques we employed to investigate cardiac structure and function clearly demonstrated that in HFD-fed mice there was a significant amelioration of myocardial dysfunction compared to RD-fed mice . This is likely due to the role of adipogenesis and lipolysis during infection . Lipolysis is a characteristic marker of acute infection in mice where there is a significant decrease in total body fat between d10 and d35d pi . HFD-fed mice increased adipogenesis and reduced the rate of adipocyte lipolysis compared to RD-fed mice . The parasite load in the WAT was higher in these HFD mice compared to RD . The pro-inflammatory marker TNF-α was higher in the WAT of HFD-fed mice at this time point of infection . Adipocyte lipolysis was significantly higher in the infected RD-fed mice as indicated by serum triglyceride and fatty acid levels . Overall , this change in fatty acid metabolism probably contributes to increased parasitemia in RD-fed mice and a higher parasite load in heart . Serum cholesterol levels were reduced with a concomitant increase in intra-organelle LDL/cholesterol levels ( Fig . 3 ) . We have demonstrated the accumulation of LDL/cholesterol in adipose tissue and the hearts [22] during infection . Accumulated lipids need to be degraded ( through lipases ) before undergoing further degradation through the β-oxidation pathway . Increased lipolysis activation through LPL was observed both in the hearts and WAT of infected mice . There was a significant increase in the LPL expression in the hearts of RD-fed mice compared with HFD-fed mice suggesting that more lipids accumulated in RD-fed mice due to increased lipolysis and increased parasite load . Hormone sensitive lipase ( Lipe ) is significantly down regulated only in the WAT of RD-fed mice . No significant change was observed in the hearts of either RD- or HFD-fed mice at d35pi . FA transport and β-oxidation are important signaling pathways in the functioning of WAT and heart . White adipocytes are not mitochondrial rich cells , unlike brown adipocytes which are mitochondrial rich , and the elevation of infection induced lipids may cause a burden on mitochondrial oxidative capacity in WAT leading to necrosis . qPCR demonstrated a down-regulation of the mRNA levels of many of the genes involved in triglyceride and FA metabolism which reflects the oxidative state in WAT of RD-fed mice . Previously , we reported a significant loss of WAT during acute infection [8] , [9] . The heart , however , could sustain this load as it is rich in mitochondria and we observed increased FA and triglyceride metabolism both in the RD- and HFD-fed mice . Interestingly , it has been recently reported that host FA metabolism is essential for the persistence of T . cruzi amastigotes [23] . Decreased expression of Insig1 , a regulator of cholesterol biosynthesis through SREBP in both the hearts and WAT of RD fed mice is also responsible for an increased HMGCR expression and cholesterol biosynthesis . De novo cholesterol biosynthesis is highly regulated and depends on intracellular cholesterol levels [24] . When there is a depletion in intracellular cholesterol , cells respond with a SREBP-activated increase in HMGCR and LDLr expression which results in endogenous cholesterol biosynthesis and LDL-mediated uptake of cholesterol . In the present study , we found increased SREBP levels and increased HMGCR and LDLr protein levels in the tissues when the cells already had elevated intracellular LDL/cholesterol levels due to parasite invasion suggesting that infection results in dysfunctional cholesterol homeostasis in tissues and changes in the regulation of these key cholesterol homeostasis genes . Expression of the ABC transporters is highly upregulated during macrophage differentiation and cholesterol loading , and they synergize to mediate cholesterol transport to Apo-A1 [25] . The expression of these genes was significantly altered during infection . Adiponectin , an adipokine secreted by the WAT , plays an important role in regulating glucose and lipid metabolism and controlling energy homeostasis in insulin-sensitive tissues [26] . Adiponectin is considerably reduced in the WAT , but , importantly produced in the heart during acute infection . Previously , we demonstrated a reduction in serum adiponectin levels at 15 days post infection ( dpi ) returning to normal by 30 dpi [8] , [9] . WAT ( fat cells ) significantly decreases during infection and thus , the amount of WAT in infection may not be sufficient to maintain normal serum adiponectin . As the adiponectin multimers are functionally active , further studies are required to confirm that the heart secretes adiponectin during T . cruzi infection , the mechanism by which this occurs and the significance of elevated adiponectin levels in the heart during acute infection . Lipogenic and adipogenic markers such as leptin , adiponectin , adipogenin and PPAR are greatly reduced in the WAT of RD-fed mice compared to HFD-fed mice , but significantly increased in the heart especially in the HFD-fed mice . This suggests that there are tissue specific of acute T . cruzi infection related to the altered lipolysis and lipogenic status of different tissues . Previously , we demonstrated the role of host LDLr in T . cruzi invasion [6] . qPCR analysis revealed ( Table 2 ) the association of other LDL receptors and modified LDL receptors such as classical LDLr , very low density lipoprotein receptor ( VLDLr ) , STAB1 , CXCL16 and SCARF1 in the hearts and WAT during acute infection . The significant increase and the decrease in the LDLr protein levels in the WAT and heart respectively , of the HFD-fed infected mice compared with RD-fed mice , suggests that WAT of HFD-fed mice and the hearts of RD-fed mice are targets of T . cruzi during the late phase of the acute infection . Thus , an increased parasite load in the hearts of RD-fed mice resulted in elevated cardiac LDL/cholesterol levels , macrophage infiltration ( F4/80 staining ) and inflammation ( TNF-α ) compared with HFD-fed mice . Overall , it is clear that HFD modulates the effect of T . cruzi infection on myocarditis and mortality in this acute model and this was confirmed by microPET and MRI analyses . HFD is known to alter the metabolic state of the host leading to diabetes and obesity . HFD induces the metabolic syndrome and this alteration in the host affects the pathogenesis of acute Chagas disease resulting in a decreased heart parasite burden and an increased survival rate . This observation in the mouse model is consistent with the “obesity paradox” that has been described for some chronic infections reflecting an observation that obesity can have a positive effect on disease outcome . To this end , the metabolic syndrome may have evolved as a response to “times of plenty” when the extra calories could be used for metabolic changes that would allows the host to better deal with chronic infectious diseases such as T . cruzi [21] . Herein , we report , for the first time on alterations in the lipid signaling net work due to diet , adipogenesis , and lipolysis in the setting of acute T . cruzi infection . We believe that these alterations contribute to the pathogenesis of acute Chagas disease . The rate of survival and the severity of the myocardial damage are related to the adipogenic and the lipolytic status of adipose tissue and the heart . Lipid and cholesterol homeostasis is completely altered by the infection which warrants further mechanistic studies to understand the pathogenic role of LDL/cholesterol in the progression of Chagasic cardiomyopathy which can now be considered , in part , to be a lipidopathy ( onset of cardiomyopathy due to abnormal intracellular lipid level ) .
|
Infection with Trypanosoma cruzi , the etiologic agent in Chagas disease , may result in heart disease . There has been an increase in obesity , diabetes , hypertension and ischemic cardiovascular disease in endemic areas . Previously , we demonstrated that adipose tissue is an early target and a reservoir for T . cruzi . T . cruzi has high affinity for lipoproteins , and that infected tissues there is an increase in intra-cellular cholesterol levels . It is likely that adipocytes and lipoproteins play a key role in the pathogenesis of Chagas disease . The role of host lipids in the pathogenesis of Chagas disease is understudied . Diet plays a major role in the regulation of systemic and whole body lipid levels including adipogenesis and lipogenesis . We report , for the first time , the effect of diet on myocardial inflammation and damage observed during acute T . cruzi infection and provide data on the role of parasite associated LDL/HDL in the regulation of systemic lipid homeostasis in white adipose tissue ( WAT ) and in the heart . Interestingly , we demonstrate that a high fat diet protects mice from the consequences of infection-induced myocardial damage through effects on adipogenesis in adipose tissue and reduced cardiac lipidopathy .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases",
"dyslipidemia",
"veterinary",
"diseases",
"zoonoses",
"medicine",
"and",
"health",
"sciences",
"nutrition",
"biology",
"and",
"life",
"sciences",
"vector-borne",
"diseases",
"diet",
"metabolic",
"disorders",
"veterinary",
"science"
] |
2014
|
High Fat Diet Modulates Trypanosoma cruzi Infection Associated Myocarditis
|
Upon viral infection , retinoic acid–inducible gene I–like receptors ( RLRs ) recognize viral RNA and trigger a series of signaling events , leading to the induction of type I interferons ( IFNs ) . These processes are delicately regulated to prevent excessive and harmful immune responses . In this study , we identified G patch domain-containing protein 3 ( GPATCH3 ) as a negative regulator of RLR-mediated antiviral signaling pathways . Overexpression of GPATCH3 impaired RNA virus- triggered induction of downstream antiviral genes , whereas its knockdown had opposite effects and attenuated viral replication . In addition , GPATCH3-deficient cells had higher IFNB1 mRNA level compared with control cells after RNA virus infection . Mechanistically , GPATCH3 was recruited to VISA in a viral infection dependent manner and the assembly of VISA/TRAF6/TBK1 signalosome was impaired in GPATCH3-overexpressing cells . In contrast , upon viral infection , the recruitment of TRAF6 and TBK1 to VISA was enhanced in GPATCH3 deficient cells . Taking together , our findings demonstrate that GPATCH3 interacts with VISA and disrupts the assembly of virus-induced VISA signalosome therefore acts as a negative regulator of RLR-mediated innate antiviral immune responses .
Recognition of conserved molecular structures of viruses by the host pattern-recognition receptors ( PRRs ) initiates innate antiviral immune responses . Several families of PRRs , including Toll-like receptors ( TLRs ) , RIG-like receptors ( RLRs ) , NOD-like receptors ( NLRs ) and recently identified DNA sensors , have been shown to sense different microbial components [1–3] . It has been demonstrated that cytosolic RNAs derived from viral genome or its replication intermediates are mainly recognized by the RLR family members retinoic acid–inducible gene-1 ( RIG-I ) and melanoma differentiation-associated gene 5 ( MDA5 ) [4 , 5] . The binding of RLRs to viral RNAs leads to rapid activation of transcription factors IRF3 and NF-κB , which collaborate to induce transcription of type I interferon ( IFN ) genes . Type I IFNs then activates JAK-STAT signaling pathways to induce the transcription of a wide range of interferon stimulated genes ( ISGs ) leading to innate antiviral responses [6] . Upon recognition of viral RNA , RIG-I and MDA5 undergo conformational changes and are recruited to the downstream mitochondria-associated adaptor protein VISA ( also known as MAVS , IPS-1 , Cardif ) [7–11] . VISA then forms large prion-like polymers and serves as platform to assemble VISA signalosome by recruiting multiple components , including TRAF proteins ( TRAF2/3/5/6 ) , TBK1 and IKK kinases [12–14] . VISA recruits the TRAF proteins through its TRAF-binding motifs [7 , 9 , 15 , 16] , which in turn recruit TBK1 and IKKs to the VISA-associated signalosome , in which TBK1 and IKKs phosphorylate IRF3 and NF-κB , respectively leading to the induction of type I IFNs and proinflammatory cytokines [12 , 13] . In addition to mitochondria , VISA has also been found to localize to peroxisomes [17] . Upon viral infection , peroxisomal VISA rapidly mounts a short-term protection by initiating the production of type III IFNs [18] . Several studies have demonstrated that dysregulation of RLR signaling is associated with autoimmune diseases [19–22] . For example , loss of function of MDA5 is associated with resistance to type I diabetes , whereas gain of function of MDA5 leads to spontaneously developed lupus-like nephritis and systemic autoimmune symptoms in mouse model [21 , 23] . As a central adaptor of RLR-mediated signaling , VISA is expected to be tightly regulated by host factors [24] . It has been shown that WDR5 plays an important role in the assembly and stability of VISA signalosome [25]whereas TOM70 and IFIT3 facilitate VISA-mediated signaling by linking TBK1 to VISA [26 , 27] . In contrast , it has been demonstrated that the virus-induced protein UBXN1 inhibits VISA-mediated signaling by interfering the assembly VISA signalosome [28] . In this study , we identified G patch domain-containing protein 3 ( GPATCH3 ) as a negative regulator of RLR-mediated innate antiviral responses . We found that deficiency of GPATCH3 significantly enhanced RNA virus-triggered induction of type I IFNs and attenuated viral replication . Biochemically , GPATCH3 was recruited to VISA in a viral infection dependent manner where it disrupts the assembly of VISA/TRAF6/TBK1 complexes . Our findings suggest that GPATCH3 negatively regulates the assembly of VISA signalosome and shed first light on the biological function of GPATCH3 .
RLRs recognize viral RNAs and initiate signal transductions leading to the induction of innate antiviral immune responses . To identify potential molecules involved in RLR-mediated signaling , we screened ~10 , 000 human cDNA expression plasmids for their abilities to regulate SeV-triggered induction of type I IFNs by luciferase assays . These efforts led to the identification of GPATCH3 , a G-patch domain containing protein , as a negative regulator of RLR-mediated signaling ( S1A Fig ) . GPATCH3 is ubiquitously expressed in mammalian tissues ( EST profile from NCBI ) . However , the function of GPATCH3 has not yet been characterized . In reporter assays , overexpression of GPATCH3 reduced SeV-triggered activation of IFN-β promoter , ISRE and NF-κB in a dose-dependent manner in 293T cells ( Fig 1A ) . However , in similar experiments , overexpression of GPATCH3 showed little effects on TNFα-triggered activation of NF-κB even at high dosage ( Fig 1B ) . These data indicate that GPATCH3 specifically inhibits RLR-mediated signaling . To further investigate the roles of endogenous GPATCH3 in RLR-mediated signaling , we constructed three GPATCH3-shRNA plasmids targeting different sequences of human GPATCH3 mRNA and examined their effects on the expression of GPATCH3 . As shown in Fig 2A , three GPATCH3-shRNA constructs could knockdown the expression of GPATCH3 to different levels . The 2# GPATCH3-shRNA markedly inhibited the expression of GPATCH3 , whereas the 1# and 3# GPATCH3-shRNA had partial or little effects on the expression of GPATCH3 respectively . In reporter assays , knockdown of GPATCH3 significantly potentiated SeV-induced activation of the IFN-β promoter , ISRE and NF-κB . Notably , the degrees of positive regulation of the IFN-β promoter , ISRE and NF-κB activation were correlated with the knockdown efficiencies of the respective shRNA plasmids ( Fig 2B ) . We then used the 2# GPATCH3-shRNA for the following experiments . Similar results were obtained with the 1# GPATCH3-shRNA . Consistent with the results of reporter assays , quantitative PCR ( qPCR ) experiments showed that knockdown of GPATCH3 significantly enhanced SeV-induced transcription of IFNB1 , ISG15 , ISG56 , CXCL10 and TNFA ( Fig 2C ) . Furthermore , knockdown of GPATCH3 enhanced SeV-induced phosphorylation of TBK1 and IRF3 ( Fig 2D ) , which are hallmarks of activation of the RLR-mediated signaling . These data suggest that endogenous GPATCH3 negatively regulates RLR-mediated signaling . To avoid the potential off-target effects of the GPATCH3-shRNA , we then performed reconstitution experiments . We generated a shRNA-resistant mutant of GPATCH3 ( mGPATCH3 ) in which silent mutations were introduced into the targeting sequence of the GPATCH3-shRNA without changing the amino acid sequence . In reporter assays , the potentiation of SeV-induced activation of the IFN-β promoter , ISRE and NF-κB by the GPATCH3-shRNA was restored by the mGPATCH3 mutant ( S2A Fig ) , indicating that the enhancement of RLR-mediated signaling was indeed caused by knockdown of GPATCH3 . To further confirm the function of endogenous GPATCH3 in other cell types , we generated two GPATCH3 siRNAs with different targeting sequences from GPATCH3-shRNAs and examined whether knockdown of GPATCH3 affects SeV-triggered induction of type I IFNs in A549 cells as well as in primary cells such as human foreskin fibroblasts ( HFFs ) and peripheral blood mononuclear cells ( PBMCs ) . Consistently , knockdown of GPATCH3 significantly enhanced SeV-induced transcription of IFNB1 , ISG15 , ISG56 and TNFA in A549 cells ( Fig 2E ) . Moreover , knockdown of GPATCH3 potentiated SeV-induced transcription of IFNB1 in HFFs and PBMCs ( Fig 2F , left two panels ) . In contrast , knockdown of GPATCH3 had little effects on DNA virus HCMV- or double strand DNA HSV120-triggered transcription of IFNB1 ( Fig 2F , right panel ) . In addition , when cotransfected with cGAS and MITA ( also known as STING ) , GPATCH3 had no significant effects on DNA sensor cGAS-mediated activation of ISRE and NF-κB ( S2B Fig ) . Since SeV-triggered induction of type I IFNs can be mediated by both RIG-I and MDA5 , we next investigated the roles of GPATCH3 in RIG-I- and MDA5-mediated signaling . In reporter assays , knockdown of GPATCH3 markedly potentiated activation of the IFN-β promoter , ISRE and NF-κB triggered by transfected low and high molecular weight poly ( I:C ) ( S2C & S2D Fig ) . It has been reported that RIG-I and MDA5 selectively recognize low and high molecular weight poly ( I:C ) respectively [29] . Therefore , these data suggest that GPATCH3 negatively regulates both RIG-I- and MDA5-mediated signaling . As a control , knockdown of GPATCH3 had no effects on TLR3-mediated signaling which was induced by poly ( I:C ) added to the cell culture media ( S2E Fig ) . Taken together , these data reveal critical negative regulatory roles of GPATCH3 in RLR- but not DNA sensor- or TLR3-mediated signaling . To further confirm these results , we generated GPATCH3-deficient 293T cells by CRISPR-Cas9 technology using a gRNA targeting the exon 1 of the Gpatch3 gene . Knockout of GPATCH3 was confirmed in two independent 293T clones ( GPATCH3-KO 1#&2# ) by immunoblotting analysis ( Fig 3A ) . As shown in Fig 3B , compared with the control cells , GPATCH3-deficiency significantly enhanced transcription of IFNB1 induced by SeV , VSV and cytosolic poly ( I:C ) . Intriguingly , when we reconstituted the expression of GPATCH3 in GPATCH3-deficient cells , the over-activated transcription of IFNB1 was restored ( Fig 3C ) . Compared with the knockout cells , reconstitution of GPATCH3 into wild type cells resulted in lower induction of IFNB1 because the total amount of GPATCH3 in wild type cells , which endogenously express GPATCH3 , was higher than that of the knockout cells ( Fig 3C ) . The 1# GPATCH3-KO clone was further used to detect SeV-induced phosphorylation of TBK1 , IRF3 and IKK α/β , which are hallmarks of their activation . As shown in Fig 3D , knockout of GPATCH3 markedly increased SeV-induced phosphorylation of TBK1 , IRF3 and IKKα/β whereas reconstitution of GPATCH3 in the knockout cells reversed such increases . These data indicate that GPATCH3 plays an important role in the negative regulation RLR-mediated signaling . Since GPATCH3 negatively regulates RLR-mediated induction of type I IFNs , we next examined whether GPATCH3 affected cellular antiviral responses . We measured replications of SeV and VSV by immunoblotting analysis using antibodies against viral proteins . As shown in S3A Fig , overexpression of GPATCH3 resulted in higher levels of SeV and VSV proteins . Consistently , knockdown of GPATCH3 markedly impaired replications of both SeV and VSV at all examined time points post infection ( Fig 4A ) . To confirm these results , replication of VSV was further measured by immunofluorescence microscopy of VSVs tagged with GFP and plague assays . Data from these experiments showed that knockdown of GPATCH3 resulted in decreased replication of VSV , as indicated by the less green fluorescence ( Fig 4B ) as well as the lower virus titers ( Fig 4C ) in the GPATCH3-shRNA transfected cells , suggesting a more robust antiviral response in GPATCH3-knockdown cells . These observations suggest that GPATCH3 negatively regulates cellular antiviral responses . To identify the potential regulatory targets of GPATCH3 , we examined the effects of GPATCH3 on activation of the IFN-β promoter , ISRE and NF-κB mediated by components of the RLR signaling pathways . The results showed that cotransfection of GPATCH3 reduced activation of the IFN-β promoter , ISRE , and NF-κB mediated by VISA and its upstream components such as RIG-I-CARD and MDA5 but had no obvious effects on activation mediated by proteins downstream of VISA such as TBK1 , IRF3-5D ( a constitutively active mutant of IRF3 ) , TRAF6 , TAK1 , TAB1 , IKKβ or p65 ( Fig 5A ) . Transient transfection and coimmunoprecipitation experiments showed that GPATCH3 interacted with VISA but not RIG-I , MDA5 or IRF3 ( Fig 5B ) . Endogenous coimmunoprecipitation experiments indicated that GPATCH3 weakly interacted with VISA in un-infected cells , but their interaction was markedly increased following SeV infection ( Fig 5C ) . These results suggest that GPATCH3 targets VISA for its inhibitory effects on RLR-mediated signaling . We generated a series of GPATCH3 truncations and examined their abilities to interact with VISA . Domain mapping analysis indicated that no specific domain of GPATCH3 examined was responsible for its association with VISA ( Fig 5D ) . A reasonable explanation is that a proper spatial conformation of full-length GPATCH3 is required for its binding to VISA . When tested for their abilities to inhibit VISA-mediated activation of the IFN-β promoter , only the truncations of GPATCH3 that interacted with VISA impaired VISA-mediated signaling ( Fig 5E ) . Taken together , these date suggest that GPATCH3 interacts with VISA following viral infection and its physical binding to VISA is sufficient and necessary for the inhibitory function . Domain mapping analysis was also performed to test which domains of VISA are responsible for its interaction with GPATCH3 . As shown in Fig 5F , GPATCH3 associated with full-length VISA as well as truncations of VISA that contain the C-terminal transmembrane ( TM ) domain . In contrast , GPATCH3 failed to interact with VISA truncations lacking its TM domain . These results indicate that the TM of VISA is required for its binding with GPATCH3 . It has been reported that VISA is located on both mitochondrial and peroxisomal membranes through its C-terminal transmembrane domain [17 , 18 , 24] . Since the transmembrane domain of VISA is responsible for its interaction with GPATCH3 , we next investigated whether GPATCH3 targets mitochondria or peroxisome localized VISA . We generated chimeric VISA expression plasmids by substituting the TM domain of VISA with the TM domain of Pex13 ( VISA-Pex ) , a protein solely locates on peroxisomes , or with the TM domain of OMP25 ( VISA-Mimic ) , a protein which was originally reported to localize on mitochondria [30] . Localizations of VISA-Pex to peroxisomes have already been demonstrated [17] . Unexpectedly , localization of VISA-Mimic have been found on both peroxisomes and mitochondria , which mimics the localization of wild-type VISA [17] . Transient transfection and coimmunoprecipitation experiments indicated that localization of VISA to peroxisomes completely abolished its binding to GPATCH3 ( Fig 6A ) . However , VISA-Mimic could still interact with GPATCH3 ( Fig 6A ) . These data suggest that GPATCH3 interacts with mitochondria-localized VISA . Since mitochondrial and peroxisomal VISA have been suggested to mediate the production of type I and type III IFNs respectively[17] , we attempted to determine whether GPATCH3 is involved in VISA-mediated production of type III IFNs . As shown in Fig 6B , overexpression of GPATCH3 had no significant effects on VISA or VISA-Pex- mediated activation of IFN-λ1 reporter . Moreover , overexpression of GPATCH3 did not affect SeV-induced activation of IFN-λ1 reporter ( Fig 6C ) . Consistently , knockdown of GPATCH3 had no significant effects on SeV-induced transcription of IL-29 ( Fig 6D ) . Collectively , these data suggest that GPATCH3 is not involved in peroxisomal VISA-mediated induction of type III IFNs . Since GPATCH3 specifically interacts with and inhibits the function of mitochondrial VISA , we next examined whether GPATCH3 localized on mitochondria by cellular fractionation assays . The results showed that GPATCH3 was localized in nucleus , cytosol and mitochondria ( S4A Fig ) . Intriguingly , the subcellular distribution of GPATCH3 did not change after virus infection ( S4A Fig ) . These data indicated that GPATCH3 was constitutively localized in mitochondria and was recruited to VISA after virus infection . To further confirm these results , we isolated mitochondria by subcellular fractionation and performed endogenous coimmunoprecipitation experiments with mitochondrial lysates . Consistently , the results showed that although GPATCH3 constitutively localized on mitochondria , it was recruited to VISA in a virus infection dependent manner ( Fig 6E ) . Collectively , these data suggest that GPATCH3 targets mitochondrial VISA to negatively regulate RLR-mediated production of type I IFNs . We next investigated the mechanism of GPATCH3 mediated inhibition of VISA . Since aggregation of VISA is important for its activation , we first examined whether GPATCH3 inhibited VISA aggregation . As shown in S5A Fig , results of coimmunoprecipitation experiments showed that overexpression of GPATCH3 had no effects on VISA oligomerization which is the first step of VISA aggregation . Furthermore , we examined the role of endogenous GPATCH3 on VISA aggregation by Semi-Denaturating Detergent Agarose Gel Electrophoresis ( SDD-AGE ) . Consistently , GPATCH3 deficiency had no significant effects on VISA aggregation ( S5B Fig ) . Collectively , these data suggest that GPATCH3 does not target the activation of VISA . We next determined whether GPATCH3 disrupted the association of VISA with downstream signaling components . We cotransfected VISA with its downstream signaling proteins such as TBK1 , TRAF6 or TRAF3 , together with an increased amount of GPATCH3 . The results of coimmunoprecipitation experiments indicated that interactions of VISA-TBK1 and VISA-TRAF6 were markedly reduced by GPATCH3 in a dose dependent manner whereas the interaction of VISA-TRAF3 was not affected ( Fig 7A ) . Furthermore , we examined the roles of endogenous GPATCH3 on assembly of the VISA signalosomes . Deficiency of GPATCH3 had no marked effects on the protein levels of VISA , TBK1 or TRAF6 . However , compared with the wild-type cells , GPATCH3-deficiency markedly enhanced recruitment of TBK1 and TRAF6 to the VISA signalosomes ( Fig 7B ) . These data suggest that GPATCH3 interferes with the assembly of VISA signalosomes , leading to the negative regulation of the RLR-mediated signaling .
The function of the G-patch domain containing protein GPATCH3 has been elusive . In this study , we identified GPATCH3 as a negative regulator of innate immune responses to RNA viruses . Knockdown of GPATCH3 significantly enhanced SeV-triggered induction of downstream antiviral genes in multiple cell lines , including primary PBMCs and HFFs but had no marked effects on TLR3- or DNA sensor-mediated signaling . GPATCH3-deficient cells showed higher induction of IFNB1 compared with wild-type cells upon SeV or VSV infection . Mechanistically , GPATCH3 interacted with VISA and impaired assembly of the VISA-associated signaling complexes . It has been demonstrated that VISA is essential for innate antiviral immune responses against to RNA viruses [31] . Previous studies suggest that VISA recruits TRAF3 and TRAF6 to activate IRF3 and NF-κB respectively [7 , 9 , 15] . However , recent gene knockout studies indicated that TRAF3 is dispensable for TBK1 and IRF3 activation whereas TRAF6 plays an important role to active both IRF3 and NF-κB [12 , 32] . In this study , we found that GPATCH3 interrupted the binding of VISA to TRAF6 but not TRAF3 and inhibited RLR-mediated activation of both IRF3 and NF-κB , which is consistent with the conclusion of gene knockout studies . Interestingly , different from the previous discovery that UBXN1 negatively regulates the function of VISA by competing for the TRAF-binding motifs of VISA to block recruitment of TRAF3 and TRAF6 [28] , our findings demonstrated that the binding of GPATCH3 to VISA and its blockade of assembly of the VISA-signalosome was independent of TRAF-binding motifs of VISA ( aa143-147 for TRAF2/3/5 , aa153-158 and aa455-460 for TRAF3/6 ) but dependent on the membrane localization of VISA . VISA is located on mitochondria and peroxisomes and the distinct cellular localization leads to different types of antiviral responses [17 , 18] . Mitochondria-localized VISA induces long term antiviral response through the expression of type I IFNs . In contrast , peroxisome-localized VISA induces the expression of type III IFNs . Notably , the C-terminal TM of VISA dictates its localization to different organelles [17 , 24 , 33] . In this study , we found that GPATCH3 specifically interacted with the mitochondrial VISA , which was sufficient and required for it to execute the inhibitory roles on RLRs-mediated signaling . The G-patch domain-containing proteins are widely found in eukaryotes [34] . A previous study has shown that the G-patch containing protein MOS2 is essential for innate immunity in Arabidopsis thaliana [35] . Different from the function of MOS2 as a positive regulator of innate immunity , we found that GPATCH3 acted as a negative regulator of RLR-mediated innate antiviral responses . Notably , while we found that the interaction of GPATCH3 with VISA was essential for its inhibitory roles , the binding of GPATCH3 to VISA was independent of its G-patch domain and a GPATCH3 mutant lack of the G-patch domain inhibited SeV-induced activation of the IFN-β promoter to similar degrees as the full-length GPATCH3 . These data suggested that the G-patch domain of GPATCH3 is not required for its inhibitory roles on innate antiviral responses . In conclusion , our findings suggest that GPATCH3-mediated disruption of VISA-associated complexes represents an important regulatory mechanism of innate antiviral responses .
HFFs and HCMV were provided by Dr . Min-Hua Luo ( Wuhan Institute of Virology , CAS ) . The 293 cells stably expressing TLR3 ( 293-TLR3 ) were provided by Katherine Fitzgerald ( University of Massachusetts Medical school , Worcester , MA ) and Tom Maniatis ( Department of Molecular and Cellular Biology , Harvard University , Cambridge , MA ) . HEK293T ( Transformed Human Embryonic Kidney 293 cell line , ATCC ) cells , A549 ( Human Lung Adenocarcinoma cell line , ATCC ) cells , PBMCs ( Peripheral Blood Mononuclear Cells , Allcells ) , Human recombinant TNFα ( R&D Systems ) , poly ( I:C ) ( InvivoGen ) , Lipofectamine 2000 ( Invitrogen ) , dual-specific luciferase assay kits ( Promega ) , mouse monoclonal antibodies against Flag and β-actin ( Sigma ) , HA ( Origene ) , AIF , LMNB1 and TBK1 ( Abcam ) , IRF-3 ( Proteintech ) , VISA ( Bethyl Laboratories ) , TRAF6 and GPATCH3 ( Y-20 ) ( Santa Cruz Biotechnology ) , phospho-IRF3 ( Ser396 ) , phospho-TBK1 ( Ser172 ) and phospho-IKKα ( Ser176 ) /β ( Ser177 ) ( Cell Signaling Technology ) were purchased from the indicated companies . Rabbit polyclonal anti-SeV and anti-VSV were previously described [36] . Mouse polyclonal antisera against GPATCH3 were raised against recombinant human C-terminal GPATCH3 fragment ( aa 299–525 ) . IFN-β promoter , ISRE and NF-κB luciferase reporter plasmids , mammalian expression plasmids for HA- or Flag-tagged RIG-I-CARD , MDA5 , VISA and its mutants , TBK1 , IRF3 , TRAF6 , TRAF3 , TAK1 , TAB1 , IKKβ , p65 were previously described [7 , 36–38] . The IFN-λ1 luciferase reporter plasmid was provided by Dr . Ying Zhu ( Wuhan University ) . Mammalian expression plasmids for human HA- , Flag-tagged GPATCH3 and its deletion mutants were constructed by standard molecular biology techniques . HSV120: 5’-AGACGGTATATTTTTGCGTTATCACTGTCCCGGATTGGACACGGTCTTGTGGGATAGGCATGCCCAGAAGGCATATTGGGTTAACCCCTTTTTATTTGTGGCGGGTTTTTTGGAGGACTT-3’ . HEK293T cells or 293-TLR3 cells ( 1x105 ) were seeded on 48-well plates and transfected the following day by standard calcium phosphate precipitation . Empty control plasmid was added to ensure that each transfection receives the same amount of total DNA . To normalize for transfection efficiency , 0 . 005 μg of pRL-TK Renilla luciferase reporter plasmid was added to each transfection . Luciferase assays were performed with a dual-specific luciferase assay kit ( Promega ) . Firefly luciferase activities were normalized on the basis of Renilla luciferase activities . Double-strand oligonucleotides corresponding to the target sequences were cloned into the pSuper-Retro plasmid ( Oligoengine ) . The following sequences were targeted for human GPATCH3 mRNA: 1#: 5’-GCAAGCGTGGATTGGGGTA-3’; 2#: 5’-CCTACCTGGCAGATATACC-3’; 3#: 5’-GTGAAGAAATACCCCAAGG-3’ . Small interfering RNAs ( siRNAs ) targeting human GPATCH3 were purchased from Ribobio . 1# siRNA: 5’-GGAACAGAGACTCCGAGAT-3’ , 2# siRNA: 5’-GTACCATGGAGAGAAGCTA-3’ . siRNA were delivered into A549s and HFFs by PepMute siRNA transfection reagent ( SignaGen ) according to procedures recommended by the manufacturer . Genome engineering using the CRISPR-Cas9 technology was previously described [39 , 40] . pGL3-U6-gRNA and pST1374-Cas9-D10A plasmids were provided by Dr . Xiao-Dong Zhang ( Wuhan University ) . GPATCH3 gRNA targeting sequence: 5’-CGCCGCGCTCTTCTCGGAAC-3’ . Total RNA was isolated and reversed transcription to cDNA for quantitative real-time PCR analysis to measure the mRNA levels of tested genes . GAPDH was used as a reference gene . Human gene specific primer sequences were as follows: GAPDH: 5’-GAGTCAACGGATTTGGTCGT-3’ ( forward ) , 5’-GACAAGCTTCCCGTTCTCAG-3’ ( reverse ) ; IFNB: 5’-TTGTTGAGAACCTCCTGGCT-3’ ( forward ) , 5’-TGACTATGG TCCAGGCACAG-3’ ( reverse ) ; ISG15: 5’-AGGACAGGGTCCCCCTTGCC-3’ ( forward ) , 5’- CCTCCAGCCCGCTCACTTGC-3’ ( reverse ) ; ISG56: 5’-TCATCAGGT CAAGGATAGTC-3’ ( forward ) , 5’-CCACACTGTATTTGGTGTCTAGG-3’ ( reverse ) ; IP-10: 5’- GGTGAGAAGAGATGTCTGAATCC-3’ ( forward ) , 5’- GTCCATCCTTGG AAGCACTGCA-3’ ( reverse ) ; TNFa: 5’-GCCGCATCGCCGTCTCCTAC-3’ ( forward ) , 5’- CCTCAGCCCCCTCTGGGGTC-3’ ( reverse ) ; IL-29: 5’-CGCCTTGGAAGAGTCACTCA-3’ ( forward ) , 5’-GAAGCCTCAGGTCCCAATTC-3’ ( reverse ) ; and GPATCH3: 5’-TGGCTGGATT CTCACGGGACTT-3’ ( forward ) , 5’- GGTGAAGGCTTCATTCTCTGCC-3’ ( reverse ) . Cells ( 5 x 106 for overexpression experiments and 2 x 107 for endogenous experiments ) were lysed in l ml NP-40 lysis buffer ( 20 mM Tris-HCl [pH 7 . 4] , 150 mM NaCl , 1mM EDTA , 1% Nonidet P-40 , 10 mg/ml aprotinin , 10 mg/ml leupeptin , and 1 mM phenylmethylsulfonyl fluoride ) . Coimmunoprecipitation and immunoblot analysis were performed as previously described [41] . These experiments were performed as described [42] . The experiments were performed with subcellular fractionation kits ( ApplyGene ) following protocols recommended by the manufacture . SDD-AGE was performed as described before [14 , 43] . In brief , mitochondria were resuspended in 1x sample buffer ( 0 . 5× TBE , 10% glycerol , 2% SDS , and 0 . 0025% bromophenol blue ) and loaded onto a vertical 1 . 5% agarose gel ( Bio-Rad ) . After electrophoresis in the running buffer ( 1× TBE and 0 . 1% SDS ) for 45 min with a constant voltage of 100 V at 4°C , the proteins were transferred to Immobilon membrane ( Millipore ) for immunoblot .
|
Virus infection triggers the host cells to produce type I IFNs and proinflammatory cytokines , which are secreted proteins important for the host to clear viruses . Previously , we identified VISA ( also named as MAVS , IPS-1 and Cardif ) as a critical adaptor of virus-triggered , RLR-mediated induction of innate antiviral responses . In this study , we further found that GPATCH3 , a functionally uncharacterized protein , interacted with mitochondria-localized VISA upon virus infection and disrupted the assembly of VISA-signalosome . Therefore , GPATCH3 acts as a negative regulator of VISA and functions as a brake of RLR-mediated antiviral innate responses . This discovery helps to understand how the innate antiviral responses are delicately regulated .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"transfection",
"luciferase",
"assay",
"molecular",
"probe",
"techniques",
"293t",
"cells",
"biological",
"cultures",
"immunoblotting",
"plasmid",
"construction",
"biochemical",
"analysis",
"enzyme",
"assays",
"immunoprecipitation",
"signalosomes",
"mitochondria",
"molecular",
"biology",
"techniques",
"bioenergetics",
"dna",
"construction",
"co-immunoprecipitation",
"bioassays",
"and",
"physiological",
"analysis",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"proteins",
"cell",
"lines",
"molecular",
"biology",
"precipitation",
"techniques",
"biochemistry",
"protein",
"complexes",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"energy-producing",
"organelles"
] |
2017
|
GPATCH3 negatively regulates RLR-mediated innate antiviral responses by disrupting the assembly of VISA signalosome
|
Sexual dimorphism in common disease is pervasive , including a dramatic male preponderance in autism spectrum disorders ( ASDs ) . Potential genetic explanations include a liability threshold model requiring increased polymorphism risk in females , sex-limited X-chromosome contribution , gene-environment interaction driven by differences in hormonal milieu , risk influenced by genes sex-differentially expressed in early brain development , or contribution from general mechanisms of sexual dimorphism shared with secondary sex characteristics . Utilizing a large single nucleotide polymorphism ( SNP ) dataset , we identify distinct sex-specific genome-wide significant loci . We investigate genetic hypotheses and find no evidence for increased genetic risk load in females , but evidence for sex heterogeneity on the X chromosome , and contribution of sex-heterogeneous SNPs for anthropometric traits to ASD risk . Thus , our results support pleiotropy between secondary sex characteristic determination and ASDs , providing a biological basis for sex differences in ASDs and implicating non brain-limited mechanisms .
Autism spectrum disorders ( ASDs ) are characterized by deficits in use of language and social communication , sensory challenges , restricted interests , and repetitive behaviors that manifest in the first years of life . ASDs are estimated to occur in 1/42 boys and 1/189 girls , and are among the most heritable common disorders[1] . Estimates of heritability for idiopathic ASDs range between 38% and 90% , and autism-related traits in the general population are similarly heritable[2–9] . An emerging body of evidence has identified a wide array of potential non-genetic risk factors[10 , 11] . Nevertheless , the biological underpinnings and relevant environmental risk factors for ASDs are mostly unknown; thus , the nearly five-fold difference in prevalence between males and females may provide critical clues . Sexual dimorphism is extensive , begins early in development , and can be mediated primarily by hormonal or genetic ( 46 , XX vs . 46 , XY ) differences or by interaction between the two . In humans , hormonal and genetic factors are difficult to dissociate and often do not correspond to animal models . Despite much speculation , there is no definitive evidence regarding why males are more susceptible to ASDs[12] . Several testable genetic models could explain the reduced risk observed in females for idiopathic ASDs . 1 ) A multifactorial liability threshold model for genetic risk loci , whereby the same alleles affect males and females equally , but females have a higher threshold ( for biological or societal reasons ) requiring more polygenic load or stronger highly penetrant mutations to be affected or diagnosed due to the modifying effects of sex; 2 ) Specific susceptibility factors encoded on the X or Y chromosome that affect males , but not females , due to lack of Y or compensatory second copy of X; 3 ) Specific autosomal risk factors with different effects in males and females due to hormonally-mediated or otherwise-mediated sexual dimorphism , i . e . ‘autism’ is to some degree a different biological disorder in males and females due to gene-sex interaction; 4 ) A major influence of androgen levels[13] . If these effects are mediated via genes responsive to steroid hormones in their expression , we can hypothesize a role for steroid-responsive genes in genetic liability to ASDs; 5 ) Pleiotropy with general mechanisms of sexual dimorphism . Since variation in secondary sex characteristics ( i . e . height , weight , hip , and waist circumference ) is strongly heritable , this model would lead to the same genetic programs showing sex-heterogeneous signals for anthropometric traits exhibiting disproportionate contribution to ASD association . These distinct models are not mutually exclusive , thus in the present report we investigated evidence that would support each of them . A liability threshold model ( 1 ) would dictate that females with an ASD diagnosis would carry more genetic risk than affected males , on average . In support of a liability threshold model , previous studies show that females with ASDs are often more severely affected , with lower IQ and more frequent co-morbidities such as epilepsy[14–16] . Similarly , the difference in prevalence between males and females is lowest for the most severely affected individuals and highest for those who are highest-functioning on the spectrum[17] . Severe features could indicate a greater burden of modest inherited risk factors ( as tested in this study ) , more highly penetrant risk factors likely to be non-inherited , or both . ( Note that these predictions about risk factors are true regardless of whether affected females comprise a fair representation of ASD traits in the population or undergo diagnostic bias resulting in recognition of only the more severe cases; the liability threshold represents the empirical one for obtaining an ASD diagnosis . ) In support of some features of the liability threshold model , many highly penetrant ASD causes , such as de novo deletions , seem to have closer to equal sex ratios[18–20] . Recent exome sequencing studies have found that female cases have a greater proportion of de novo loss-of-function mutations and that single nucleotide variants ( SNVs ) identified in female cases exhibit an excess of deleterious predictions[21–24] . A model whereby females require higher inherited genetic loading to be affected than males would suggest that females should have an increased burden of family history . This model has some suggestive support in recent studies[25 , 26] , but lack of evidence in other studies[27] . Thus , it could be feasible that although individually strong risk factors like de novo mutations are enriched in females with ASDs , modest polygenic influence of common polymorphisms could contribute proportionately more or solely to male risk if they are generally insufficient to achieve the higher female threshold . Thus , there are two opposite but equally plausible models that can be simultaneously evaluated: 1 ) The majority of both male and female ASD is heritable ( not captured in de novo mutations ) ; strong sex bias is present in those likely to have risk from SNPs; therefore , the female liability threshold only considering SNPs may be increased compared to males ( it requires a higher SNP burden for a female to be affected ) . 2 ) For ASD overall , the observation of increased de novo mutation in females and more severe ID in females with ASD may imply that SNP risk is not sufficient for a female to be affected; therefore , considering only SNPs , female genetic burden may appear decreased compared with males ( based on SNPs , ASD would appear not to be heritable ) . Although rare genetic events causing ASDs have been identified on the X chromosome , such as mutations in the NLGN3 , NLGN4X , ARX , MECP2 , FMR1 genes , microdeletion , and aneuploidy[28] , there is little evidence that common risk factors of strong effect for ASDs lie on the X or Y chromosomes to support the sex chromosome risk model ( 2 ) [29–31] . A recent study of single nucleotide polymorphism ( SNP ) -based heritability estimated a disproportionately low contribution of the X chromosome to polygenic risk based on its length[32] . An exome sequencing study has estimated that 1 . 7% of male ASDs may be comprised of individuals with rare X-linked loss-of-function SNVs[33] . With respect to model 3 , male-specific autosomal linkage consistent with autosomal gene-sex interaction ( 3 ) has been identified , including a replicated region of chromosome 17[34–36] . In addition , autosomal dominant single-gene RASopathy syndromes have gene-sex interaction with NF1 showing male bias in ASD symptoms and Noonan syndrome showing a lack of sex bias[37] . However , gene-sex interaction has not been investigated in modern genome-wide association study ( GWAS ) datasets . Theories for excess male hormones characterizing ASDs ( 4 ) have led to investigation of testosterone levels in ASDs , with varied results[38–41] . A recent study found evidence for increased levels of steroid hormones in the amniotic fluid samples of subjects who went on to develop an ASD[42] . However , the androgen theory of ASDs has not yet been comprehensively investigated at the genetic level . To our knowledge , no one has studied the relationship of secondary sex characteristics and behavioral sexual dimorphism ( 5 ) . Here , we investigated five genetic models of sexual dimorphism in ASDs: 1 ) We examined evidence for a higher common polymorphism genetic load in the lower-prevalence sex . 2 ) We investigated sex-heterogeneity and association enrichment specific to the X chromosome . 3 ) We assessed the contribution of g x sex interaction across the autosomes . 4 ) We evaluated the role for genes whose expression is influenced by steroid hormones or sexually-dimorphic in the brain . 5 ) Finally , we estimated whether SNPs exhibiting sex-heterogeneous association with anthropometric traits contribute to ASD risk implicating pleiotropy with secondary sex characteristics .
In order to test the different hypotheses of sex-specific genetic architecture , we obtained the largest sex-specific datasets currently feasible . Recent analyses support the strategy of combining datasets with different ascertainment or diagnostic criteria to maximize power; increased sample size appears to have much greater impact than decreased homogeneity[43 , 44] . In order to achieve maximal sample size , we utilized previously published GWAS data [Autism Genetic Resource Exchange ( AGRE ) -Weiss[45] , AGRE-Wang[46] , Autism Genome Project ( AGP ) [47] , Early Markers for Autism ( EMA ) [48] , SSC[43]; N = 6 , 567 trios ( 16% female ) , N = 625 cases ( 21% female ) , and N = 377 controls ( 19% female ) ] . To these data , we added samples we genotyped at University of California San Francisco ( UCSF ) and/or via collaboration with a number of other consortia [UCSF , Childhood Autism Risks from Genetics and the Environment ( CHARGE ) [49] , Study to Explore Early Development ( SEED ) [50] , Autism Phenome Project ( APP ) , Tummy Troubles ( TT ) [51 , 52] , Interactive Autism Network ( IAN ) [53]; N = 195 trios ( 44% female ) , N = 1 , 259 cases ( 16% female ) , and N = 1 , 127 controls ( 37% female ) ] ( Table 1 , see URLs ) . Within each genotyping technical batch ( Table 2 ) , quality control was performed , and each dataset was imputed to the 1000G reference panel , with an additional round of quality control for imputed data ( Materials and Methods ) . All imputed datasets were then merged , and SNPs present in 90% of the total dataset were retained . From this mega-dataset , we extracted all complete trios ( N = 6 , 762 ) for transmission disequilibrium test ( TDT ) analysis and utilized the remaining data ( N = 3 , 388 ) in a case-control ( CC ) analysis . Finally , we performed a meta-analysis of the TDT and CC results for the complete combined-sex dataset ( N = 10 , 150 ) , as well as the male-specific ( trios with male probands and male cases vs . male controls , N = 8 , 207 ) and female-specific ( trios with female probands and female cases vs . female controls , N = 1 , 943 ) datasets . One SNP met genome-wide significance ( P = 5 x 10−8 ) in the combined-sex dataset ( rs7836146 near EXT1 ) ( Table 3 , S1 Fig ) . Two SNPs in one locus met genome-wide significance in the male-specific dataset ( rs7836146 and rs7835763 near EXT1 ) ( Table 3 , Fig 1A ) . Notably , patients with rare mutations in EXT1 have been previously described to have ASDs[54] . Three SNPs in one locus reached genome-wide significance in the female-specific dataset ( rs60443693 , rs12614637 , and rs140431641 in between CTNNA2 and SUCLG1 ) ( Table 3 , Fig 1B ) . Of the top association SNPs ( P < 10−6 ) , each of the independent loci in females show strong sex-heterogeneity ( Cochran’s Q , P < 10−3 ) and two of the five male independent loci ( both on the X chromosome ) show sex-heterogeneity ( P < 0 . 05 ) . In the combined-sex association results , one locus additionally shows sex-heterogeneity ( P < 0 . 05 ) ( S1 Table ) . None of the top sex-specific associations show within-sex differences comparing high vs . low IQ groups , suggesting the sex-specificity is not confounded by ASD severity differences ( S1 Table ) . Our first mechanistic hypothesis contributing to sex differences is increased genetic load in the lower-prevalence sex . Although it has been suggested in several studies that rare , highly penetrant risk variants are more strongly enriched in female probands compared to male probands , common polymorphism data has not been examined for sex differences in genetic load . We first assessed potential evidence for enrichment of genetic association signal in female cases ( compared to parental genotypes or female controls ) versus in male cases ( compared with parental genotypes or male controls ) . In order to adjust for the differential power of these datasets ( affected males N = 8 , 207; affected females N = 1 , 943 ) , we utilized sex permutations , whereby sex classifications were permuted within technical batch/dataset and study design ( trio or case-control ) to obtain permuted datasets of mixed sex but equal power to the true male and female datasets ( Materials and Methods ) . We assessed enrichment by setting a false discovery rate ( FDR ) threshold ( q = 0 . 8 ) and comparing the proportion of SNPs exceeding this threshold ( note that we use FDR only as a metric for comparison , not to assess significance ) . Male autosomal datasets did not show any enrichment compared to sex-permuted datasets ( Fig 1A ) . However , only 8% of sex-permuted female datasets exceeded the true female autosomal association enrichment ( P = 0 . 08 ) ( Fig 2A ) . This trend could occur due to heterogeneity ( e . g . some female-specific association loci not shared by males ) or due to increased genetic load in affected females . Next , we assessed whether SNP-based additive heritability ( h2g ) would support a liability-threshold model resulting in increased female genetic load . To do this we utilized our family-based dataset and compared proband genotypes to pseudo-controls ( non-transmitted parental alleles ) . Because the female-specific dataset is underpowered for heritability comparison ( Materials and Methods ) , we set aside the female-specific dataset and a matched male-specific dataset and utilized the remaining independent male-specific dataset for heritability estimation . We then added increasing numbers of female cases in a step-wise manner . For comparison , we did the same with the matched set-aside male-specific dataset . We observed no difference in correlation between the number of female versus male cases included and the observed-scale heritability estimates ( Spearman's rank correlation: female rho = 0 . 195 , P = 0 . 06; male rho = 0 . 195 , P = 0 . 06 ) . When using female and male pseudo-controls only , we see a negative correlation between number of male and female pseudo-controls added and observed-scale heritability estimates ( Spearman's rank correlation: female rho = -0 . 951 , P < 2 . 2 x 10−16; male rho = -0 . 863 P < 2 . 2 x 10−16 ) . These results suggest that on an individual basis females show equivalent SNP-based heritability to males ( Fig 2B ) . Finally , to assess individual-level risk burden distributions in males and females , we utilized the genome-wide genetic relationship matrix to predict the linear aggregate genetic risk for each individual . In order to carry out sex comparisons , we set aside our complete female datasets and a matched male subset . We used the remaining male data to generate best linear unbiased prediction ( BLUP ) solutions for each SNP , and applied these to our reserved independent male and female datasets ( Materials and Methods ) . We found that in both males and females , cases showed significantly higher mean SNP risk scores than controls ( Pmale = 1 . 82 x 10−5; Pfemale = 6 . 46 x 10−7 ) . However , neither male and female controls nor cases differed significantly from each other in a male-derived risk score ( Fig 2C ) . Similarly , within-sex high vs . low IQ ASD-affected groups did not differ by risk score ( S2 Fig ) . A second plausible genetic mechanism underlying sex differences could be genetic risk encoded on the X chromosome . We identified significant and suggestive association at several loci on the X chromosome , in or between the following genes: SPANXC , PRR32 / ACTRT1 , PRKX / NLGN4X , and MAGEC2 / SPANXN4 ( Table 3 ) . Thus , we wanted to test whether the X chromosome is overall enriched in association signal compared with similarly sized chromosomes and whether the X chromosome shows association that is specific to males with ASDs . First , we assessed association enrichment on the non pseudo-autosomal X chromosome utilizing a similar FDR-threshold strategy as above , in comparison with chromosome 7 ( similar physical size ) and chromosome 17 ( similar SNP representation ) . In females , chromosome X shows equivalent signal to the comparison chromosomes , and in the male-specific dataset , chromosome X shows slightly increased enrichment compared to chromosome 17 ( P = 0 . 09 ) ( S2 Table ) . Performing sex permutations , we observed enrichment in male-specific data compared with sex-permuted data ( P = 0 . 04 ) ( S2 Table ) . This could occur if X-linked loci have stronger effects in males compared to females or sex-limited effects that do not extend to females . We assessed association heterogeneity on the X chromosome via Cochran’s Q statistic . The vast majority of SNPs with heterogeneity P < 10−3 have larger absolute effect size in females; however , this is attributable to the smaller sample size , as our permuted datasets showed similar results ( Fig 3A ) . We examined 20 independent SNPs that were most significant in the male and female associations on the X-chromosome individually . The true distributions for heterogeneity among top SNPs were compared with sex-permutations as above , but adjusting the FDR level to account for SNPs ascertained for association ( FDRfemale = 0 . 01; FDRmale = 0 . 2 ) . The male top hits were significant for sex-heterogeneity ( P < 0 . 01 ) and the female top results did not show heterogeneity compared to the sex-permuted association results . As heterogeneity can be due to differing effect sizes in the same direction or no effect ( or opposite effect ) in one sample , we performed a binomial sign test . Female and male top 20 independent results on the X chromosome were suggestively or significantly depleted of same-direction effects compared with sex-permuted datasets ( Pfemale = 0 . 06 , Pmale < 0 . 01 ) ( Fig 3B ) . Third , we wanted to test a hypothesized mechanism of global autosomal gene-by-sex interaction . In contrast to the X chromosome , female and male autosomal top 100 independent SNPs did not show any difference from sex-permuted datasets for direction of effects ( Fig 3B ) . Together , the similar heritability and risk scores for male and female cases and lack of difference in the sign test suggest that much of the autosomal genetic signal is derived from common associations across males and females . In the extreme case of completely different genetic risk determinants , one would expect the heritability and association signal to decrease for both males and females in combined-sex datasets . To verify this prediction , we tested for heterogeneity among the 100 strongest independent autosomal association results for each sex by calculating Cochran’s Q statistic . We then set FDR thresholds adjusted for ascertaining via test statistics ( FDRfemale = 0 . 001; FDRmale = 0 . 2 ) and compared to our sex-permuted datasets , which have an even proportion of males and females per permuted set ( Materials and Methods ) . No significant evidence of heterogeneity was observed for top male or female SNP results . Our fourth and fifth hypotheses would result in enrichment of genetic association or sex-heterogeneity concentrated in specific limited sets of genetic variation , such as those likely to differ based on hormonal milieu or those shared with brain-specific or anthropometric secondary sex characteristics . To assess whether specific biological gene sets are likely to show sex differences , we obtained lists of genes with gene expression levels that are 1 ) androgen-responsive ( AR ) [55] , 2 ) estrogen-responsive ( ER ) [56] , 3 ) sexually-dimorphic in early brain development ( SD ) [57] , or 4 ) sex-correlated ( SC ) ( Materials and Methods ) . Significance of enrichment was determined by permutation with matched-length genes ( Materials and Methods ) and significance of sex differences was determined by sex permutation , as above . Lastly , we examined 5 ) SNPs showing sex heterogeneity in association to anthropometric traits ( AH; heterogeneity P < 10−3 for height , weight , BMI , hip , or waist measurements from the Genetic Investigation of ANthropometric Traits ( GIANT ) datasets[58] ) . Significance of enrichment for this set was determined by permutation matched by test statistic in the combined-sex GIANT dataset to control for SNP ascertainment via trait association , and empirical significance for sex differences was calculated by sex permutation ( Materials and Methods ) . SNPs contained within AR or ER genes experimentally determined to be hormone-responsive did not show enrichment for association signal in either sex or for heterogeneity via sex permutation . Nor did SD genes with sexually-dimorphic expression level or SC genes whose fetal brain expression is correlated or anti-correlated with ‘male-ness’ ( e . g . Y-encoded gene expression ) show excess association or sex differences . In order to test whether the same genetic influences resulting in sexually-dimorphic anthropometric traits are enriched in sexually-dimorphic behavioral disorders like ASDs , we tested whether SNPs showing sex heterogeneity of association signal to anthropometric traits ( AH ) were enriched for association signal or sex differences in ASDs . Indeed , in females with ASDs , AH SNPs showed enrichment for association signal compared to SNPs equally associated with anthropometric traits but not ascertained for sexual-dimorphism ( Pfemale < 0 . 01 ) ( Table 4 ) . To determine if this result is specific to females , we assessed enrichment of AH SNPs in the ASD combined-sex dataset and observed similar results ( Pall < 0 . 01 ) . We wanted to determine whether this observation might be exclusive to ASD , so we obtained summary statistics for two equivalently imputed datasets from the psychiatric genomics consortium ( PGC ) , schizophrenia ( SCZ ) [59] and bipolar disorder ( BIP ) [60] . Although neither disorder has overall sex prevalence differences , we observed similar AH SNP enrichment for BIP ( Pall < 0 . 01 ) , but not for the well-powered SCZ dataset . Previous work has revealed sexual dimorphism in onset , course , and co-morbidities of BIP[61] and in animal models , and suggested that it may be driven by endocrine systems[62]; thus the AH SNP enrichment in ASD is not unique but is also unlikely to be an artifact manifest in all GWAS data . In order to understand the functional characteristics of the AH SNPs , we tested for overlap with our gene sets , and found that they showed significant overlap with AR and ER datasets compared with permuted SNP lists ( P < 0 . 01 , each ) , although the amount of overlap was small . We also found suggestive overlap with predicted binding sites for hormone-responsive transcription factors AR , ESR1 and LEF1 ( P < 0 . 1 , each ) , but not for ESRRA , ESRRB , or NKX3-1[63] . We did not find disproportional overlap with SD or SC genes in the developing brain .
In this study , we set out to assess sex-specific mechanisms in common polymorphism association signal for ASDs . We gathered the largest feasible dataset to do so , however the strength of our conclusions are limited by the sample size we achieved and diverse study designs of component datasets , including mixed ancestry and ethnicity . We mitigated the impact of study differences on our sex-specific conclusions to the degree possible by permuting within technical batches/datasets and study designs to exclude some foreseeable sources of confounding , but our overall study may be reduced in power by the heterogeneity present . On the other hand , our use of diverse-ancestry datasets may render our results applicable to a broader group of populations . Our primarily family-based datasets may similarly contribute to relatively robust results due to perfect genetic matching between parents and offspring , but at the same time features such as assortative mating may reduce overall power compared with using population-based controls[32] . In addition , multiple analyses were performed in order to assess five potential hypotheses , calling for replication of each individual finding in the future when sufficient datasets become available . Despite these limitations , we describe results below providing evidence ( or absence thereof ) for genetic mechanisms of 1 ) increased genetic load in the lower-prevalence sex , 2 ) disproportionate ASD risk contained on the X chromosome , 3 ) global autosomal gene-by-sex interaction , 4 ) hormone-driven genetic sex differences , and 5 ) general pleiotropy resulting in shared mechanisms between ASD risk and secondary sex characteristics . Our first hypothesis was a major difference in genetic load between males and females . We showed similar heritability estimates on the observed-scale when females were included , and similar male-derived risk prediction scores in female and male cases . However , we observed a trend towards enrichment of association signal in female-only analysis compared with permuted-sex analyses . Thus , we do not see a decrease in common polygenic load in affected females , which might have been consistent with a disproportionate or nearly exclusive role for rare or de novo genetic variants contributing to female ASDs and associated with the more severe phenotypic manifestations . On the other hand , we do not observe a striking excess of genetic load or enrichment in females as has been demonstrated for de novo loss of function variants in overlapping datasets[18–24] . We do not see differences in genetic load comparing within-sex low vs . high IQ groups . Despite our limited power to detect modest relative differences , we do observe an extremely clear case-pseudocontrol difference , demonstrating our power to detect large differences even in a mixed-ancestry dataset . Together these results suggest that each component of genetic architecture ( rare variants , common polymorphisms , inherited , de novo , etc . ) should be considered separately for sex differences in ASDs , and potentially in other disorders . Our second hypothesis was a major role for X chromosome polymorphisms . Male-specific analyses revealed sex-heterogeneity specific to the X chromosome and several X-linked loci associated at genome-wide significant levels . In addition , we find association in the combined-sex dataset with polymorphism in NLGN4X , previously reported as having rare inherited variants associated with ASDs in males[64 , 65] . Female X chromosome associations also show suggestive results in the sign test , indicating that there may be female-acting and female-specific risk loci on the X chromosome . These results , taken together , suggest a role for common polymorphisms on the X chromosome in addition to the more well-described role for rare X-linked loci or Mendelian diseases in ASD[28] ( 33 ) . Although many complex trait studies exclude the X chromosome when examining genome-wide autosomal SNPs , our results indicate additional analysis of SNPs on the X chromosome in sex-specific datasets may be worthwhile . Third , we proposed global autosomal sex-heterogeneity . Despite the relatively smaller sample size , we identify a genome-wide significant association signal in females when analyzed alone , which shows strong sex-heterogeneity ( P < 10−8 ) , but no difference in low vs . high IQ groups ( S1 Table ) . This locus is near the CTNNA2 locus encoding alpha-2 catenin , a key neurodevelopmental gene . In addition , one of the loci identified in the combined-sex association results at P < 10−6 , near EXOC4 , showed significantly stronger effect in female cases . This locus , as well as male-identified EXT1 and NRSN1 loci in our ASD data , have recently been implicated in population-based learning and memory GWAS[66 , 67] . Our results thus suggest that potential sex differences should be investigated in these cognitive phenotypes . We did not find evidence for global sex-heterogeneity in association or heritability for the autosomal genome; nor did we identify a locus on chromosome 17 that might explain previous sex-specific linkage findings [34–36] . However , our datasets are limited in power to detect subtle effects that might become evident with increased sample sizes and our study design is complicated by combining datasets with different ascertainment biases . Fourth , we assessed whether a hormone-driven mechanism might be evident in genetic set enrichment . We were unable to identify genetic support for an androgen-driven mechanism for ASD risk loci , represented by genes with expression levels influenced by androgens . Nor did we find evidence for markedly increased influence of genes with sexually-dimorphic brain expression in early development . As these represent small and imperfectly-selected sets of SNPs to represent functional categories , our power may be limited to conclude a lack of effect from these mechanisms . In addition , it is possible that dimorphic gene expression may arise through sex-related differential environmental effects but not show association with genetic variation in these genes . Our final hypothesis was substantial pleiotropy between anthropometric and complex disease sex differences . We found strong evidence that variants with sexually-dimorphic effects on anthropometric traits contribute disproportionately to ASD association . Our interpretation of this result is that the same mechanisms acting on secondary sex characteristic differences later in life may influence ASD risk in early development . As these loci were identified via anthropometric traits such as height , weight , and waist/hip measurements , our finding suggests general pleiotropy rather than brain-limited or behavioral-specific influences on sex-specific ASD risk . Although we obtained similar results in bipolar disorder , independent replication of these results in additional ASD datasets would be ideal . There has been much discussion of potential diagnostic bias towards males influencing the observed prevalence differences by sex[68] . However , the disproportionate enrichment ( particularly in affected females ) of anthropomorphic-heterogeneous SNPs indicates that general biological mechanisms related to sexual dimorphism contribute prominently to ASD risk and could be investigated in other sex-biased behavioral and developmental disorders . Further work may clarify the functions of this set of SNPs and the means by which they act on ASD risk , and could help to quantify limitations on the effects of diagnostic bias in observed prevalence differences . Overall , our study complements recent identification of rare variants in ASD-affected females by assessing polygenic common SNP association in a sex-specific framework[24] . We report comprehensive evidence of common polymorphic X-linked loci contributing to ASD risk and sex-heterogeneity specific to X-linked loci . Notably , our data highlight the importance of general mechanisms of sexual dimorphism in the etiology of ASDs , and future research may be able to clarify specific biological mechanisms involved and to what degree our findings here may apply to other sex-biased disorders .
Information about diagnosis and inclusion/exclusion criteria for each dataset is described in Supplemental Note 1 . Genotype data for each dataset are summarized in Table 1 . Previously published GWAS data included Autism Genetic Resource Exchange ( AGRE ) -Weiss[45] , AGRE- Wang[46] , Simons Simplex Collection ( SSC ) [43] , Autism Genome Project ( AGP ) [47] , and Early Markers for Autism ( EMA ) [48] . We obtained data by application to AGRE , SSC , dbGAP ( AGP ) , or as study investigators ( EMA ) ( see URLs ) . Genotype data and phenotype data were utilized as provided , with additional quality control steps described below . Normalized intelligence quotient ( IQ ) or developmental quotient ( DQ ) data indicating low ( <70 ) or high ( >80 ) functioning categories were available for 3 , 571 affected males ( 2 , 017 low , 1 , 554 high ) and 619 affected females ( 405 low , 214 high ) and used for secondary analyses ( see S1 Note ) . Genotyping was performed at the University of California San Francisco ( UCSF ) genomics core facility for unpublished trios and case-control samples from UCSF/Weiss , UCSF/Hendren[69 , 70] , Tummy Troubles ( TT ) [51 , 52] , Interactive Autism Network ( IAN ) [53] , Childhood Autism Risks from Genetics and the Environment ( CHARGE ) [49] , Autism Phenome Project ( APP ) , and for a portion of the multisite Study to Explore Early Development ( SEED ) [50] study ( see URLs ) . Affymetrix Axiom EUR arrays were used , according to manufacturer protocols[71] . Additional unpublished data from the SEED cohort were genotyped on the Illumina Omni1M Quad BeadChip at the Johns Hopkins SNP Center , according to manufacturer protocols . For this dataset , quality control measures were applied within technical batches ( Table 2 ) , stratified by ancestry . These measures included removal of samples with a call rate less than 98% , a sex discrepancy , relatedness ( PI-HAT > 0 . 2 ) , or excess hetero- or homozygosity . [Note that previous studies have shown inflated PI-HAT estimates in multi-ethnic datasets , thus our relatively high PI-HAT threshold is appropriate for this study design[72] . Additionally , SNPs with a missing call rate greater than 1% , monomorphic , with minor allele frequency ( MAF ) less than 1% , or which deviated significantly ( P < 1 . 0x10-10 ) from Hardy Weinberg Equilibrium ( HWE ) were removed . All datasets were anonymized and patient identifiers , except for affection status and sex , were removed in the genotyping datasets used by the investigators . Saliva and blood samples collected for patients recruited specifically for this study for the dataset UCSF/Weiss were approved for research use by UCSF Committee on Human Research ( IRB #: 10–02794 ) . We obtained informed consent and HIPAA authorization for all participants . We have made these data available on The National Database for Autism Research ( NDAR ) ( see Accessions , see URLs ) . According to the following criteria set by the UCSF Committee on Human Research ( 1 ) coded private information or specimens not collected specifically for the current research project , and for which ( 2 ) by agreement or by IRB-approved written policies the key to coded human subjects data will not be released to investigators analyzing the data , the other datasets utilized for this study were considered as non-human subject data by the UCSF Committee on Human Research . Marker quality was assessed within technical batches; exact thresholds for HWE , call rate , MAF , and Mendel errors for marker exclusion in the different datasets are noted in Table 2 . Technical batches were merged within sub-studies to assess individual identity or to check for known and unknown relationships . Relationships indicating confounding family structure were corrected or individuals contributing to confounding relationships were removed . Remaining individuals were assessed for individual call rate , heterozygosity and sex; those who had unresolvable sex ( F-het > 0 . 3–0 . 35 ) , increased heterozygosity , or genotyping rate < 0 . 95% were removed . All individual and marker quality control was carried out using PLINK ( see URLs ) [73 , 74] . Genotype datasets mapped to hg18 positions were updated to hg19 using the LiftOver tool available from University of California Santa Cruz ( UCSC ) Genome Browser ( see URLs ) . Post-quality control datasets , separated by genotyping platform , were checked against 1000G phase1v3 reference data using SHAPEIT’s—check function[75 , 76] ( see URLs ) . Markers that received an error warning had alleles flipped using PLINK’s—flip option; flipped data was rechecked against the reference panel , and finally any markers still receiving an error warning using SHAPEIT’s—check were then excluded from consideration . Refined datasets were then phased utilizing SHAPEIT and 1000G phase1v3 reference data , specifying—duohmm -W 5 to take advantage of pedigree information when available . Phased genotyping datasets were imputed with IMPUTE2 specifying HapMapb37 as the recombination map , 1000G phase1v3 as the reference panel , and an effective population size of 20 , 000 using the–Ne flag[77] ( see URLs ) . Chromosomes were processed separately in consecutive chunks of 5MB per chunk for imputation . Chunks were concatenated across entire chromosomes and converted back to PLINK binary file format from Oxford gen/sample format for each chromosome separately , keeping only calls with a imputation quality score of >90% . All marker calls were then matched to the reference panel’s marker ID and position to ensure only properly imputed markers remain; any marker presenting an ID and position that were not exact matches to the reference panel were excluded from further consideration . Separated chromosomes were then merged for each dataset . Quality control filters were applied separately for each dataset , eliminating markers with HWE P< 1x10-10 , call rate of 0 . 95 and greater than 10 Mendel errors where applicable ( Table 2 ) . Additionally , SNPs with large differences in MAF between datasets or indication of being flipped between datasets were removed . All datasets were then merged , applying an additional call rate filter of 0 . 9 and MAF of 0 . 01 to include only common variants genotyped for the majority of individuals for analysis . Association was assessed in trio-family ( unaffected mother and father with ASD affected child ) designed studies using the transmission disequilibrium test ( TDT ) for 6 , 762 affected probands ( 1 , 113 females and 5 , 649 males ) . Association was tested in case-control ( CC , ASD affected probands and unrelated unaffected controls ) datasets using logistic regression considering ten principal components as covariates in order to control for population stratification ( S3 Fig ) . The ten principal components were calculated using PLINK—mds-plot 10—cluster options . No other covariates were used for the analyses . 1 , 884 cases and 1 , 504 controls were used for the logistic regression analysis ( 338 female cases and 492 female controls; 1 , 546 male cases and 1 , 012 male controls ) . Primary association analyses were carried out using PLINK v1 . 90[73] . The TDT and logistic regression summary statistics were then used as input into METASOFT[78] ( see URLs ) for a fixed-effects meta-analysis to find combined-sex association results , male-specific association results ( trios with male probands and male CC ) , and female-specific association results ( trios with female probands and female CC ) . The standard GWAS significance threshold of P ≤ 5 . 0x10-8 was used to identify genome-wide significant SNPs accounting for approximate independent common variants[79 , 80] . In order to test sex-specificity for each analysis relevant to a potential mechanism of sexual dimorphism , sex permutations were performed by randomly permuting sex classifications ( i . e . male or female ) for each individual . Individuals were permuted within their respective genotype technical batch/dataset and study design ( trio or CC ) ( Table 2 ) to account for batch effects . The total number of individuals included in each permuted-sex set was matched to the actual number of male or female probands in the batch to account for the difference in power between the sexes ( R script available in GitHub ) . Then , the TDT and logistic regression association tests were performed on the male-permuted and female-permuted ( sex-permuted ) datasets , and meta-analysis of the TDT and logistic regression summary statistics was implemented . Association signal was calculated as the percent of SNPs that surpassed a given FDR q-value of 0 . 8 . The FDR threshold was determined by finding the common threshold for all datasets that had a reasonable number of SNPs to utilize for empirical comparison ( S3 Table ) . Note that this FDR is not used to assess significance , only as a metric for comparison . The observed sex-specific association signal was compared to 100 sex-permuted results . The empirical P-value for sex specific association was calculated as the proportion of permuted datasets more extreme than the observed data . First , pseudo-controls were created based on our trio dataset using PLINK ( —tucc ) software[74] . A single proband from each family was used , and individuals showing any relatedness ( PI_HAT > 0 . 1 ) were removed . The final multi-ethnic dataset consisted of 5 , 311 trio probands and 5 , 311 pseudo-controls to be used for heritability and risk prediction analysis . Unrelated case-control datasets were excluded from these analyses , as they would be challenging to match precisely by genetic ancestry . Using Genome-wide Complex Trait Analysis ( GCTA ) , we created the genetic relationship matrix ( GRM ) between all pairs of individuals based on all autosomal SNPs ( see URLs ) [81] . We calculated the heritability based on the GRM and ten principal component analysis ( PCA ) eigenvectors as quantitative covariates to account for population stratification[81–83] . Heritability on the observed scale , defined as the genotypic variance divided by the phenotypic variance , was estimated using GCTA program’s unconstrained restricted maximum likelihood ( REML ) analysis . To assess the effect of female cases on the heritability estimate , we performed the REML analysis in GCTA for differing proportions of added female cases . Starting with a base set of 6 , 810 male trio probands and pseudo-controls , we added female probands and their matched pseudo-controls in a step-wise manner from 0 to 1 , 906 , the maximum number of females . A total of 97 sets were created , where each additional set contained all the individuals from the previous set plus up to ten pairs . For comparison , we performed a similar step-wise analysis , adding an equal number of male proband and matched pseudo-controls to the base male dataset ( R script available in GitHub ) . As a negative control and to account for sample size , we performed the step-wise heritability analysis with pseudo-controls only . We did this first with female pseudo-controls designated as “cases” and male pseudo-controls designated as “controls” , and then switched female and male pseudo-controls . To avoid technical batch effects , males and females that were added to the base effects were ascertained from the same genotyping technical batch ( Table 2 ) . The observed scale heritability estimate was calculated for every set with male or female cases added . Spearman's rank correlation test was conducted to assess significance . To determine the genetic risk score for individuals , first , we divided the male-specific dataset into a discovery set and a test set . To avoid technical batch effects , we matched the male test set to the number and technical batches of the female set , as done for the heritability analysis . The discovery set contained 6 , 810 males ( 3 , 405 probands and 3 , 405 pseudo-controls ) . We predicted the total genetic effect of all SNPs in the male discovery set using best linear unbiased prediction ( BLUP ) method in GCTA ( —reml-pred-rand ) , and then transformed the solutions for individual autosomal SNPs ( —blup-snp ) [81–83] . Finally , we predicted the risk score utilizing these SNP-solutions using PLINK ( —score ) for an independent test male-specific dataset and female-specific dataset ( 953 probands and 953 pseudo-controls each ) . To determine the significance of difference in mean predicted risk score between cases and controls and between males and females , we conducted an independent two sample t-test in R ( see URLs ) . Similarly , we assessed mean differences in low IQ ( <70 ) and high IQ ( >80 ) groups within-sex by t-test for individuals with IQ data available overlapping with the independent male and female test datasets ( S2 Fig ) . We also verified that strongly associated SNPs are not the main contributing factor to the difference in the distribution of risk scores between cases and pseudo-controls . This was determined by performing the analysis excluding SNPs with combined-sex ASD association P < 1 . 0x10-6 . Sex- specific association signal enrichment was tested for autosomes , chromosome 7 , chromosome 17 , and the non-pseudoautosomal X chromosome . For the case-control association component , the X chromosome was coded in standard PLINK format where male genotypes are A = 0 and B = 1 , and female genotypes are AA = 0 , AB = 1 , and BB = 2 ( see URLs ) . For mixed-sex analyses ( e . g . combined-sex and permuted datasets ) sex is also included as a covariate . No changes are required to the TDT for the X chromosome . Male-specific and female-specific association results per chromosome were assessed for enrichment of genetic signal compared to sex-permuted datasets , derived as above ( see Assessment of Sex Specificity ) . In the same manner , association signal for each chromosome was calculated as the percent of SNPs that surpassed the FDR q-value of 0 . 8 . To assess heterogeneity between males and females on the X chromosome , we calculated the Cochran's Q statistic and P-value using METASOFT[78] ( see URLs ) . The Cochran’s Q statistic[84] for each SNP is the weighted sum of squared differences between the effect estimates in the sex-specific analyses and the combined sex meta-analysis . Cochran’s Q follows a chi-square distribution with 1 DF . A significant P-value indicates there is a difference in the SNP effect estimates between the male and female specific datasets . We looked at these heterogeneity results in three ways . First , for chromosome X SNPs with a suggestive Cochran’s Q result ( P < 1 . 0x10-3 ) , we calculated the proportion with a greater absolute effect size , as indicated by the beta from the fixed-effects meta-analysis , in females versus males . We performed the same analysis with the sex-permuted association results to account for power differences in the male and female datasets . Second , we examined the 20 most significant linkage disequilibrium ( LD ) independent X chromosome results separately in males and females . We used PLINK–clump option to LD prune the SNPs based on the sex-specific association P-values . Separately for males and females , we found the corresponding Cochran’s Q P-value for the top 20 SNPs , and calculated the percent of SNPs that surpassed a given FDR q-value of 0 . 2 in males and 0 . 01 in females . We determined the FDR q-value based on the value that produced a reasonable percent ( between 50–80% ) for comparison ( S4 Table ) . We performed the same analysis in the 100 sets of permuted-sex datasets , and compared the observed male-specific and female-specific results to derive an empirical P-value . We also verified that individual associated SNPs are not the main contributing factor to heterogeneity by performing the analysis excluding SNPs with male-specific and female-specific ASD association P < 1 . 0x10-6 . Lastly , we examined the direction of effect , as indicated by the beta from the meta-analysis , of the LD independent top 20 SNPs for each sex . We conducted a binomial sign test , and compared the results to the 100 permuted-sex results to assess significance ( R script available in GitHub ) . In order to assess gene-by-sex interaction across the autosomes , we performed genome-wide heterogeneity analysis via Cochran’s Q test , in the same manner as above for the X chromosome . We examined the most significant 100 independent autosomal results for males and females , which were filtered using PLINK—clump . In the same method as described above , for these top 100 SNPs in the sex-specific association results , we calculated the proportion that had a Cochran’s Q result above or equal to an FDR q-value of 0 . 2 in males and q-value of 0 . 001 in female autosomes . The FDR q-value threshold was chosen separately in males and females to avoid saturation of results and allow for reasonable comparison ( S4 Table ) . Using the same method , we calculated heterogeneity levels of the sex-permuted association results to derive an empirical distribution for comparison . We compared our true sex-specific heterogeneity enrichment values to the empirical distribution to calculate an empirical P-value . Next , for the same set of 100 SNPs , we compared between the sexes the direction of effect by implementing a binomial sign test . We compared the proportion of SNPs associated in the same direction in the true results to the sex-permuted datasets to calculate an empirical P-value . We defined several autosomal gene sets of interest , including a 5kb flanking region when defining each gene . For all five gene sets , we removed genes based on the following criteria ( 1 ) duplicated gene name listed , ( 2 ) no corresponding SNPs in the genotype dataset , ( 3 ) gene length greater than 92Mb for appropriate length-matching , and/or ( 4 ) on the X or Y chromosome . Androgen-responsive ( AR ) gene list was gathered from Androgen Responsive Gene Database ( ARGDB , see URLs ) for a total of 2 , 613 genes[55] . Of these 2 , 613 genes , 2 , 070 genes met our criteria . Estrogen-responsive ( ER ) gene list was gathered from Estrogen Responsive Gene Database ( ERGDB ) , with a total of 1 , 384 genes[56] , of which 1 , 092 genes met our criteria . Sexually-dimorphic ( SD ) genes were defined as those previously shown to have sex-biased expression patterns in the fetal brain , for a total of 285 genes , of which 227 genes met our criteria[57] . Sex-correlated ( SC ) genes were defined based on a number of fetal brain gene expression datasets: 1 ) ABI . RNAseq . 21 . to . 26: RNAseq data from a variety of cortical areas and individuals aged 21 to 26 post-conception weeks ( PCW ) ; 2 ) Sestan . STHB . 19 . to . 37: Affy Exon data from a variety of cortical areas and individuals aged 19 to 37 PCW[57]; 3 ) ABI . 4CTX . Cingulate: Agilent arrays analyzing laser micro-dissected samples spanning the entire developing wall of cingulate cortex from four individuals ( 15–22 PCW ) ; 4 ) STHB . STR . 8 . to . 22: Affy Exon data from ventral telencephalon , individuals aged 8 to 22 PCW[57]; 5 ) AFFYEXON . 4to6 . mo: Affy Exon data from various brain regions in 4 to 6 month individuals[57] . [Note that these datasets may not be entirely independent of each other . ] For each dataset , we found the coexpression module most significantly enriched with a set of genes previously found to be differentially expressed between males and females in human cerebral cortex[85] . These modules were summarized by their first principal component and all genes ( or probes ) in each dataset were correlated to PC1 . These correlations were Fisher-transformed and averaged across datasets with weights corresponding to sample sizes . These values were then converted into 'average' correlation coefficients ( r ) using the reverse Fisher transformation and ranked genome-wide . Y-chromosome genes dominate the signature , thus genes were considered male-correlated with a Pearson correlation coefficient r > 0 . 3 and male anti-correlated with r < -0 . 3 to include moderate and strong association . Based on our criteria , we found a total of 826 autosomal male correlated genes and 58 autosomal male anti-correlated genes with corresponding SNPs in the imputed genotype dataset . For each gene set ( AR , ER , SD , and SC ) , SNPs falling within +/- 5 kb of each gene were extracted for analysis and filtered to contain no duplicate SNPs . Anthropometric-heterogeneous ( AH ) SNPs were defined in the GIANT datasets [body mass index ( BMI ) , hip circumference ( HIP ) , HIP adjusted for BMI ( HIPadjBMI ) , waist circumference ( WC ) , WC adjusted for BMI ( WCadjBMI ) , waist-to-hip-ratio ( WHR ) , WHR adjusted for BMI ( WHRadjBMI ) , height , and weight][58] as showing sex difference in association with any anthropometric trait ( P < 10−3 ) . The SNP list was filtered to contain no duplicates and was LD-pruned for a final total of 8 , 140 SNPs , of which 3 , 238 overlap with the imputed ASD genotype dataset . In order to assess significance of enrichment in the AR , ER , SD , and SC gene sets of interest , 100 permuted gene sets with individually length-matched genes were chosen to match the true gene sets . Gene and size information were downloaded from RefSeq database—UCSC genome browser ( see URLs ) . For each gene in the set , a gene was randomly selected from the 100 genes most similar in length to the gene of interest . For permuted gene sets , SNPs falling within +/- 5 kb of each length-matched gene were extracted for analysis . CCSER , CNTNAP2 , CSMD3 , CTNNA2 , DPP6 , GRID2 , LRP1B , and MACROD2 genes were too large to be matched for permutation and therefore were excluded from all gene set investigation . In order to assess significance of enrichment for the AH SNPs , 100 permuted lists of SNPs equally associated with the anthropometric traits for which the AH SNPs show sexual dimorphism were generated . For each AH SNP of interest , a SNP was randomly selected for the permuted list from 100 SNPs with the most similar trait association P-value . Association signal in the true biological sets of interest were compared to the permuted lists to derive an empirical P-value . We used the consistent FDR q-value = 0 . 8 threshold to determine association enrichment ( see above ) . In addition , we tested for binding site enrichment of AH SNPs compared to permuted lists of SNPs equally associated with anthropometric traits but not ascertained for sexual-dimorphism . The hormone-responsive transcription factors ( TF ) we tested included: estrogen receptor 1 ( ESR1 ) , estrogen receptor 2 ( ESR2 ) , estrogen-related receptor alpha ( ESRRA ) , estrogen-related receptor beta ( ESRRB ) , NK3 homeobox 1 ( NKX3-1 ) , lymphoid enhancer-binding factor 1 ( LEF1 ) , and androgen receptor ( AR ) . UCSC Hg19 Table Browser[86] was used to get the 50bp upstream and downstream DNA sequence surrounding each SNP . These sequences for the AH SNPs and for the permuted SNPs lists were used as input into Deepbind[63] which used deep learning techniques to predict the binding of the hormone-responsive TF to the specified sequences . For each TF , we compared the number of sequences in permuted lists with binding scores above a threshold corresponding to the top hundredth sequence in the true AH sequences to reach an empirical P-value for the TF binding site enrichment . Power analysis conducted prior to analysis suggested that for our study goal of 2 , 000 affected individuals in the female-only ( smallest ) dataset for a trio design , we would have approximately 80% power at P-value 5 x 10−8 to detect a genotype relative risk of at least 1 . 35 for common alleles ( MAF 30% ) . This effect size was in the range of reported effects for other GWAS studies at the time , particularly considering that our hypothesis was that the lower-prevalence sex might contain stronger risk alleles . This analysis is simplistic , considering that our study design of meta-analyzing a small case-control cohort with the larger trio dataset is not accounted for [per affected individual , power is increased for our case-control subset– 0 . 78X cases are required for equal power] . Further , the power calculation was performed in order to assess the adequacy of our sample size and thus considers only a single genome-wide association analysis and none of the other kinds of analyses we performed and tested empirically . We determined that power was insufficient for direct comparison of heritability between male and female ASD-affected probands , since for a 10% difference in heritability , we would have only approximately 30% power[87] . As risk scores can be analyzed much like any quantitative trait , our power for a t-test was adequate to detect large case vs . control differences ( 80% power for 0 . 13 SD ) . However , power was limited to detect more subtle potential male vs . female differences ( magnitude of mean difference would need to be 67% of that observed for male case vs . control mean difference to achieve 80% power; empirical sex difference in means was 16% of the case-control difference ) . Although each individual analysis is adequately corrected for multiple testing either by significance threshold or permutation , we have not accounted for the three datasets utilized ( male , female , all ) , the five major hypotheses we are testing , nor the multiple approaches used to assess evidence for each hypothesis . Therefore , our results should be interpreted in light of the limitations of our multi-faceted study design . The accession number for the UCSF ASD genotype data reported in this paper is The National Database for Autism Research ( NDAR ) ID 1883 . Sex permutation datasets . https://github . com/michelaTra/ASD_SS_Mitra_I_2016/blob/master/sex_permutation_CC_trios_creator . R Spiked datasets . https://github . com/michelaTra/ASD_SS_Mitra_I_2016/blob/master/risk_score_spike_set_creator . R Sign test . https://github . com/michelaTra/ASD_SS_Mitra_I_2016/blob/master/sign_test . R Additional R functions and utilities . https://github . com/michelaTra/ASD_SS_Mitra_I_2016/blob/master/pipeline_function . R https://github . com/michelaTra/ASD_SS_Mitra_I_2016/blob/master/utils . R https://github . com/michelaTra/ASD_SS_Mitra_I_2016/blob/master/pulling_variant_windows_function . R 1000G phase1v3 reference data: https://mathgen . stats . ox . ac . uk/impute/data_download_1000G_phase1_integrated . html Androgen Responsive Gene Database ( ARGDB ) : http://argdb . fudan . edu . cn/ Autism Genetic Resource Exchange ( AGRE ) : http://agre . autismspeaks . org/site/c . lwLZKnN1LtH/b . 5332889/k . B473/AGRE . htm Autism Genome Project ( AGP ) : http://www . autismspeaks . org/science/initiatives/autism-genome-project Autism Phenome Project ( APP ) : http://nationalautismnetwork . com/research/research-initiatives/autism-genome-project . html Childhood Autism Risks from Genetics and the Environment ( CHARGE ) : http://beincharge . ucdavis . edu/ DeepBind Predictive Models: http://tools . genes . toronto . edu/deepbind/ Genome-wide Complex Trait Analysis ( GCTA ) : http://cnsgenomics . com/software/gcta/ HapMap b37: http://www . shapeit . fr/files/genetic_map_b37 . tar . gz IMPUTE2: https://mathgen . stats . ox . ac . uk/impute/impute_v2 . html Interactive Autism Network ( IAN ) : http://iancommunity . org/cs/ian_research/ian_genetics LiftOver—University of California Santa Cruz ( UCSC ) Genome Browser: https://genome . ucsc . edu/cgi-bin/hgLiftOver LocusZoom: http://locuszoom . sph . umich . edu/locuszoom/ METASOFT: http://genetics . cs . ucla . edu/meta National Database for Autism Research ( NDAR ) : https://ndar . nih . gov/ PLINK: http://pngu . mgh . harvard . edu/~purcell/plink/index . shtml R—A language and environment for statistical computing: http://www . R-project . org/ RefSeq Genes Database–UCSC: http://hgdownload . cse . ucsc . edu/goldenPath/hg19/database/knownToRefSeq . txt . gz Simons Simplex Collection ( SSC ) : http://sfari . org/resources/autism-cohorts/simons-simplex-collection Study to Explore Early Development ( SEED ) : http://www . cdc . gov/ncbddd/autism/seed . html UCSC Table Browser: http://genome . ucsc . edu/cgi-bin/hgText
|
Autism Spectrum Disorders ( ASDs ) make up a debilitating neurodevelopmental disorder class . It has been known for a long time that more males than females are affected , but despite much speculation there is no clear etiological reason for this sex bias . As ASDs are highly heritable , we examined evidence in single nucleotide polymorphism ( SNP ) data for five plausible genetic models that could generate sex bias . We identified distinct genome-wide significant loci in each sex-specific dataset , and evaluated support in five analyses: 1 ) In contrast to rare variant contribution , we find no evidence for increased SNP genetic load in females . 2 ) Sex-heterogeneity is demonstrated on the X-chromosome . 3 ) We uncover no evidence for hormone-responsive genes being overrepresented in association signals . 4 ) We identify no signature for genes differentially brain-expressed between males and females contributing to ASDs . 5 ) We observe a strong signal of excess association in the same regions of the genome showing sex-heterogeneity in anthropometric traits . This latter finding is striking , implicating general sexual dimorphism as opposed to brain- or behavior-specific origins for sex differences contributing to ASDs .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] |
[
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"social",
"sciences",
"autism",
"anthropology",
"developmental",
"psychology",
"neuroscience",
"developmental",
"biology",
"mathematics",
"genome",
"analysis",
"autism",
"spectrum",
"disorder",
"sexual",
"dimorphism",
"morphogenesis",
"discrete",
"mathematics",
"combinatorics",
"developmental",
"neuroscience",
"sex",
"chromosomes",
"genomics",
"neurodevelopmental",
"disorders",
"anthropometry",
"chromosome",
"biology",
"x",
"chromosomes",
"genetic",
"loci",
"psychology",
"permutation",
"cell",
"biology",
"physical",
"anthropology",
"sexual",
"differentiation",
"heredity",
"neurology",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"genetic",
"load",
"chromosomes",
"human",
"genetics"
] |
2016
|
Pleiotropic Mechanisms Indicated for Sex Differences in Autism
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.